Works in Progress Podcast

0:00:01.200,0:00:09.040
Place yourself in the 1980s. This is 
kind of hard because, as a scientist,

0:00:09.040,0:00:16.480
you had just found out that retroviruses 
could infect humans. Now you find out

0:00:16.480,0:00:22.960
that HIV is a retrovirus. You're like, 
wait, I didn't even know that RNA could

0:00:22.960,0:00:29.280
be turned into DNA until 10 years ago. So 
that's quite a tricky position to be in.

0:00:33.920,0:00:39.760
Making effective new medicines isn't easy. 
Welcome to Hard Drugs. I'm Saloni Dattani,

0:00:39.760,0:00:43.920
a researcher on global health at Our World 
in Data, and one of the founders of Works

0:00:43.920,0:00:49.600
in Progress magazine. And I'm hosting this podcast 
with Jacob Trefethen, who leads science and global

0:00:49.600,0:00:55.360
health R&D funding at Open Philanthropy, and is 
one of the most fun, interesting people I know.

0:00:55.360,0:00:59.360
This show is about medical innovation: 
how to speed it up, how to scale it up,

0:00:59.360,0:01:04.240
and how to make sure lifesaving tools reach the 
people who need them the most. It all started with

0:01:04.240,0:01:09.440
a conversation, a shared instinct that this was 
the right time to start a podcast, to dive deep

0:01:09.440,0:01:14.960
into how to technologies for malaria, cancer, 
AIDS, and other diseases, actually came to be.

0:01:14.960,0:01:22.400
Today we're going to talk about HIV. Making an HIV 
vaccine has been the holy grail for many of the

0:01:22.400,0:01:27.360
world's top scientists over the last generation. 
It has proven one of the most challenging

0:01:27.360,0:01:33.200
scientific problems too, and we don't yet have a 
vaccine. But last year, one drug company announced

0:01:33.200,0:01:39.200
they'd gone a completely different route. They 
made a drug you get injected with once every

0:01:39.200,0:01:45.200
six months, or maybe only once a year, like a 
flu shot, giving you almost perfect protection

0:01:45.200,0:01:57.840
against HIV. So how did we get here and what does 
it mean for one of the world's deadliest diseases?

0:01:57.840,0:02:07.040
I'm super excited to talk about HIV, lenacapavir 
and other HIV drugs today with Jacob. Hello.

0:02:07.040,0:02:09.200
Hey, how are you doing?

0:02:09.200,0:02:15.680
Great. Yeah, so I'm super excited about this. I 
think we have a bunch of things that we're going

0:02:15.680,0:02:22.080
to talk about in the episode, maybe starting with 
just what HIV is, how it infects people, and then

0:02:22.080,0:02:28.800
moving on to the history of drug development 
in the field, how lenacapavir was develops,

0:02:28.800,0:02:36.240
what lenacapavir actually does, and then where we 
are now, how to scale it up to people who need it.

0:02:36.240,0:02:41.600
That sounds good to me. I feel like I first 
really heard about or realised what a big

0:02:41.600,0:02:47.680
deal lenacapavir was from a tweet from you. 
So I get to be the lucky guy who gets to be

0:02:47.680,0:02:51.280
taught some of that history of HIV and how 
it all fits together from you. Hopefully I

0:02:51.280,0:02:58.400
can chip in some of my knowledge from working 
at Open Philanthropy on Global Health R&D as

0:02:58.400,0:03:05.280
well. That's this conversation. There's a lot to 
cover. I hope we cover the stuff that matters,

0:03:05.280,0:03:09.360
and get to the finish line of what are 
these magical new drugs really doing.

0:03:09.360,0:03:13.920
There are a lot of, I think, subplots that we're 
going to go through. The whole process of drug

0:03:13.920,0:03:23.200
discovery in this field has been really amazing. 
If you think about what HIV was like in the 1980s,

0:03:23.200,0:03:28.960
where people would only have a few years of 
survival after being diagnosed to now, where,

0:03:28.960,0:03:35.200
if people take treatment early enough, they can 
expect to live almost a normal life expectancy.

0:03:36.080,0:03:41.840
I think talking about how that's happened, how 
that's been made possible, is really important.

0:03:41.840,0:03:44.800
I agree. It sounds like we're 
going to have to start with:

0:03:44.800,0:03:47.440
what the heck is HIV itself? 
Should we begin there?

0:03:51.440,0:03:59.280
HIV is a virus that causes AIDS, the 
acquired immunodeficiency syndrome,

0:03:59.280,0:04:02.640
which is associated with lots of 
different infectious diseases,

0:04:02.640,0:04:09.360
cancers, and conditions that people suffer from 
if they've been infected with the virus for long

0:04:09.360,0:04:17.200
enough. But what's interesting about HIV, to begin 
with, is that it's not a typical type of virus.

0:04:17.200,0:04:26.400
It's something called a retrovirus. "Retrovirus" 
itself is a new concept to a lot of scientists,

0:04:26.400,0:04:34.880
historically speaking. The first retrovirus that 
infected humans was only discovered in 1979, which

0:04:34.880,0:04:41.920
is just two years before the first reported cases 
of HIV in the US. Before that, people had no idea-

0:04:41.920,0:04:43.440
And that was not HIV.
That was not HIV.

0:04:43.440,0:04:44.960
Okay. That was a different retrovirus.

0:04:46.400,0:04:52.080
That was human T lymphotropic virus, 
the first human retrovirus discovered.

0:04:52.080,0:04:57.600
I think that itself has a really interesting 
story. When I was first reading about this,

0:04:57.600,0:05:01.360
I was really struck by the fact that there 
was only this two year gap. We essentially

0:05:01.360,0:05:08.160
figured out what [human] retroviruses were in the 
first place just shortly before discovering this

0:05:08.160,0:05:14.560
deadly new disease that was caused by one. I 
think that's an important thing to think about,

0:05:14.560,0:05:19.280
when we're thinking about what scientists 
at the time would've been working on,

0:05:19.280,0:05:22.400
and how they would've figured 
out that it caused AIDS.

0:05:22.400,0:05:28.560
What is similar about that virus 
and HIV? What makes it a retrovirus?

0:05:28.560,0:05:35.520
A retrovirus — maybe I could kind of step 
back a bit and talk about how the usual

0:05:35.520,0:05:43.280
process of DNA works, just to give you some 
context. Great. Almost every cell in our body

0:05:43.280,0:05:51.760
has DNA. DNA is the genetic code to tell us 
which proteins to make, but it contains all

0:05:51.760,0:05:57.680
of that genetic code. All of our cells don't 
need to be producing all of those proteins,

0:05:57.680,0:06:02.160
and they don't need to be producing them 
all of the time. So instead of using the

0:06:02.160,0:06:09.760
entire genetic code, we use our enzymes to 
find segments of the DNA, to turn into RNA,

0:06:09.760,0:06:14.880
which is this intermediate molecule that's 
also used for various other things. And then

0:06:14.880,0:06:22.080
we turn this RNA into protein; proteins that 
are used in all kinds of processes in our body.

0:06:23.040,0:06:31.600
This direction — from DNA to RNA to protein 
— was how biologists and scientists typically

0:06:31.600,0:06:40.560
understood cells and how biological life worked. 
And that was overturned with the discovery of

0:06:40.560,0:06:46.720
retroviruses. What happened here was that 
people found out that there were certain

0:06:46.720,0:06:56.000
viruses that could turn their RNA into DNA, 
using an enzyme called reverse transcriptase.

0:06:57.040,0:06:59.360
They're in the other direction.

0:06:59.360,0:07:06.080
Yeah. This was a huge discovery in 1970 
by Howard Temin and David Baltimore. They

0:07:07.040,0:07:12.880
discovered this enzyme, reverse transcriptase. 
They discovered that retroviruses could reverse

0:07:12.880,0:07:21.040
transcribe RNA into DNA, and then they tried to 
find other retroviruses that infected humans.

0:07:21.040,0:07:28.480
For a long time, no one succeeded in finding 
any of these viruses, until 1979. So it was

0:07:28.480,0:07:37.520
about almost 10 years of people trying to find 
real examples of these, and they couldn't.

0:07:37.520,0:07:49.840
And is that related to HIV in 1981? We 
just got lucky? Or yeah, why so close?

0:07:49.840,0:07:55.200
So I think we did get lucky, and I think it 
would've been really difficult to figure out

0:07:55.200,0:08:04.400
that HIV was causing AIDS if not for that — or to 
even know where to look. What's really interesting

0:08:04.400,0:08:10.880
about this is, the scientist who discovered 
the first retrovirus that infected humans,

0:08:10.880,0:08:18.560
Robert Gallo, was also one of the scientists 
who discovered HIV as being the cause of AIDS.

0:08:20.720,0:08:24.880
I really like kind of reading 
through reviews, or retrospectives,

0:08:24.880,0:08:30.240
written by scientists themselves on how they 
figured out something, what research they did,

0:08:30.240,0:08:37.200
and so I was reading this retrospective that he 
wrote on the discovery of retroviruses and HIV,

0:08:37.200,0:08:40.880
and it was really interesting because 
he talks about how people were really

0:08:40.880,0:08:44.320
sceptical that there were any other 
retroviruses that infected humans.

0:08:44.880,0:08:51.200
The reason for that was: people had found 
retroviruses that infected other animals,

0:08:51.200,0:08:57.520
other primates. And in those primates, it seemed 
to be pretty abundant — or ubiquitous — across

0:08:57.520,0:09:02.960
their organs, but that wasn't the case in 
humans, and it was just hard to find them.

0:09:02.960,0:09:08.240
So they assumed that maybe there's something 
that prevents us from being infected by them.

0:09:09.520,0:09:19.760
Scientists also found that if you put the animal 
retroviruses into human blood — human serum — that

0:09:19.760,0:09:27.760
would immediately inactivate those viruses, using 
our complement system, which is one component of

0:09:27.760,0:09:32.720
our immune system. So this suggested that maybe 
there's some way that we're just protected

0:09:32.720,0:09:38.400
against them; they're not going to affect us. 
I think Robert Gallo just had this idea that

0:09:38.400,0:09:44.320
that might be wrong: maybe there are some other 
types of retroviruses that we hadn't studied.

0:09:44.320,0:09:52.160
And so he started looking at T-cell cancers.
In animals, the retroviruses that infected

0:09:52.160,0:09:58.000
those animals would typically cause T-cell 
leukemias; T-cells are a type of white

0:09:58.000,0:10:05.200
blood cell. So he thought maybe they're also 
causing leukemias in humans, and he started

0:10:05.200,0:10:13.920
working on finding patients with T-cell leukemias. 
Eventually he did actually discover a retrovirus

0:10:13.920,0:10:20.880
in them. This retrovirus was using reverse 
transcriptase to turn its RNA back into DNA.

0:10:20.880,0:10:30.960
And the reason that this is important for HIV is 
that they actually developed the tools — to test

0:10:30.960,0:10:38.240
out whether there are retroviruses infecting a 
sample — as part of that process. They started

0:10:38.240,0:10:45.520
working on potential drugs that could be used 
to target reverse transcriptase. They also just

0:10:45.520,0:10:51.200
generally had the idea that humans can be 
infected by retroviruses, and retroviruses

0:10:51.200,0:10:58.560
infect humans in their T-cells. And as we'll 
come to later on, that's very relevant in HIV.

0:10:58.560,0:11:04.640
Okay, so just stepping back for a second. 
Beforehand, scientists thought: well, maybe these

0:11:04.640,0:11:09.840
retroviruses that we're seeing in other animals 
aren't infecting humans, and the human immune

0:11:09.840,0:11:15.120
system's basically winning against them. We've 
got this under control. But then, it looked like

0:11:15.120,0:11:21.280
they may infect some cell types and we might not 
be winning fully. I mean, my question then is — it

0:11:21.280,0:11:28.640
sounds like such a good strategy for a virus. If 
I'm a virus, I would love to reverse transcribe

0:11:28.640,0:11:35.920
and integrate in your DNA. So how come if it's 
possible for HIV, we don't see this in lots of

0:11:35.920,0:11:42.720
viruses? I mean other viruses, other than HTLV, 
you mention, how come there are not lots and lots?

0:11:42.720,0:11:48.800
That's a really interesting question. And 
retroviruses, it turns out are really ancient.

0:11:49.520,0:11:57.520
Parts of their genomes are integrated into 
our normal DNA and they've just been passed

0:11:57.520,0:12:03.760
down over time — so this is called endogenous 
retroviruses. I don't really know much more

0:12:03.760,0:12:10.400
than that about that topic. But on your question 
on why aren't there more retroviruses infecting

0:12:10.400,0:12:18.880
humans? I think there's potentially three or four 
things going on. I am sort of wary of saying that

0:12:18.880,0:12:23.200
something's not possible, because sometimes 
people say that, even in the case of this,

0:12:23.200,0:12:28.800
and then they figure out that it's wrong.
We just haven't found those other retroviruses.

0:12:28.800,0:12:39.120
But some of the reasons, probably: one, it's like 
an error-prone process. If you're a virus — sorry,

0:12:39.120,0:12:49.440
you're not a virus. But for a virus that's 
transcribing its RNA genome into DNA,

0:12:49.440,0:12:56.320
the reverse transcriptase enzyme is not very 
precise in how it does that. It introduces

0:12:56.320,0:13:04.160
errors into the code. That's probably a little bit 
dangerous for the viruses themselves. Secondly,

0:13:04.160,0:13:11.840
they have this RNA genome, they then transcribe 
it into DNA, and then they get our cells to

0:13:11.840,0:13:18.480
transcribe the DNA back into RNA — which 
just seems a little bit inefficient. It

0:13:18.480,0:13:25.280
would just take a longer amount of time. It 
introduces errors. That's not super useful,

0:13:25.280,0:13:30.400
maybe, but obviously in some cases, it 
actually is, and it's worth that trade off.

0:13:30.400,0:13:37.280
You mentioned T-cells — it's going after these. I 
mean this is really, it's clever, but it's creepy.

0:13:37.280,0:13:44.720
So there's a reverse transcription where it's 
then going to integrate into my DNA, which is

0:13:44.720,0:13:50.720
disturbing in its own way. And then, additionally, 
you're telling me it's going to do that not just

0:13:50.720,0:13:56.080
anywhere, but in one of my immune cells — which 
is what's meant to be fighting infections. It's

0:13:56.080,0:14:02.240
going to hijack and integrate there. So 
is that right? And those are the T-cells?

0:14:02.240,0:14:13.120
That's right. So HIV infects various immune cells, 
but usually a specific type of T-cell called a CD4

0:14:13.120,0:14:25.760
T-cell. And the "CD4" just describes one receptor 
on the surface of this T-cell that is very crucial

0:14:25.760,0:14:32.480
in signalling. But also kind of defines that type 
of T-cell. And these types of T-cells- I guess I

0:14:32.480,0:14:37.280
should just describe what those actually are.
We have different types of white blood cells

0:14:37.280,0:14:44.480
in our body; T-cells are an important type. In 
this case, what they do is: they help the body

0:14:44.480,0:14:52.800
recognise pathogens or things that we've seen 
before, by presenting parts of that pathogen to

0:14:52.800,0:15:00.560
our other immune cells that can last much longer 
in the body and remember them if they appear

0:15:00.560,0:15:08.240
ever again. HIV is essentially infecting these 
quite important white blood cells in our body.

0:15:08.240,0:15:15.280
What happens then? We've entered the cell. Why 
is that a problem? How does it cause disease?

0:15:15.280,0:15:21.920
We've entered the cell. Well, there's quite a 
long process. Initially, when people get infected,

0:15:21.920,0:15:28.160
they have this short term infection — 
some kind of fevers, flu-like symptoms,

0:15:28.160,0:15:34.640
things like that; and the virus quickly 
replicates itself, multiplies into many

0:15:34.640,0:15:42.720
copies. Those copies then infect other T-cells. 
They then go into our lymph nodes, which are

0:15:43.520,0:15:48.160
basically these little hubs of immune 
activity. There's some in your neck,

0:15:48.160,0:15:52.720
some under your arms, other parts of 
your body. It infects these different

0:15:52.720,0:16:01.760
immune related tissues and depletes them.
And that means that people become more vulnerable

0:16:01.760,0:16:10.240
to all sorts of other infections that are normally 
mild to people, or just infections in general,

0:16:10.240,0:16:17.600
and some cancers. Our white blood cells are also 
useful in detecting tumour cells and trying to

0:16:17.600,0:16:25.440
eradicate them. And by depleting those important 
cells, people also have a higher risk of certain

0:16:25.440,0:16:31.360
cancers. So we have this short-term infection 
that immediately depletes a lot of our immune

0:16:31.360,0:16:37.200
cells. Eventually, there's this kind of slowdown 
in how much it replicates, and you get to this

0:16:37.200,0:16:43.440
equilibrium — but that equilibrium is still much 
worse than if someone hadn't been infected. And

0:16:43.440,0:16:52.800
over a long period of time, this reduction in 
immune cells means that people are vulnerable to

0:16:52.800,0:16:59.840
various diseases. And as time goes by, they get 
sicker and sicker from those other infections.

0:16:59.840,0:17:04.560
Before HIV, were there infections that were known

0:17:04.560,0:17:10.960
to cause cancers now something 
that we're more familiar with,

0:17:10.960,0:17:19.600
but was that connection a surprise and made it 
harder to figure out what the real cause was?

0:17:19.600,0:17:25.360
HTLV, the first retrovirus that was discovered 
just two years before that was the first

0:17:25.360,0:17:30.240
pathogen that was clearly causing cancer 
[in humans]. After that, there were, I mean,

0:17:30.240,0:17:35.280
there've been a bunch of other pathogens 
that are now known to be cancer-causing.

0:17:37.120,0:17:44.160
One is hepatitis B, which you do a lot of research 
on; HPV, human papillomavirus, that causes various

0:17:44.160,0:17:52.000
cancers including cervical cancer. There's 
Helicobacter pylori: this bacterium that causes

0:17:52.000,0:17:58.560
stomach cancer. I think there's more; maybe 
you remember some others. And there's HTLV.

0:17:58.560,0:18:00.080
Hepatitis C, which causes hepatitis.

0:18:00.080,0:18:02.640
Oh, hepatitis C. That's right.

0:18:02.640,0:18:06.240
Yeah. Those hepatitis viruses are sneaky.

0:18:06.800,0:18:15.760
So coming back to your question: the cancers 
were the first, I think, surprising thing about

0:18:15.760,0:18:22.400
people who had AIDS. One of the types of cancers 
that they became vulnerable to was called Kaposi

0:18:22.400,0:18:30.080
sarcoma, which is this tumour. What was surprising 
about that was: usually doctors who saw patients

0:18:30.080,0:18:37.040
with this type of cancer would see them in 
quite old- elderly patients, or people with,

0:18:37.040,0:18:41.680
I don't know, severe immune deficiencies 
and things like that. And in this case,

0:18:41.680,0:18:48.400
they were seeing them in young adults, which 
was really surprising. They were also quite

0:18:48.400,0:18:53.760
severe cancers: they were hard to treat with 
the usual treatments that were available.

0:18:54.960,0:19:01.680
And the fact that this was growing in prevalence 
was also really surprising and worrying to people.

0:19:01.680,0:19:07.360
So I think that was one of the first noticeable 
kind of warning signs that there was some kind of

0:19:07.360,0:19:16.320
epidemic spreading. It was probably an epidemic 
disease that was caused by some pathogen.

0:19:16.320,0:19:22.880
I think there were, maybe, a few months or a year 
or so before people realised that it was probably

0:19:22.880,0:19:30.640
caused by a virus. And I think the reason for that 
was that there were some cases of people being

0:19:30.640,0:19:38.640
infected through blood transfusion; so they had no 
other connection to other people with the disease,

0:19:38.640,0:19:44.640
and they had no other environmental risk factors 
or anything like that. But they had recently had

0:19:44.640,0:19:48.880
a transfusion, or an organ transplant, or 
something like that, and then suddenly,

0:19:48.880,0:19:55.920
they got infected. And the reason that this links 
to being a virus is because you can usually filter

0:19:55.920,0:20:05.600
out or purify some of the blood that you're 
using, or the organs, with filters that get rid

0:20:05.600,0:20:11.760
of bacteria — which are bigger. But that doesn't 
always work for viruses, which are much smaller.

0:20:11.760,0:20:15.440
So the viruses were getting through and they 
were still infecting people. And the fact that

0:20:16.880,0:20:21.680
this was infecting people far away, with no 
other connection, suggested that it was a virus.

0:20:21.680,0:20:25.360
I see. I mean, maybe that's a good 
point to just talk a bit about

0:20:25.360,0:20:31.280
transmission. So how does the virus 
transmit? I've got some ideas, but.

0:20:31.280,0:20:37.920
There are different types of modes of 
transport. One is, as we just talked about,

0:20:37.920,0:20:45.200
blood transfusions and organ transplants. So if 
there is contaminated blood with HIV, the risk

0:20:45.200,0:20:53.520
of an infection to someone else is quite high. 
It's some- I was looking at the data from the

0:20:53.520,0:21:00.320
systematic review and they estimate that the risk 
of infection is about 92% from a transfusion of

0:21:00.320,0:21:06.080
contaminated blood, which is quite high. Mother to 
child transmission? For mothers who are infected

0:21:06.080,0:21:16.080
with HIV, a quarter of them would pass HIV down 
to their baby, and this was before treatment

0:21:16.080,0:21:23.440
was available. Now, the chances are much lower if 
people use antiviral treatments around the time of

0:21:23.440,0:21:31.840
pregnancy and childbirth; but that was obviously 
very scary. And then there's sexual transmission,

0:21:31.840,0:21:38.880
which is probably the most common route that 
most people have heard of for HIV. And then,

0:21:38.880,0:21:45.280
finally, injection and drug use; using 
shared contaminated needles with HIV.

0:21:45.280,0:21:48.560
So before drug development, how come we 
need drugs? How come the immune system

0:21:48.560,0:21:54.000
doesn't control HIV better? I mean, I have a 
stereotype that it's extremely hard to control.

0:21:55.040,0:22:00.960
If infection gets established, it's 
really tough for us. So why is that?

0:22:00.960,0:22:04.560
One really interesting thing that 
I learned while trying to read

0:22:04.560,0:22:10.560
about this was that HIV is usually 
caused by a single virus particle,

0:22:11.360,0:22:16.160
and that's one particle replicates enough 
that it can cause a whole infection.

0:22:16.160,0:22:22.080
You're scaring me, Saloni. That's scary; 
that's scary. So do you mean that one

0:22:22.080,0:22:28.800
particle entering my bloodstream is all 
it takes, or you mean something else?

0:22:28.800,0:22:34.400
I actually mean something else. If you 
look retrospectively at people who have

0:22:34.400,0:22:40.160
HIV — you take a sample of their blood and 
you look at the different virus particles;

0:22:40.160,0:22:45.360
if you then trace back the 
genetics of those virus particles,

0:22:45.360,0:22:53.280
you then find that they all have one common 
ancestor, in around three quarters of cases.

0:22:53.280,0:22:54.240
I see. I see.

0:22:54.240,0:23:01.200
But that doesn't necessarily mean that just one 
particle is enough to infect someone, because we

0:23:01.200,0:23:07.600
have enough barriers in our immune system. I mean, 
if you think about a skin infection or something,

0:23:07.600,0:23:14.160
we have our skin, we have several layers of skin. 
We have immune cells protecting us within our

0:23:14.160,0:23:19.760
body. There are various barriers that prevent 
a bacterium or something from infecting us,

0:23:19.760,0:23:25.920
and that's the same is true with HIV. So I think 
what's happening here is that: if you think about

0:23:25.920,0:23:31.040
this from the perspective of probability, 
there are many barriers, but eventually one

0:23:31.040,0:23:36.560
of them might be able to cross all of those 
barriers and cause an infection, and if it's

0:23:36.560,0:23:43.600
able to do that, it can replicate very quickly.
I think that gets us to why it's difficult to

0:23:43.600,0:23:51.760
control an infection. Even though we do have these 
barriers, HIV is just very fast at replicating,

0:23:51.760,0:24:00.640
mutating. The reason that HIV can mutate so 
quickly — it basically becomes really genetically

0:24:00.640,0:24:07.680
diverse within a person who is infected, and 
that means that our immune cells might be able

0:24:07.680,0:24:15.200
to recognise some of the HIV strains that are 
in our body, but it's very difficult for it

0:24:15.200,0:24:22.480
to keep up with the rapid evolution and increased 
diversity. But there are several reasons for that.

0:24:22.480,0:24:30.000
One is the reverse transcriptase enzyme that we 
talked about that turns RNA into DNA; that enzyme

0:24:30.000,0:24:37.280
is not very precise, and that introduces errors. 
The errors allow it to potentially get beneficial

0:24:37.280,0:24:45.680
mutations sometimes, and that means that it can 
genetically diverge. The other is that the HIV

0:24:45.680,0:24:54.160
particle has two copies of RNA inside it, and 
those two copies can recombine with each other.

0:24:54.160,0:25:01.440
Wait, hold on, hold on, hold on. It 
comes in with two of the same thing?

0:25:02.160,0:25:04.800
Two of almost the same thing. It has two,

0:25:05.360,0:25:09.760
in the same way that we have two 
sets of chromosomes in our cells.

0:25:09.760,0:25:13.913
Okay, fair enough. I guess 
that's true, yes. Oh, wow, okay.

0:25:13.913,0:25:15.760
It has two copies of RNA.

0:25:16.320,0:25:19.580
It's not a double-stranded RNA; they're separate.

0:25:19.580,0:25:25.840
Yeah, that's right. They're two single-stranded 
RNA viruses, two single-stranded RNA particles;

0:25:25.840,0:25:32.080
and they can get reverse transcribed 
separately. They can also recombine

0:25:32.080,0:25:39.360
with each other. So I was reading this review 
paper about how all this worked — why there was

0:25:39.360,0:25:44.880
such rapid mutation — and they said- they were 
talking about this recombination and they said,

0:25:44.880,0:25:50.040
"This can be considered a primitive form 
of sexual reproduction." And I was —

0:25:50.040,0:25:55.600
No, no, no, no. This is a virus. If 
there's one thing I know it's that

0:25:55.600,0:26:02.520
that is too complicated. I don't 
believe this review paper. Okay.

0:26:02.520,0:26:10.880
Right, that's crazy. And there's a third thing 
actually, which is that our enzymes introduce

0:26:10.880,0:26:19.520
errors and mutations into HIV. We have this 
family of enzymes called the APOBEC family;

0:26:19.520,0:26:26.640
they insert mutations into HIV to try to 
damage it. What they do is: they change

0:26:26.640,0:26:35.520
the G's in our bases, in our DNA, into A; and 
they do that on a single-stranded DNA particle.

0:26:36.560,0:26:45.360
The HIV virus has these two single-stranded 
RNA molecules: they get turned into DNA,

0:26:45.360,0:26:50.880
then our enzymes introduce errors into 
it, by turning some of the G's into A's.

0:26:52.400,0:26:57.760
Just taking a step back, you're saying that 
the virus itself mutates a lot? So the reverse

0:26:57.760,0:27:04.080
transcription stage introduces errors, and that, 
actually, in a sense, helps the virus evade our

0:27:04.080,0:27:09.200
immune system. So it is introducing errors that 
make it hard for us, and then you're saying we

0:27:09.200,0:27:14.720
also are introducing errors that make it easier 
for us, or rather harder for the virus. So we are

0:27:14.720,0:27:19.040
both sort of fighting fire with fire of: I'm 
going to make you different, and it's saying,

0:27:19.040,0:27:24.320
no, I'm going to make myself different. And 
we're in a kind of ratchet situation there.

0:27:24.320,0:27:29.520
Yeah, it's very funny. I learned 
about this first, actually,

0:27:29.520,0:27:41.280
in 2022, this process, during the Mpox epidemic. 
I'd just been following the literature — what the

0:27:41.280,0:27:47.520
virologists are working out on the epidemic 
and so on — and they found out at that point

0:27:47.520,0:27:54.320
that the viruses from Mpox were mutating much 
faster than expected, than had been seen before.

0:27:55.680,0:28:00.960
And it turned out that the types of mutations that 
they were seeing, that were happening rapidly,

0:28:00.960,0:28:09.760
were very similar to this, APOBEC- kind of 
known-mutation change, and that led scientists

0:28:09.760,0:28:18.048
to think: maybe our enzymes are also working 
on this Mpox virus. So learning about that —

0:28:18.048,0:28:18.076
Oh I see what you're saying.

0:28:18.076,0:28:24.800
— testing that out, helped to figure out: 
we were introducing errors into the Mpox

0:28:24.800,0:28:31.920
virus very quickly, which made it mutate much 
faster than usual; and that knowledge helped

0:28:31.920,0:28:38.560
to figure out better when Mpox actually 
emerged, and when it started to spread.

0:28:38.560,0:28:42.960
Oh, interesting. You can somehow 
work backwards from that information.

0:28:42.960,0:28:43.920
Right.

0:28:43.920,0:28:50.080
Wow. Okay. Okay. Well, that's it. I am 
feeling grateful for my immune system;

0:28:50.080,0:28:55.040
has a bunch of tricks I didn't even 
know about, so thank you, immune system.

0:28:55.040,0:29:01.040
I think, one more thing is, the 
types of infections and cancers

0:29:01.040,0:29:07.760
that HIV makes people vulnerable to. I think, 
probably, that's not very obvious to people;

0:29:07.760,0:29:11.760
maybe they've heard of a few of them. 
So we talked about Kaposi sarcoma,

0:29:11.760,0:29:17.600
there are also a bunch of others. There's PCP, 
which is this fungal lung infection. There's

0:29:17.600,0:29:25.200
toxoplasmosis, which is this parasitic brain 
infection. There's cytomegalovirus retinitis,

0:29:25.200,0:29:32.240
that can cause blindness. There's tuberculosis 
— we already know about tuberculosis, probably,

0:29:32.240,0:29:38.320
most people here know about that — that is 
the leading cause of death in people with HIV

0:29:38.320,0:29:45.280
worldwide. The reason is that, because of this 
immune suppression, because of the fact that

0:29:45.280,0:29:52.160
HIV is depleting our immune cells, it makes us 
more vulnerable to infections like tuberculosis,

0:29:52.160,0:30:00.320
which are often easier to clear for people.
And then, some of these are also related — I think

0:30:00.320,0:30:06.800
I talked about a few that were brain infections 
or that cause blindness — some of these can also

0:30:06.800,0:30:14.160
cause AIDS-related dementias, where people lose 
their ability to think, and make decisions, and

0:30:14.960,0:30:21.680
generally lose their memory. And then there are 
a bunch of cancers as well that HIV is associated

0:30:21.680,0:30:29.200
with. As we talked about earlier, some of our 
T-cells are really important in trying to find

0:30:29.200,0:30:35.440
potentially cancerous cells and eliminating them. 
If those T-cells are depleted, that makes us more

0:30:35.440,0:30:41.360
vulnerable to cancers continuing to grow. So that 
includes Kaposi sarcoma, which we talked about,

0:30:41.360,0:30:50.400
and then non-Hodgkin lymphoma, and it also 
increases the risk of cervical cancer.

0:30:50.400,0:30:54.080
So I think I have a rough understanding 
of the virus and a rough understanding of

0:30:54.080,0:30:59.040
how it leads to AIDS, and to disease, and 
opportunistic infections. Maybe quickly,

0:30:59.040,0:31:03.520
where did it come from originally? 
Where did this virus start out?

0:31:03.520,0:31:11.760
So... this, I think, has had a lot of research 
go into it, and the likely answer is that it

0:31:11.760,0:31:22.880
came from chimpanzees and gorillas in Central and 
Western Africa. There are different types of HIV.

0:31:22.880,0:31:28.640
As we talked about, HIV mutates very quickly; 
creates this huge amount of genetic diversity,

0:31:28.640,0:31:35.120
and that also means that there are multiple 
different types of HIV. The one that's found

0:31:35.120,0:31:42.480
across the world is called HIV-1. There's 
also an HIV-2. Both of them are from different

0:31:42.480,0:31:50.720
kinds of primates from Central and Western 
Africa. And I think the current understanding

0:31:50.720,0:32:03.200
is that HIV one came from a type of chimpanzee 
somewhere near the southeastern area of Cameroon.

0:32:03.760,0:32:12.080
And probably — so this is interesting: the way 
that we understand this is by collecting a lot

0:32:12.080,0:32:21.200
of samples of HIV from different people, ideally 
as early as possible. The earlier the cases are,

0:32:21.200,0:32:28.080
the easier it is to try to estimate where they 
all came from, or when they converged back in

0:32:28.080,0:32:36.480
history. What was super interesting to me, that I 
was reading about recently, was that the earliest

0:32:36.480,0:32:48.960
genome of HIV was recovered from someone who died 
in 1966 in Africa. That's 15 years before the

0:32:48.960,0:32:55.680
first cases were reported. There were definitely 
cases of HIV before that; probably for decades

0:32:55.680,0:33:02.240
before that. If you use this sample, but also 
all of the other early samples that we have,

0:33:02.240,0:33:09.040
you can kind of trace back where they have shared 
ancestry — in the same way that you might be able

0:33:09.040,0:33:16.560
to do with a family tree. If you do that with 
genomics, you can try to trace back where they

0:33:16.560,0:33:25.360
have similarities, and that suggests that the 
pandemic originated at the turn of 20th century.

0:33:26.320,0:33:28.720
Hundred years ago.

0:33:28.720,0:33:36.560
Yeah. Well, so we don't have — because 
the earliest case we have is from 1966,

0:33:36.560,0:33:41.920
that's still not very early, and that 
means that there's still some uncertainty

0:33:41.920,0:33:46.000
in when it actually originated. 
There's this uncertainty between,

0:33:46.000,0:33:53.280
so somewhere between 1881 and 1918 is 
probably around when it first emerged.

0:33:53.280,0:34:00.720
Okay, well, and in humans. So basically, there 
was this virus that infected other primates,

0:34:00.720,0:34:06.400
chimps; maybe goes back much longer, I don't 
know if that's known. And then at some point,

0:34:06.400,0:34:10.480
someone was probably hunting 
a chimp and there was a blood,

0:34:11.120,0:34:20.480
the blood sort of got into their food, or was 
it, you're eating uncooked chimp or? Oh God.

0:34:20.480,0:34:28.320
That's probably it. So probably through hunting 
primates and consuming them from the, kind of,

0:34:28.320,0:34:36.080
butchery of that process, and being exposed to 
their infected body fluids. So the virus that

0:34:36.080,0:34:43.280
infects chimpanzees and other monkeys is called 
Simian Immunodeficiency Virus rather than human.

0:34:43.920,0:34:44.720
SIM [SIV], okay.

0:34:44.720,0:34:53.120
That, it seems like, has crossed over 
to humans dozens of times over history.

0:34:53.120,0:34:57.520
As in, since the 1900s?

0:34:57.520,0:34:57.920
In total.

0:34:57.920,0:35:00.400
Okay. Okay. Interesting. I didn't know that.

0:35:01.040,0:35:08.080
And I think there's around four of those 
"spillover events", is what they're called.

0:35:08.080,0:35:13.520
At least four of those spillover events 
have led to sustained epidemics that —

0:35:13.520,0:35:13.960
Oh my gosh.

0:35:13.960,0:35:18.160
— people are still infected by today.

0:35:18.160,0:35:25.360
Wow, okay. So it's been with us for a long time, 
and we haven't known the extent of it so well

0:35:25.360,0:35:32.720
as a species until the more recent epidemic 
starting in the '80s. Is that a fair summary?

0:35:32.720,0:35:37.040
That is a fair summary. I think, this 
really reminds me of some of the work

0:35:37.040,0:35:44.480
and writing I do about missing data.
I mean, on a lot of topics in health, but

0:35:44.480,0:35:52.160
also in other areas, we tend to have much better 
data collection and understanding of epidemics,

0:35:52.160,0:36:00.800
but also other diseases, in richer countries, 
because of the institutions that can collect

0:36:00.800,0:36:12.880
that data. Having the resources and the people to 
collect that data is not super easy. The fact that

0:36:12.880,0:36:20.560
we know about the first cases of AIDS that had 
been reported were in the US, is not because it

0:36:20.560,0:36:29.360
started in the US — it's because the US had good 
detection and disease control research going on.

0:36:29.360,0:36:34.800
Oh yeah. I mean, it reminds me of the 
extremely frustrating initial COVID graphs,

0:36:34.800,0:36:41.840
where reports of number of cases would actually be 
number of confirmed cases — vastly different than

0:36:41.840,0:36:48.400
what you could interpolate must be happening 
in reality. But, you know, whoever gets tested

0:36:48.400,0:36:55.280
ends up getting reported and whoever doesn't 
doesn't. It can lead to missed inferences.

0:36:55.280,0:36:56.880
Exactly. That's right.

0:36:56.880,0:37:05.120
I feel like we should get to drugs. And as you've 
described HIV so far, as a drug developer, which

0:37:05.120,0:37:12.960
I've never been yet, but moonlight as in my head, 
I'm thinking: the different parts you've described

0:37:12.960,0:37:20.480
are each potential targets that I could maybe 
make a chemical small molecule to interfere with,

0:37:20.480,0:37:27.040
to sort of mess up its lifecycle. I know 
my immune system's not going to sort it

0:37:27.040,0:37:34.000
out all on its own. So some of those non-human 
chemicals might be pretty useful too. So firstly,

0:37:34.640,0:37:38.160
is that right? And then, am I thinking about 
that right? And maybe that means we're going

0:37:38.160,0:37:43.680
to have to go in a bit more detail about the 
infection cycle, because the virus is going to

0:37:43.680,0:37:51.440
look different at different points and that will 
maybe give us a clue about what drugs we can make.

0:37:51.440,0:37:57.680
No, exactly. I think, actually, it would 
be probably easier to think about where

0:37:57.680,0:38:03.760
scientists were at the time, how much they 
knew, and how they were developing drugs,

0:38:03.760,0:38:07.840
and then we can talk about the 
broader life cycle of the virus.

0:38:07.840,0:38:12.880
Let's do it. Sounds good.

0:38:12.880,0:38:23.440
So place yourself in the 1980s. This is kind of 
hard because, as a scientist, you had just found

0:38:23.440,0:38:32.480
out that retroviruses could infect humans. Now you 
find out that HIV is a retrovirus. You're like,

0:38:32.480,0:38:40.080
wait, I didn't even know that RNA could be turned 
into DNA until 10 years ago. So that's quite a

0:38:40.080,0:38:47.600
tricky position to be in. At this point, in the 
early 1980s, there were no antiviral drugs for

0:38:47.600,0:38:57.760
retroviruses. Also, antivirals in total were kind 
of new. I did not know this before reading about

0:38:57.760,0:39:07.160
it for this episode, but the first antiviral drug 
of any kind was approved in 1963, which is, again-

0:39:07.160,0:39:10.640
1963, really? Not even flu. We 
didn't have nothing. That's crazy.

0:39:10.640,0:39:16.880
No, we didn't. We had flu vaccines. We had a bunch 
of vaccines before that. But the first antiviral

0:39:16.880,0:39:24.960
drug was for treating the herpes simplex virus 
— infections of the eye — and that was in 1963.

0:39:24.960,0:39:25.680
Wow.

0:39:25.680,0:39:31.760
Also, if you were in the early 1980s, 
you wouldn't have PCR. PCR is polymerase-

0:39:31.760,0:39:33.263
Of course. Yes, of course.

0:39:33.263,0:39:39.120
-chain reaction, which is used to multiply 
samples of genetic material so you can study it

0:39:39.120,0:39:46.240
more easily. And that would've been really useful 
for being able to detect the level of infection,

0:39:46.240,0:39:52.800
and the level of virus in someone. You would 
be very new to knowing about retroviruses to

0:39:52.800,0:39:56.880
begin with. You wouldn't have any antivirals 
for them. You wouldn't have a great idea of

0:39:56.880,0:40:03.120
how to even develop antivirals at all. 
You didn't have PCR, you didn't have

0:40:03.120,0:40:09.200
multi-center trials, so you wouldn't be 
able to test drugs in multiple hospitals.

0:40:09.200,0:40:09.440
Multi-center? Oh, I see. Got it. Yep, got it.

0:40:09.440,0:40:15.600
You wouldn't be able to test whether 
people were resistant to the virus,

0:40:15.600,0:40:24.960
through the genetics of the virus, again, because 
of the lack of PCR testing. And so there was just

0:40:24.960,0:40:32.720
this complete lack of — what are we going to do 
now? We have no idea. This is super new to us.

0:40:32.720,0:40:39.760
I think a lot of people at the time might've 
thought it's impossible to treat this disease.

0:40:39.760,0:40:46.320
It's so new, it's so different. How 
are we gonna make any progress on it?

0:40:46.320,0:40:50.240
You don't have a proof point 
you can hang your hat on.

0:40:50.240,0:40:52.000
That's right.

0:40:52.000,0:40:55.200
Very... yeah, the unknown beckons.

0:40:55.200,0:41:02.320
And so again, I was trying to find retrospective 
written by someone who works in this field,

0:41:02.320,0:41:06.800
and I found one that was really 
interesting by Samuel Broder.

0:41:06.800,0:41:13.760
He was one of the scientists who 
developed AZT, the first HIV drug.

0:41:13.760,0:41:14.260
AZT.

0:41:15.120,0:41:21.440
His team found that it was effective against 
HIV. They also developed various other drugs

0:41:21.440,0:41:29.360
against it. And he wrote this retrospective on 
how they discovered the first antivirals and

0:41:29.360,0:41:34.480
how pessimistic people were at the time — that 
it was possible to make any treatments against

0:41:34.480,0:41:43.760
it. So maybe a bit of background on who he was. 
Samuel Broder was this cancer and immunology

0:41:43.760,0:41:51.360
researcher at the National Cancer Institute in 
the US. And this itself is quite interesting.

0:41:51.360,0:41:57.200
You're thinking HIV, this is an infectious 
disease. But the people who were studying it,

0:41:59.040,0:42:03.760
who made these first effective 
antivirals, were cancer researchers.

0:42:03.760,0:42:10.800
So he was part of the NIH, the division that 
works on cancer. And so he was employed by

0:42:10.800,0:42:17.360
the US government and had a lab in the 
cancer part. And okay, got it. Keep going.

0:42:17.360,0:42:21.920
So I think the reason that he 
was working on it was, as I said,

0:42:21.920,0:42:28.480
the first thing that was noticed in AIDS patients, 
that was surprising to people, was Kaposi sarcoma,

0:42:28.480,0:42:35.600
this type of cancer of the skin. And so he was 
trying to figure out what was going on here.

0:42:37.520,0:42:46.960
I read this book about the drug development in 
HIV and AIDS called How to Survive a Plague. And

0:42:46.960,0:42:55.840
there they describe this period very humorously 
to me. I mean, it's obviously not humorous,

0:42:55.840,0:43:02.960
but the way that they say it, was that he 
was really excited to see this first HIV

0:43:02.960,0:43:09.920
patient come to the National Cancer Institute. 
He saw it as this once-in-a-lifetime scientific

0:43:09.920,0:43:15.840
challenge that brought together two of his 
interests of immunodeficiency and cancer.

0:43:15.840,0:43:22.160
And they thought, if we're able to crack 
this, this is going to be really important.

0:43:22.160,0:43:32.080
So I think the first thing they noticed was 
the virus was probably infecting CD4 T-cells.

0:43:32.080,0:43:38.320
The reason that they thought this was one of the 
first signs that you would see in someone who was

0:43:38.320,0:43:46.720
infected was that their T-cell count would drop. 
They at least had some blood testing and they had

0:43:46.720,0:43:53.280
some of the tools and technology to measure CD4 
T-cells at the time. So they knew that there's

0:43:53.280,0:43:59.520
this massive drop, and they thought maybe it's 
because the virus is multiplying in these cells.

0:43:59.520,0:44:02.000
Seems like a link to make.

0:44:03.280,0:44:07.840
So they thought, okay, let's — we've 
hypothesised that the virus is replicating

0:44:07.840,0:44:14.960
in these cells — maybe we should test compounds 
to interrupt this process, to prevent it from

0:44:14.960,0:44:22.720
replicating in these CD4 T-cells. And so he then 
approached different pharmaceutical firms to try

0:44:22.720,0:44:28.720
to get funding to work on this project, and 
also to potentially commercialise a drug if

0:44:28.720,0:44:37.280
they found one. I think he approached several 
firms. Most of them said no, but one of them

0:44:37.280,0:44:43.989
said yes. And that company was Borroughs 
Wellcome. So we probably haven't heard of-

0:44:43.989,0:44:47.760
Yes, which now has a philanthropy, 
Burroughs Wellcome fund.

0:44:47.760,0:44:53.920
And this was a pharmaceutical firm that no 
longer exists. It was merged into what became

0:44:53.920,0:45:00.560
GSK or GlaxoSmithKlein, and they were the 
only company at this time that were willing

0:45:00.560,0:45:10.400
to consider funding or commercialising HIV 
drugs. But they were very afraid that their

0:45:10.400,0:45:17.200
researchers or their scientists would 
get infected with stab that people were

0:45:17.200,0:45:22.400
working with. So they refused to work with 
these live virus samples, and they said-

0:45:22.400,0:45:22.960
Oh my god.

0:45:22.960,0:45:26.960
-nope, you've got to work on it yourself. 
They were saying this to Samuel Broder and

0:45:26.960,0:45:33.600
his team. And so Samuel Broder's team had 
to do all of the screening of these drugs,

0:45:33.600,0:45:39.280
running the trials all on their 
own, despite Burroughs Wellcome

0:45:39.280,0:45:44.312
then getting the credit for it, and 
then being able to commercialise it.

0:45:44.312,0:45:51.440
Oh god. Well, it reminds me of streptomycin, the 
first TB drug, where there was Albert Schatz,

0:45:51.440,0:45:56.640
the PhD student, and Waxman, what was his 
first name? Henry Waxman, something? Waxman-

0:45:56.640,0:45:57.760
Selman Waxman.

0:45:57.760,0:46:03.760
Selman Waxman, there we go, the professor 
who then got the Nobel Prize. They didn't

0:46:03.760,0:46:06.400
share it in the end, did they? I don't 
know. I'm going to forget the story.

0:46:06.400,0:46:12.000
They didn't share it. I also read the story 
because I was writing about antibiotics,

0:46:12.000,0:46:15.440
and it was really interesting, because he

0:46:15.440,0:46:20.640
discovered this group of bacteria 
that produced antibiotics, right?

0:46:20.640,0:46:22.080
Right, in soil? Yeah.

0:46:25.120,0:46:29.200
It's so weird to think about bacteria that are 
producing antibiotics, but basically they're

0:46:29.200,0:46:36.240
producing it to compete with other bacteria. 
And somehow he found that this specific type,

0:46:36.240,0:46:41.120
or this group, of bacteria were producing a lot 
of antibiotics. He thought, okay, maybe there's

0:46:41.120,0:46:45.920
something there. They seem to be killing the other 
bacteria around them, maybe we could use that as a

0:46:45.920,0:46:51.600
treatment for our own bacterial infections. 
And he started to recruit PhD students to

0:46:51.600,0:46:56.231
work on that. One of them was Albert Schatz. 
And what was really interesting about this-

0:46:56.231,0:46:58.000
This was Second World War era, right?

0:46:58.000,0:47:04.640
This was during the Second World War. Well, the 
1930s, I think, and he had Albert Schatz start

0:47:04.640,0:47:11.989
to work on this project in a basement's room 
that he never, I think, Waxman never visited.

0:47:11.989,0:47:16.400
He's like, try this out on 
some TB, I'll be upstairs.

0:47:16.400,0:47:19.360
And so Schatz was working on this,

0:47:19.360,0:47:24.320
and then what was really strange was 
that Schatz was drafted into war!

0:47:24.320,0:47:24.880
Oh, yes.

0:47:24.880,0:47:30.800
And the project, basically, was on 
pause for a few months. Apparently

0:47:30.800,0:47:33.754
he then got a back injury and then was sent home.

0:47:33.754,0:47:34.880
Thank goodness!

0:47:34.880,0:47:37.920
And thank goodness, he discovered streptomycin.

0:47:37.920,0:47:38.720
Wow.

0:47:38.720,0:47:41.440
That was the first, I think, first 
antibiotic compound that was found

0:47:41.440,0:47:45.920
from this type of bacteria, 
which is called actinomycetes,

0:47:45.920,0:47:52.240
and that group also led to the 
discovery of various other antibiotics.

0:47:52.240,0:47:59.120
Okay. Well, Albert Schatz, legend. Grad students 
have been abused for decades, it turns out. And

0:47:59.840,0:48:04.880
I guess, back to HIV, which we are currently 
dealing with, it sounds like a very virtuous

0:48:04.880,0:48:09.120
academic group. And I don't want to insult 
Burroughs Wellcome. I'm very glad they brought

0:48:09.120,0:48:14.720
this to market. So kudos that there was at least 
one company willing to stand up. So maybe I'll

0:48:14.720,0:48:20.400
reserve my vitriol in case there are other bad 
forces I need to get mad at, later in the story.

0:48:20.400,0:48:25.040
Well, yeah. So this is also not quite as similar

0:48:25.040,0:48:30.480
because I think Schatz and Waxman 
then got into this big fight.

0:48:31.120,0:48:36.720
Yeah, I remember this, yeah. And Waxman was 
withholding royalties from Merck or yeah.

0:48:36.720,0:48:40.000
There was something like that. But in this case,

0:48:40.000,0:48:45.360
I think they seem to work together fine. 
In this case, in Samuel Broder's team,

0:48:45.360,0:48:50.800
there were a few scientists who were really 
involved in this. One was Hiroaki Mitsuya,

0:48:50.800,0:48:58.880
and he was doing the day-to-day research on 
potential drugs that could work against HIV.

0:48:58.880,0:49:05.360
And I said, day to day, but actually they were 
doing this research in the night, after the other

0:49:05.360,0:49:11.584
colleagues at the National Cancer Institute went 
home, because apparently they were also afraid-

0:49:11.584,0:49:11.600
Oh my gosh. Wow.

0:49:11.600,0:49:19.520
-of potentially being contaminated or 
getting infected. Also, there was another

0:49:19.520,0:49:31.280
scientist called Robert Yarchoan. So these 
two scientists tested over 180- well, okay,

0:49:31.840,0:49:37.040
let's back up to what we said before. You're 
in the 1980s, you have no idea how to tackle

0:49:37.040,0:49:43.280
this disease. You have various reasons to doubt 
whether you can even develop a treatment at all.

0:49:43.280,0:49:47.440
And the only clue that you have at this 
point, really, is that it infects CD4

0:49:47.440,0:49:55.680
T-cells. So that probably helps you to test 
things in a lab. You can probably test how

0:49:55.680,0:50:01.680
these different drugs affect HIV's ability to 
infect these CD4 T-cells, but you don't really

0:50:01.680,0:50:08.160
have anything else to go on. So what would 
you do? And the thing that they did was,

0:50:08.160,0:50:13.360
they just tried anything. They just 
tried any drug compounds that they had.

0:50:14.320,0:50:20.960
So Mitsuya tested over 180 different compounds 
and they would be coded with different code names,

0:50:21.520,0:50:28.080
and you would ask people for whatever they 
thought might be potentially effective.

0:50:28.960,0:50:33.920
So there's no unifying hypothesis per se, it's 
more like, ask around, see what people think might

0:50:33.920,0:50:40.160
work, see what you have lying in the fridge 
at the cancer institute, that kind of thing?

0:50:40.160,0:50:44.480
Well, even if you did have hypotheses, 
many of them would just turn out to

0:50:44.480,0:50:47.760
not work and then you would have 
to try something else. So okay,

0:50:47.760,0:50:52.720
there were a bunch of hypotheses, but there 
were also just, let's just see what happens,

0:50:52.720,0:50:57.360
let's just use this trial-and-error kind 
of approach and see if anything works.

0:50:58.000,0:51:03.360
Well, now I think about high-throughput 
screening where you screen hundreds of thousands,

0:51:03.360,0:51:07.600
sometimes more, drug candidates 
or molecules all at once. Should

0:51:07.600,0:51:13.840
I be visualising that, or that comes much 
after, and we're dealing with hundreds?

0:51:13.840,0:51:20.880
Well, yeah. So I think this might've been before 
high-throughput screening became much more

0:51:20.880,0:51:27.120
popular. This is probably early days where you're 
doing this one at a time. You have these different

0:51:28.080,0:51:33.920
cultures of the virus in the lab and you're 
just testing out random drugs, one for each

0:51:33.920,0:51:40.000
one or whatever, and maybe you have a bunch for 
each one, just to see if it's really working.

0:51:40.000,0:51:47.040
And at one point they find this one little 
vial, or a bottle, with something called

0:51:47.040,0:51:55.360
"compound S" — so that's the code name — and that 
somehow seems to keep the infected cells alive.

0:51:55.360,0:52:05.680
So this was a real breakthrough and this compound 
turned out to be AZT, or azidothymidine. What was

0:52:05.680,0:52:11.680
this compound? This was a compound that already 
existed, and it was developed in the 1960s — in

0:52:11.680,0:52:20.960
1964 — by another cancer researcher called Jerome 
Horwitz. And I had found- I was trying to look up,

0:52:20.960,0:52:27.440
who was the discoverer of each important antiviral 
in HIV? And I found this name, and I thought,

0:52:27.440,0:52:31.360
okay, well let me try to find a retrospective 
written by him and I couldn't find one.

0:52:32.240,0:52:38.480
And the reason was that, when he was working 
on this in the 1960s, he was developing this

0:52:38.480,0:52:44.800
as a potential cancer drug. He had this 
idea that — if you think about cancers,

0:52:44.800,0:52:50.480
the thing that people know about cancers is 
they grow quickly. They have these tumours

0:52:50.480,0:52:55.360
and the tumours grow quickly, and the 
way that they do that is by replicating;

0:52:55.360,0:53:04.000
they need to replicate their DNA in order to 
divide. So his idea was, if you have the DNA code,

0:53:04.000,0:53:13.280
and the DNA code is duplicated by adding 
these bases one at a time, by our enzymes,

0:53:13.280,0:53:21.600
into this longer DNA structure. What if, instead 
of a normal base, you had a fake base that was

0:53:21.600,0:53:30.360
kind of like a normal base except it didn't allow 
any more bases to join to it. And so he found-

0:53:30.360,0:53:31.120
Sounds clever.

0:53:31.120,0:53:40.720
-a compound that, it essentially was this type of 
fake base — di-deoxynucleoside. And he thought,

0:53:40.720,0:53:44.720
okay, maybe this is going to stop the 
cancers from growing. But it turned out,

0:53:44.720,0:53:50.800
it didn't work for cancer, and he was so 
disappointed with it that he apparently

0:53:50.800,0:53:56.160
threw away his lab notes — essentially 
just trashed it and forgot about it;

0:53:56.160,0:54:01.760
didn't even apply for a patent. So it was 
just in the National Cancer Institute,

0:54:01.760,0:54:07.280
where he also worked, and it was just there as 
one of the compounds that had been developed.

0:54:07.280,0:54:11.520
Don't throw away your lab notes. Don't 
throw away your lab notes. But I'm glad

0:54:11.520,0:54:17.520
he didn't throw away the samples? 
That sounds great to me. I'm happy.

0:54:19.200,0:54:25.360
We said that the AZT — this new 
compound — it was able to mimic

0:54:25.360,0:54:32.640
the bases in our DNA. So why did it 
work for HIV, but not for cancer?

0:54:32.640,0:54:38.880
That's something I don't know. But what it 
does here is almost the same process. When

0:54:38.880,0:54:47.200
HIV's reverse transcriptase is turning the RNA 
into DNA, so that it can integrate into our own

0:54:47.200,0:54:53.920
genome, it introduces this fake base, 
which blocks the chain from getting longer;

0:54:53.920,0:55:00.960
it blocks the rest of the DNA from forming, 
and that halts the virus's replication.

0:55:00.960,0:55:03.400
That's epic. Go AZT.

0:55:03.400,0:55:10.400
Super interesting. Yeah, it was just, I think 
it's this trial-and-error approach that sometimes

0:55:10.400,0:55:16.800
works. What's really useful about things like this 
is that, once you do find a compound that works,

0:55:16.800,0:55:24.480
you can then try to make modifications that 
create new related drugs that you now will

0:55:24.480,0:55:29.760
hope will also work. That is possible because 
if there's something about the structure that

0:55:29.760,0:55:34.480
is allowing it to have this function, then 
making these different modifications could

0:55:34.480,0:55:39.520
lead to additional compounds, or maybe it could 
make it more effective, or more safe in some way,

0:55:39.520,0:55:45.200
and so you could now have this wider 
range of compounds that you can work with.

0:55:45.200,0:55:53.360
And my stereotype of AZT is that it was not 
very safe, in the sense of, many side effects?

0:55:53.360,0:56:00.000
That's right. Yeah, I mean, I think 
if you were a patient at the time,

0:56:00.000,0:56:09.520
you would still see it as much better than the 
prospects of a continued progression of HIV,

0:56:09.520,0:56:19.360
but it was pretty toxic. It affected people's bone 
marrows, it led to anaemia; also just made them

0:56:19.360,0:56:26.720
feel quite physically weak in some ways. But it 
did clear out some of the virus from their bodies;

0:56:26.720,0:56:33.680
it restored their immune function, it cleared 
infections. One interesting thing I read was

0:56:33.680,0:56:41.760
that it, surprisingly, also reversed some of 
these AIDS-related dementias that I mentioned.

0:56:42.480,0:56:48.880
Because those dementias were actually caused by 
infections, if you can clear some of the HIV,

0:56:48.880,0:56:55.280
which is reducing your immune function, which was 
previously suppressing these infections — if you

0:56:55.280,0:57:02.080
can, kind of, revert that immune depletion, 
then you can now fight off these infections

0:57:02.080,0:57:09.760
that caused brain dysfunction and so on.
This was really astonishing, I think,

0:57:09.760,0:57:15.120
according to Samuel Broder, to the doctors 
who saw people who were being treated with

0:57:15.120,0:57:19.600
AZT at the time — they were genuinely 
shocked that this was even possible.

0:57:19.600,0:57:26.160
And we went from not knowing if any antiviral 
was going to be possible to... you're actually

0:57:26.160,0:57:35.200
seeing as a doctor people reverse even 
some of the cognitive effects. What a time.

0:57:35.200,0:57:41.040
What a time, yeah. And also, the other 
downside was, not just the side effects,

0:57:41.040,0:57:47.680
but this drug seemed to work for at least a few 
months in people, and then they would start to get

0:57:47.680,0:57:54.720
worse again, and the reason was that HIV would 
find a way to evade the action of this drug.

0:57:54.720,0:58:01.520
Because it was mutating so quickly, it was able 
to find ways to either get rid of this drug, or to

0:58:01.520,0:58:09.200
develop certain changes in its proteins that would 
mean that the drug was no longer able to work.

0:58:09.760,0:58:17.360
So that rapid mutation that made it so hard 
for our immune system to operate against HIV,

0:58:17.360,0:58:24.160
now is making it hard for AZT 
to durably operate against HIV.

0:58:24.160,0:58:27.520
But it was really important 
because it was the first

0:58:27.520,0:58:32.720
drug. It was a drug against a disease 
that people thought was untreatable.

0:58:32.720,0:58:33.040
Yes, totally.

0:58:33.040,0:58:39.600
And this completely shifted the perception 
of the disease. Samuel Broder has this line

0:58:40.640,0:58:45.520
in his retrospective review where he says, 
"The question at that point was no longer

0:58:45.520,0:58:51.680
whether HIV-1 could ever be successfully 
treated, but rather how fast more therapies

0:58:51.680,0:58:58.880
could be developed." And their drug, AZT 
moves from research in the lab to drug

0:58:58.880,0:59:03.920
approval within just two years, and this 
is partly a result of how the trials work,

0:59:03.920,0:59:09.200
but it's also partly because of activism 
around trying to make it available quickly.

0:59:09.200,0:59:12.560
I mean, it's... that's both so inspiring and

0:59:12.560,0:59:20.720
so infuriating. So when was AZT 
available, did you say? In 19-?

0:59:20.720,0:59:22.720
I think 1987.

0:59:22.720,0:59:27.600
1987, okay, so from 1981 to 1987. 
If the clock starts at 1981,

0:59:27.600,0:59:31.920
when HIV was discovered; if there 
had been more energy earlier,

0:59:31.920,0:59:41.440
more funding, more support, if it only took 
two years once you started investigating...

0:59:41.440,0:59:42.623
It is really frustrating.

0:59:42.623,0:59:44.720
Anyway, I should celebrate 
it was two years, but...

0:59:44.720,0:59:51.840
It's really frustrating because, so I have 
been reading this book, the audiobook version,

0:59:51.840,1:00:00.160
of How to Survive a Plague by David France, which 
is this amazing, very well written book on drug

1:00:00.160,1:00:07.840
development. "How scientists and activists came 
together to treat AIDS" is the tagline, I think.

1:00:08.960,1:00:20.320
It starts, I think, in 1981, and it's genuinely 
so depressing to read it — obviously — for several

1:00:20.320,1:00:26.800
years. You're getting through this book, and 
you're just so frustrated with how slow people

1:00:26.800,1:00:32.160
are; how unresponsive, how much they don't 
treat it as an urgent problem — even when

1:00:32.160,1:00:38.800
it's clearly an epidemic disease that's growing 
exponentially over time. That people are just

1:00:38.800,1:00:44.400
unwilling to consider that there are potential 
treatments out there, or they're in these petty

1:00:44.400,1:00:50.080
arguments with each other about what we should be 
doing. Should we be saying enough? Are we scaring

1:00:50.080,1:00:56.800
people by telling them that this is a deadly 
disease? And so on. And it was just, it was really

1:00:56.800,1:01:00.400
frustrating to read. It was a very well-written 
book, but it was very frustrating to read.

1:01:01.280,1:01:08.080
I remember I read "And The Band Played On" 
by Randy Shilts covering some of the initial

1:01:08.080,1:01:14.080
years and had the same experience 
— just a very, very tough read.

1:01:14.080,1:01:21.120
But I think one interesting thing about Samuel 
Broder and the, kind of, cancer approach to

1:01:21.120,1:01:27.680
studying HIV is that, I think, — so we said 
that the reason that they were studying this

1:01:27.680,1:01:32.800
was because of Kaposi's sarcoma, which was 
one of the cancers that HIV made people more

1:01:32.800,1:01:38.320
vulnerable to. But I actually think that being a 
cancer researcher was probably the right mindset

1:01:38.320,1:01:43.360
that you needed to have, as a scientist, 
if you wanted to develop drugs against HIV.

1:01:44.560,1:01:49.920
One reason for that was, cancer research 
at the time, they were, I think,

1:01:49.920,1:01:57.280
the only group in the NIH that were experienced 
with drug development. But the other was just:

1:01:57.280,1:02:03.360
you're facing this horrible disease 
that's very rapidly progressing,

1:02:04.000,1:02:11.200
similar to cancer. You're also in the situation 
where action is much more important than inaction,

1:02:11.200,1:02:16.320
because it's just going to get worse. You're 
also in the situation where you're willing to

1:02:16.320,1:02:25.600
take drugs that have toxic side effects, even if 
that, because they might be able to slow down the

1:02:25.600,1:02:31.520
disease, and that's more important right now, 
because the disease progression is so deadly.

1:02:32.080,1:02:39.280
But I think the next thing is, because you 
would realise that what was really important

1:02:39.280,1:02:46.640
here was not just using a single drug. Just like 
with cancer, just like the connection to Jerome

1:02:46.640,1:02:53.200
Horowitz, who discovered AZT, who was also a 
cancer researcher, you would know that cancer

1:02:53.200,1:02:59.840
and HIV were rapidly able to evolve to mutate 
and develop resistance against any drug that

1:02:59.840,1:03:07.040
you developed. So the aim would not be to develop 
a single drug, but to use a combination of drugs,

1:03:07.040,1:03:13.760
and that was the goal that these researchers had 
even in the 1980s, even though they developed

1:03:13.760,1:03:20.480
AZT — it did work, but people eventually 
started to develop resistance against it,

1:03:20.480,1:03:24.468
but that was okay from their perspective because 
they knew that this was not the end goal.

1:03:24.468,1:03:25.426
This is the beginning.

1:03:25.426,1:03:29.040
It was not to develop one 
drug, we had to develop many.

1:03:29.040,1:03:33.920
We have to develop many. Not just 
because they're going to get used

1:03:33.920,1:03:37.840
one by one and then become resistant, but 
to use in combination from the beginning.

1:03:37.840,1:03:42.240
So they were thinking that way 
from the beginning. Yeah, okay.

1:03:42.240,1:03:50.080
Yeah, I mean this was so interesting to me 
just as a 'how to develop drugs', what is the

1:03:50.080,1:03:54.800
mindset that's required? What is the type of 
approach you use? Of just trying everything,

1:03:54.800,1:03:59.600
essentially having different hypotheses, just 
seeing what works. I thought it was really

1:03:59.600,1:04:07.200
interesting to read about. This, I think, then 
spurred a lot of other pharmaceutical firms and

1:04:07.200,1:04:12.400
researchers to work in the area to develop 
other types of drugs. Samuel Broder's team

1:04:12.400,1:04:20.720
then developed a bunch of other similar drugs. 
We just talked about AZT, which is a type of

1:04:20.720,1:04:29.440
nucleoside reverse transcriptase inhibitor, NRTI, 
and as we said, it's a drug that is this "mimic",

1:04:29.440,1:04:36.160
or this fake, version of a nucleoside 
base of the DNA molecule of HIV.

1:04:36.160,1:04:41.280
Let's pause there, and let me see if I 
can remember everything I just learned.

1:04:41.280,1:04:48.320
So I am putting myself in the headspace of a drug 
developer who doesn't have the tools of 2025,

1:04:48.320,1:04:55.840
when we're recording today. And there's quite 
a few tools I don't have. I don't have PCR.

1:04:55.840,1:05:00.320
I don't have, and don't have modern 
genomics. I probably don't have high

1:05:00.320,1:05:05.920
throughput screening. I definitely don't have 
knowledge of what HIV looks like, in terms of,

1:05:07.760,1:05:14.720
visually as a 3D structure — that's probably 
far away. And I don't really know the whole

1:05:14.720,1:05:18.880
process of the lifecycle of the virus. 
But what I do know is that, probably,

1:05:18.880,1:05:24.160
CD4 T-cells are implicated, because I'm seeing 
these counts drop. I've taken blood samples,

1:05:24.160,1:05:29.760
and those counts are not looking so good for 
patients. And what I do know is that, if I

1:05:29.760,1:05:34.480
rummage around, there are going to be some failed 
cancer drugs somewhere that I can at least try.

1:05:34.480,1:05:40.880
And so, sure enough, and because well, 
in addition, if I'm a cancer researcher,

1:05:40.880,1:05:48.080
I think about resistance and I think about 
combination drugs. So what I'm going to do is,

1:05:48.080,1:05:55.520
I'm going to go around and try a bunch of stuff. 
I mean, my takeaway from this is, it was just

1:05:55.520,1:06:03.760
incredibly empirical. You didn't have much in 
the way of theory, beyond the CD4 implication,

1:06:04.880,1:06:12.640
link, and you would try and stuff, and a bunch of 
stuff probably did not work, and then guess what?

1:06:12.640,1:06:21.200
One thing did work, and that gave everyone 
some hope, and changed things going forward.

1:06:21.200,1:06:27.120
Yeah, I mean, it's so amazing to read about 
drug development during that time and what

1:06:27.120,1:06:32.240
happened after that, as well. So maybe 
this is the time to actually talk about

1:06:32.240,1:06:41.030
the HIV life cycle and what the other types of 
drugs that have been developed are. And so...

1:06:41.030,1:06:41.040
I'm ready.

1:06:41.040,1:06:46.160
We should start with how an infection 
happens, at a molecular level.

1:06:46.160,1:06:46.560
Great.

1:06:46.560,1:06:52.320
So we have the HIV virus particle. 
I don't know if people have seen an

1:06:52.320,1:06:58.960
image or diagram or something of 
HIV, but essentially it has this-

1:06:58.960,1:07:04.080
I'm holding up my hands for people 
watching the video. Does this look right?

1:07:04.080,1:07:11.840
That looks right. So this is a spherical particle, 
it has a bunch of proteins coming out of it,

1:07:11.840,1:07:22.320
and inside the spherical particle is a bunch 
of stuff including a capsid. This capsid is

1:07:22.320,1:07:30.073
the core of the HIV virus. You can think of 
it as looking — oh wow! Is that an actual-?

1:07:30.073,1:07:35.440
I just picked up something to mislead people. 
It's a sun-bleached version of a vaccine,

1:07:35.440,1:07:39.920
but that mimics a viruses structure, so 
that you can present to the immune system.

1:07:39.920,1:07:44.240
This is a COVID vaccine that the 
Institute for Protein Design made.

1:07:44.240,1:07:45.840
Wait, can you hold it up to the camera?

1:07:45.840,1:07:52.720
Yeah, oh yes, I'm looking at it for myself. Anyone 
watching the video here: this looks like a virus,

1:07:52.720,1:08:01.040
it is not a virus. It presents the receptor 
binding domain of the spike protein of COVID,

1:08:01.040,1:08:06.480
or SARS-CoV-2, to the immune system on lots of 
different places so you can get antibodies that

1:08:06.480,1:08:11.920
bind. It doesn't look that far off. I mean, 
it's better than when I held my hands up, so.

1:08:11.920,1:08:13.440
It looks kind of cute, also.

1:08:13.440,1:08:15.040
It is cute. It is cute.

1:08:15.040,1:08:23.760
I had a stuffed toy version of the 
coronavirus that I got from this museum,

1:08:23.760,1:08:27.600
and I thought it would be really 
funny to get this as a gift,

1:08:27.600,1:08:32.160
and then to give it to someone 
and say, Ha! I've given you COVID.

1:08:32.160,1:08:38.240
You know, it's crucial to get 
good bits in, so I support.

1:08:38.240,1:08:48.480
So the HIV virus, you showed this spherical 
particle. We have this envelope that is a sphere,

1:08:48.480,1:08:54.960
and then it has some protein sticking out of 
it. Inside it, it has a capsid. The capsid sort

1:08:54.960,1:09:02.720
of looks like a thimble, or maybe more like 
a bullet. This kind of interesting because

1:09:02.720,1:09:08.720
the bullet — or the capsid — contains a bunch 
of the really important stuff, for the virus.

1:09:08.720,1:09:15.280
It contains the RNA molecules that's its genetic 
code. It also contains a bunch of other enzymes

1:09:15.280,1:09:20.240
that it needs to do important stuff, including 
reverse transcriptase, which it needs to turn its

1:09:20.240,1:09:29.440
RNA into DNA, and a bunch of other enzymes that 
we'll come to. We have this HIV virus particle,

1:09:29.440,1:09:34.160
this spherical thing with the protein sticking 
out of it. One of those proteins is called

1:09:34.160,1:09:44.880
GP-120. That protein — when the virus gets into 
our body, it targets our white blood cells,

1:09:44.880,1:09:56.880
our T-cells primarily, and this GP-120 protein 
attaches to a CD4 receptor on our T-cells.

1:09:56.880,1:09:57.440
Got it.

1:09:57.440,1:09:59.200
They get attached.

1:09:59.200,1:10:05.840
So the virus is currently outside of the 
cell, and it attaches to CD4 on the outside.

1:10:05.840,1:10:10.400
Then, it starts to also attach to another protein,

1:10:11.120,1:10:23.920
CCR5 or CXCR5. There's a — initially, it starts 
by infecting CCR5 T-cells. It uses these two

1:10:23.920,1:10:32.720
receptors, it binds to these two receptors, 
and then it injects itself into our cells.

1:10:32.720,1:10:38.720
So it binds to two proteins on the outside of 
the cell; it uses that as a way to get inside.

1:10:38.720,1:10:46.560
So it fuses with our cell membranes. 
Inserts the contents of this HIV

1:10:46.560,1:10:53.200
virus into our cells. That includes the 
capsid, the bullet, the bullet-like thing.

1:10:53.200,1:10:53.440
Bullet

1:10:53.440,1:10:58.800
But actually, I think a really good analogy, 
maybe, is like a rocket. You know how, when a

1:10:58.800,1:11:07.680
rocket launches, most of it falls off, but there's 
this core part of the rocket that continues going.

1:11:07.680,1:11:14.160
Right, and usually that's something I like 
because it contains astronauts. In this case-

1:11:14.160,1:11:14.960
It's not.

1:11:14.960,1:11:22.080
I don't like it. I don't like it. So instead 
of space, we're now in the cytoplasm.

1:11:22.080,1:11:28.320
That's right. So we're now in the cytoplasm 
— the inside of the cell. At this point,

1:11:28.320,1:11:36.800
the capsid then makes its way to our cell's 
nucleus. This was really interesting,

1:11:36.800,1:11:41.920
because we found out — so I had watched this 
video to try to understand what was going on,

1:11:41.920,1:11:46.560
what was the pathway? I feel like videos 
kind of help me remember things better,

1:11:46.560,1:11:53.920
and this video was from 2010, I think. Then I 
started reading about this process separately,

1:11:53.920,1:11:58.160
in research papers, and they described 
it differently. And it turned out that

1:11:58.160,1:12:05.760
our understanding of this life lifecycle has 
actually changed in the last five years, right?

1:12:05.760,1:12:15.280
Yes. I actually talked to a friend who did her 
PhD, who I think graduated in 2018, and did

1:12:15.280,1:12:24.960
her PhD on the HIV capsid. And she was saying to 
me, oh, back when we were doing it back in 2018,

1:12:24.960,1:12:34.080
all those centuries ago, we actually didn't 
yet know that the capsid, at least sometimes,

1:12:34.080,1:12:40.960
makes it all the way intact into the nucleus! 
I was like, what? But that's so basic,

1:12:40.960,1:12:48.640
that's the whole game. It turns out no, even 
now, we are getting tools that are making it

1:12:48.640,1:12:54.320
easier to actually see what the heck is going 
on inside these incredibly busy cells. And yeah,

1:12:54.320,1:12:59.520
it's makes you wonder, if you do your PhD in five 
years? What the heck are we going to know now?

1:12:59.520,1:13:06.000
I mean, I think one of the reasons for this is 
that, it's really hard to observe an infection

1:13:06.000,1:13:12.560
happening. It's obviously very harmful and 
probably very unethical to infect someone

1:13:13.120,1:13:19.840
directly with HIV, if you wanted to study what 
happens in this early part. The people with HIV,

1:13:19.840,1:13:24.240
that have been part of research, have 
obviously been much further along than

1:13:24.240,1:13:29.120
just being infected. So it's hard to 
actually study those earliest stages,

1:13:29.120,1:13:33.680
and that's especially true because 
HIV doesn't infect other animals.

1:13:33.680,1:13:40.560
The closest that we could use is SIV — simian 
immunodeficiency virus — which is slightly

1:13:40.560,1:13:46.960
different. That means that there are various 
things about this early stage that, I think,

1:13:46.960,1:13:54.480
weren't very clear. And I think that the change 
in the last few years was better microscopy.

1:13:54.480,1:14:02.560
That is what my friend was basically 
saying. She was saying, before we had

1:14:03.120,1:14:09.200
cryo-electron microscopy, we just couldn't 
visualise things as well. She was using other

1:14:09.200,1:14:15.280
techniques to do her best, and now you have 
this atomic resolution of these systems that

1:14:15.280,1:14:21.520
we haven't ever seen. I mean, it's beautiful. You 
can actually see what's happening. We never knew.

1:14:21.520,1:14:22.320
Yeah,

1:14:22.320,1:14:28.640
It's genuinely crazy. I mean, like, I've 
been reading about vaccine development;

1:14:28.640,1:14:33.840
biology over the 20th century, and what is 
really surprising to me is that — we didn't have

1:14:33.840,1:14:41.680
any way to visualize viruses until the 1930s.
Before that- so we have the smallpox vaccine,

1:14:41.680,1:14:46.080
which is against a virus, but this is 
before anyone knows what viruses are;

1:14:46.080,1:14:52.320
that's before germ theory was developed. We just 
happened to get quite lucky with observation

1:14:52.320,1:15:00.080
and testing. But it's only in the 1930s that 
we actually got this ability, this type of new

1:15:00.080,1:15:06.080
microscopy technique called "electron microscopy" 
that allowed us to see things at the resolution

1:15:06.080,1:15:13.920
that would let us see viruses that are much 
smaller than other bacteria, parasites and so on.

1:15:13.920,1:15:17.440
Okay, we're getting too excited, and we 
need to focus on the lifecycle. I can tell,

1:15:17.440,1:15:20.000
I can tell, I can tell. Because 
right now, I'm a capsid and I'm

1:15:20.000,1:15:22.720
about to enter the nucleus. I'm in the 
nucleus. What's happening after that?

1:15:22.720,1:15:30.240
Okay, so let's recap. So the virus attaches 
to the cell with GP120, it attaches to CD4

1:15:30.240,1:15:38.160
and CCR5. It inserts itself, fuses with our 
cell membrane, inserts its content into our

1:15:38.160,1:15:44.080
cells. That includes the rocket, or the 
rocket core, or the capsid, or the bullet,

1:15:44.080,1:15:53.280
whatever. That capsid makes its way to our 
cell's nucleus. It then actually gets inside

1:15:53.280,1:16:02.400
the nucleus. So our nucleus has these entry 
points, which are called pores, and the capsid

1:16:02.400,1:16:09.200
kind of snuggles through, kind of wiggles through 
those. And this is going to be important later on,

1:16:09.200,1:16:19.040
the wiggling. So it gets into the nucleus and then 
it starts reverse transcription. So at this point-

1:16:19.040,1:16:23.920
Question! Question, question. Does it do 
another? So I came as a full package and

1:16:23.920,1:16:29.120
then I unfilled myself for the inner package. Do 
I unfill myself again? So the capsid, kind of,

1:16:29.120,1:16:32.160
lets all the inner contents out into the nucleus?

1:16:32.160,1:16:37.520
I think so. And I am afraid of 
saying anything too definitively,

1:16:37.520,1:16:42.640
because I'm thinking, what if this 
knowledge changes in a few years or

1:16:42.640,1:16:47.200
something like that? But I think the 
capsid also dissolves at this point.

1:16:47.200,1:16:47.840
Okay.

1:16:47.840,1:16:56.480
Within the capsid we said that there was the RNA 
— the two RNA molecules — and there's the enzymes,

1:16:56.480,1:17:01.760
including reverse transcriptase. 
Reverse transcriptase turns the

1:17:01.760,1:17:09.680
RNA molecule into DNA. Then, now 
it has DNA, we also have DNA!

1:17:09.680,1:17:17.440
It can then insert itself into our cell's 
DNA, using an enzyme called integrase,

1:17:17.440,1:17:27.600
which integrates it. Makes sense. So the virus 
is now integrated itself into our cell's DNA.

1:17:27.600,1:17:36.400
Now, at some point, our cells will decide 
to turn our DNA into other RNA molecules,

1:17:36.400,1:17:41.200
and to proteins, because we need parts of our 
genetic code to do stuff at different times.

1:17:41.200,1:17:43.920
I don't want to brag, but 
I'm doing that all the time.

1:17:43.920,1:17:47.840
We're doing it all the time. I don't 
know the maths on this, but I know

1:17:47.840,1:17:55.680
there's a lot of it going on at any given time.
So it basically uses our own cells' machinery to

1:17:55.680,1:18:05.600
turn its DNA now into its RNA particles, and also 
to transcribe the other proteins and enzymes that

1:18:05.600,1:18:13.440
it needs for its functions. These proteins and 
enzymes and the RNA molecule somehow make their

1:18:13.440,1:18:23.360
way to the surface of our cells. They then bud 
out of the cell. The cell membrane of our cell,

1:18:23.360,1:18:31.840
which previously was fused with the previous virus 
particles, they bud into this new little particle,

1:18:32.400,1:18:38.960
and there's now an immature 
virus, a new HIV virus particle.

1:18:38.960,1:18:45.360
But there's lots of them, because it's not just 
that our body has transcribed one of these. It's

1:18:45.360,1:18:54.320
transcribing loads of these proteins and enzymes 
at a time. So this one HIV that has infected our

1:18:54.320,1:19:00.960
cells and integrated into our DNA can then 
multiply into many, many more that, and but out of

1:19:00.960,1:19:08.640
the cell. But at this point, it's still immature. 
It's still not able to cause an infection,

1:19:08.640,1:19:17.600
because it hasn't- the proteins that we've made 
for the virus are actually in this big compound,

1:19:17.600,1:19:25.920
of what is called a "polyprotein". So it's 
multiple proteins that are fused together.

1:19:25.920,1:19:27.040
At this point,

1:19:27.040,1:19:33.600
So it was maybe more efficient for 
the virus to do 'em all at once,

1:19:33.600,1:19:37.440
but they're now in a big string, so you 
can't actually- they aren't going to

1:19:37.440,1:19:41.040
perform their function. There are many different 
proteins that it wants from you. For example,

1:19:41.040,1:19:45.040
the integrase, maybe it's still 
part of that? Is that right?

1:19:45.040,1:19:48.640
They're all part of- Well, no, I 
think there are multiple polyproteins,

1:19:48.640,1:19:53.040
but one really big one has 
reverse transcriptase, integrase,

1:19:53.040,1:19:58.800
and a bunch of other important proteins. And 
they're all kind of in this huge polyprotein,

1:19:58.800,1:20:05.200
and then there's a separate enzyme 
that's produced called "protease" in HIV.

1:20:05.200,1:20:09.520
That's what I would do, if I had something that- 
I think I know what's going to happen. Keep going.

1:20:09.520,1:20:16.080
Okay. So this protease is what is 
commonly called a "molecular scissor".

1:20:16.080,1:20:16.240
Right!

1:20:16.240,1:20:24.080
It cuts this giant polyprotein into its components 
— into the individual enzymes that it then needs.

1:20:24.080,1:20:33.680
And that also creates, that also cuts off 
capsid proteins, which then form a new capsid.

1:20:33.680,1:20:37.600
Oh wow. Sorry, we're inside of the 
envelope, right now, of the virus?

1:20:37.600,1:20:38.708
We are now inside of the envelope.

1:20:38.708,1:20:43.680
But there's just a lot of mess in the immature 
state. We don't have a capsid yet. Okay.

1:20:43.680,1:20:51.280
So we now have- the protease has cut this 
giant protein, it has cut them into lots of

1:20:51.280,1:20:57.840
capsid proteins. The capsid proteins 
start to assemble into a new capsid.

1:20:57.840,1:21:01.440
Which, in itself, I mean, that's so cool. 
Because there are loads of different proteins,

1:21:01.440,1:21:07.760
right? That's going to be lots of different 
individual units that are due to, you know,

1:21:07.760,1:21:11.200
thermodynamics, I guess, sort of 
gravitating towards this configuration,

1:21:11.200,1:21:14.560
that is a bullet or that. Yeah, it's wild.

1:21:14.560,1:21:22.320
I mean, it's crazy. So this protease is what I 
wanted to talk about a little more. And I think it

1:21:22.907,1:21:28.720
is probably really useful to hear about this whole 
lifecycle in order to know what protease even is.

1:21:28.720,1:21:35.760
And so the HIV's protease, which is cutting 
up this giant protein into its components,

1:21:35.760,1:21:43.920
that is what I think is the next big advance in 
HIV drug development. There are a bunch of other

1:21:44.480,1:21:49.520
nucleoside analogues, the mimics like 
AZT, that are developed around this time.

1:21:49.520,1:21:58.560
A bunch of other drugs are developed, but I 
think the proteases are the next big step.

1:21:58.560,1:22:05.600
Alright. So we talked about protease, and 
the reason that's important is because the

1:22:05.600,1:22:14.000
next big advance, in my view, is drugs that 
targeted HIV's protease enzyme. I think I'll

1:22:14.000,1:22:19.440
talk a little bit about the first one that 
was developed. This was called saquinavir,

1:22:19.440,1:22:30.160
and it was developed by scientists at Roche. 
They were trying to figure out if there were any

1:22:30.160,1:22:35.440
drugs that could target this protein. So they 
knew that this was probably important, because

1:22:35.440,1:22:44.800
they could see that- I think they were able to 
test whether it was present in people with HIV.

1:22:44.800,1:22:52.720
And they knew that it was important in the process 
of breaking down that giant protein polypeptide

1:22:52.720,1:23:01.200
into the smaller components. And so they started 
to study its structure and its cutting pattern.

1:23:01.200,1:23:08.240
As we said, protease is often called a molecular 
scissor. It doesn't literally look like a scissor,

1:23:08.240,1:23:18.640
I assume it just looks like a blob or something. 
But when it's trying to cut down this giant

1:23:18.640,1:23:25.920
polyprotein, it slightly changes shape, I guess, 
it opens up. It gets into this transition stage,

1:23:26.720,1:23:35.120
attaches to the polyprotein and then snips it into 
separate proteins. And what they were trying to do

1:23:35.120,1:23:42.560
was, they were trying to find something that could 
jam that transition state, so it couldn't actually

1:23:42.560,1:23:53.520
cut the protein. You have to look at what this 
transition state might be, what specific part of

1:23:53.520,1:24:03.600
the protease enzyme is doing that snipping, and 
then, can we fit something into this little gap?

1:24:03.600,1:24:09.760
Okay, so we've got this protease 
— scissors — and we got my long

1:24:09.760,1:24:16.320
string of proteins — paper — and 
we're going to jam a rock in there,

1:24:16.320,1:24:19.840
and rock beats scissors. And that's 
one thing I learned many years ago.

1:24:19.840,1:24:21.040
Very good.

1:24:21.040,1:24:23.040
Thank you.

1:24:23.040,1:24:32.000
So what was interesting, I think, to them. I 
think, at this point, there is PCR testing,

1:24:32.000,1:24:39.280
because this is 1986. So now, PCR is available, 
but they could also try to look at where exactly

1:24:39.280,1:24:45.200
the protease was typically cutting. And 
they found that if you looked at protease

1:24:46.000,1:24:53.920
in other proteins in the lab, it was 
cutting at specific sequences in a protein.

1:24:53.920,1:24:59.360
A protein is made of many amino acids 
joined together, and it was typically

1:24:59.360,1:25:04.880
cutting in places with a tyrosine, which is 
one type of amino acid, or a phenylalanine,

1:25:04.880,1:25:11.600
which is another — and either of those followed 
by a proline. So it was a sequence of either

1:25:11.600,1:25:17.520
tyrosine or phenylalanine followed by proline. 
And this combination, of cutting at this point,

1:25:17.520,1:25:23.040
is something that human enzymes almost never do, 
which is really useful, because it means that if

1:25:23.040,1:25:30.400
they're able to target this something that is 
cutting at this point, then they're hopefully

1:25:30.400,1:25:37.600
not going to be affecting any human enzymes that 
are important to us, for our other functions.

1:25:39.280,1:25:49.280
So what they then did was try to look for 
other molecules that could fit into this

1:25:49.280,1:25:58.080
transition state, where the enzyme is snipping 
the polyprotein. As part of designing the first

1:25:58.080,1:26:05.040
protease drug, they also had to develop tools to 
test how well their drugs were working against it.

1:26:05.040,1:26:11.760
So they developed a dye reaction test, to 
detect these proline-containing fragments.

1:26:13.120,1:26:19.920
They also worked on cloning and purifying the 
protease enzyme, using recombinant DNA methods,

1:26:19.920,1:26:27.520
which were also fairly recent. The first 
recombinant DNA that was produced was in 1972,

1:26:27.520,1:26:36.000
and the first time that was used for human enzymes 
was insulin in 1978 — so this was the first time

1:26:36.000,1:26:45.200
that we could produce insulin in bacteria, 
instead of extracting it from the pancreases?

1:26:45.200,1:26:46.160
Pigs, was it?

1:26:47.840,1:26:55.040
I think pigs. Previously to that, it was 
dogs, and it was also cows and other mammals,

1:26:55.040,1:27:00.560
which is horrible. But there was no other 
way to treat diabetes except to extract

1:27:00.560,1:27:05.520
insulin from various animals, until the 
1970s. So this was a huge development

1:27:05.520,1:27:11.840
that was also very useful for testing 
out potential drugs against protease.

1:27:11.840,1:27:17.520
I mean, yeah, we now take this for granted. 
In any lab you're in, or most labs you're in,

1:27:17.520,1:27:23.600
you'll have some way of growing up proteins you 
want to study in a biological system — probably

1:27:23.600,1:27:28.480
in a bacteria, but maybe in mammalian cells, maybe 
in yeast. And that means you can study proteins

1:27:28.480,1:27:34.320
all day long. But back then, this would've 
been recent. You could only do that from the

1:27:34.320,1:27:43.120
seventies as you said. So yeah, that's another 
nice intersecting biotech improvement there.

1:27:43.120,1:27:49.360
It's so interesting to me, just how much 
the technology kind of happens along side

1:27:49.360,1:27:55.360
and how much is dependent on other tools 
being available and what new things that

1:27:55.360,1:28:01.840
allows you to, allows scientists to, do.
Okay, so back to protease inhibitors.

1:28:01.840,1:28:08.320
We're trying to test any drugs that fit 
into these. The transition state — the

1:28:08.320,1:28:15.760
little wedge — where the scissors are cutting. 
They tried a bunch of different drugs. One type,

1:28:15.760,1:28:25.200
called hydroxyethylamines, worked especially 
well. When they found that that was working,

1:28:25.200,1:28:30.640
they started making adjustments to it, to 
see if they could improve on that result.

1:28:30.640,1:28:35.360
They changed the ends of the molecules, 
they tweaked the sizes of chemical rings,

1:28:35.360,1:28:42.320
they swapped side chains, and they found 
that having a larger, fused ring structure

1:28:42.320,1:28:48.400
made it much easier for this compound to latch 
onto the protease and block it from cutting.

1:28:48.400,1:28:55.440
I find this kind of stuff crazy, just because 
the tweaking is so important. We're talking

1:28:55.440,1:29:02.480
about very small molecules, well, small 
areas we're tweaking. We're talking about

1:29:02.480,1:29:07.680
not that many... atoms! We're talking about- 
you add up all the atoms at the cleavage site,

1:29:07.680,1:29:14.480
I dunno, not that many. So yeah, it's 
wild that chemistry gets so specific,

1:29:14.480,1:29:21.680
and that means that you can have these small 
changes that have enormous changes for patients.

1:29:21.680,1:29:25.920
I mean, I think it shows how much 
different fields of research come

1:29:25.920,1:29:32.320
together in developing new drugs. We have 
people who work in the clinic with patients,

1:29:32.320,1:29:37.200
and they might see something that seems to be 
having an effect, so they try it out. Then,

1:29:37.200,1:29:42.480
there's the people working on microscopy, 
who are really important, the chemists, the

1:29:42.480,1:29:47.760
pharmacologists, who are testing out toxicity and 
drug reactions and things like that. It's like,

1:29:47.760,1:29:54.080
everything comes together, and that is 
really important here. And so this drug

1:29:54.080,1:30:05.360
that they then developed after these dozens of 
adjustments was initially called Ro-31-8159.

1:30:05.360,1:30:09.120
Ro-31-8159. It rolls off the tongue!

1:30:09.120,1:30:16.320
That was later named saquinavir, which is the 
first protease inhibitor that was approved. As

1:30:16.320,1:30:23.280
I said, this drug, because it was targeting 
the enzyme, that was cutting in the specific

1:30:23.280,1:30:32.240
place that was not the case in human enzymes. 
It was extremely selective to HIV protease,

1:30:32.240,1:30:39.520
and barely affected human enzymes, even at very 
high concentrations. That meant that it was

1:30:39.520,1:30:45.760
much safer. But at the same time, what 
I found kind of interesting was that,

1:30:45.760,1:30:54.560
this drug, if you gave it to people, most of 
the drug was excreted very quickly. About 96%,

1:30:54.560,1:31:01.680
I think, was excreted in the urine. So having a 
high concentration of the drug, thankfully, didn't

1:31:01.680,1:31:06.720
have these side effects, because you'd have to 
compensate for the amount that just gets peed out.

1:31:08.560,1:31:17.680
I think this is also interesting because we're 
at this point now — this was approved in 1995,

1:31:17.680,1:31:25.280
saquinavir. And at this point, there are, I 
think, around a dozen different antiviral drugs,

1:31:25.280,1:31:30.640
which again is amazing, because 
just 10 years before that,

1:31:30.640,1:31:37.760
people thought no drugs would work against 
this disease, but now they have 10, or so.

1:31:37.760,1:31:39.200
Progress.

1:31:39.200,1:31:44.720
So, at this point, there are a bunch of drugs, 
but none of them really work in a long lasting

1:31:44.720,1:31:50.240
way. People develop HIV, it manages to 
evolve resistance against the drugs that

1:31:50.240,1:31:56.720
are being used after a few months. And this 
seems like it's just another one of those,

1:31:56.720,1:32:01.440
okay, we've got a new drug, but is it really 
going to make a difference in the long term?

1:32:01.440,1:32:08.960
And I think people were, in terms of people 
with HIV, they were probably quite pessimistic,

1:32:08.960,1:32:16.080
in some ways. They want new drugs to be developed 
for their condition, but how is this going to

1:32:16.080,1:32:23.760
make a difference after a few months? But this 
is actually where things change, because now

1:32:23.760,1:32:33.680
that we have a protease inhibitor and we have 
nucleoside — the fake nucleosides — like AZT,

1:32:33.680,1:32:40.000
and we have a few other drugs, we can now combine 
them, and give them as combination treatment.

1:32:40.000,1:32:44.400
Ding, ding, ding, here we go. 
I don't know if, literally,

1:32:44.400,1:32:50.080
the combination involved AZT or some 
of the other antibodies you mentioned,

1:32:50.080,1:32:56.800
but I did notice that the protease is operating 
right near the end of the viral lifecycle. AZT,

1:32:56.800,1:33:01.360
as you described, it's operating much earlier. 
These are really different parts of the lifecycle,

1:33:01.360,1:33:07.280
and it's quite unlikely intuitively the virus 
would, at the same time, mutate against both.

1:33:09.040,1:33:14.080
So I think this is something that the cancer 
researchers, or the virologists, working on

1:33:14.080,1:33:20.000
this would now have realised: that we're now 
working on different aspects of the virus's life

1:33:20.000,1:33:27.120
cycle. It's fairly unlikely that it's going to be 
able to resist all of these drugs coming at it,

1:33:27.120,1:33:35.200
in different parts. And I think this is when 
this combination therapy started to be used, and

1:33:35.200,1:33:44.880
it was being tested alongside these new protease 
inhibitors. This new type of combination therapy

1:33:44.880,1:33:52.000
is called HAART, which stands for "highly active 
antiretroviral therapy". It's a combination of,

1:33:52.000,1:33:59.920
typically, one nucleoside reverse transcriptase 
inhibitor, like AZT, another drug inhibitor that

1:33:59.920,1:34:06.720
directly inhibits reverse transcriptase — so it's 
not a fake base — and then, a protease inhibitor.

1:34:06.720,1:34:12.000
There were multiple protease inhibitors that 
were introduced around the same time in 1995,

1:34:12.000,1:34:17.040
like a bunch of different pharmaceutical firms 
essentially racing to get their to market.

1:34:17.040,1:34:25.120
This is a huge change in terms of how HIV 
treatment works, how effective it is, in the US.

1:34:25.120,1:34:32.560
I'm showing this chart that I worked on earlier, 
just to show what impacts it had. You can see this

1:34:32.560,1:34:42.560
massive rise in mortality rates from HIV and AIDS 
from the 1980s to the 1990s; rapidly grows as an

1:34:42.560,1:34:51.600
epidemic disease growing exponentially. In 1995, 
in December, highly active antiretroviral therapy,

1:34:51.600,1:34:59.520
the combination treatment, is introduced, 
and it's just this huge drop in death rates.

1:35:00.480,1:35:07.280
The way that people talk about it is, as if people 
are on death row and they're coming back to life

1:35:07.280,1:35:12.240
suddenly with this new combination therapy, 
because it's something that the virus is very

1:35:12.240,1:35:21.440
hard- not able to evolve resistance to. I wanted 
to bring this up because I was reading that book,

1:35:21.440,1:35:29.680
How to Survive a Plague by David France, and 
he talks about his own- he was a reporter at

1:35:29.680,1:35:38.080
the time, and he was at one of these scientific 
conferences on protease inhibitors learning about

1:35:38.080,1:35:44.400
the science, what new drugs were available.
He describes one of his experiences towards

1:35:44.400,1:35:52.400
the end of the book, and this is what he says. 
He says: "One of the scientists interrupted his

1:35:52.400,1:35:59.120
presentation abruptly and he said, 'Maybe you 
are not understanding what I am saying. This

1:35:59.120,1:36:04.320
is the biggest news ever in this epidemic. 
This stuff is actually clearing virus out

1:36:04.320,1:36:09.040
of people's bodies. People are getting better. 
We don't know for sure yet, but we think these

1:36:09.040,1:36:15.200
drugs — this whole class of drugs — might allow 
people to live a normal life. This is what we've

1:36:15.200,1:36:19.840
been working for all these years. They're not 
a cure. We don't know what they are, in effect,

1:36:19.840,1:36:25.600
but this is the first major piece of good news 
we've had in all these years. They're calling it

1:36:25.600,1:36:31.680
the Lazarus effect. People who were in hospitals 
on their last breath are getting up and going back

1:36:31.680,1:36:38.720
to work. We've never seen anything like it.'"
And that's just an incredible change, I think,

1:36:38.720,1:36:48.000
from how scary it must have been in the 1980s, 
seeing some drugs really promising. But eventually

1:36:48.000,1:36:54.320
they start to fail and then you get this 
combination of therapy that changes everything.

1:36:54.320,1:37:01.200
A complete change, and it breaks your heart 
to think of people who didn't make it to see

1:37:01.200,1:37:11.280
that change. Oh, what a graph. Okay, 
you're going to have make some nice,

1:37:11.280,1:37:16.000
less emotionally-intense 
graphs for me to calm down now.

1:37:16.720,1:37:23.120
There are a bunch of other drugs that I think we 
aren't going to talk about. But apart from the

1:37:23.120,1:37:29.040
protease inhibitors, the nucleoside analogues, 
and other reverse transcriptase inhibitors,

1:37:29.040,1:37:34.880
there are other drugs that targets how 
HIV enters the cell. I think there are

1:37:34.880,1:37:41.680
also some that targets the integrase enzyme 
— that allows its DNA to integrate into our

1:37:41.680,1:37:49.200
DNA — and then there are some others as 
well. But I think that kind of covers

1:37:49.200,1:37:56.480
much of the major story in the 1980s 
and '90s on HIV and drug development.

1:37:56.480,1:38:04.240
Okay, so taking a step back, we know what HIV 
looks like. We know roughly where it came from.

1:38:04.240,1:38:08.880
We now know a bit about the lifecycle, 
and we know about combination drugs that,

1:38:08.880,1:38:16.640
together, prevent HIV from taking 
over, and are less prone to resistance,

1:38:16.640,1:38:24.480
because there's many of them you're on at once. 
Okay, what about, what's happening with vaccines

1:38:24.480,1:38:30.400
at this point? And are people talking about 
curing HIV? I mean, these drugs control HIV.

1:38:31.040,1:38:39.200
I think this is interesting because we don't 
have a vaccine for HIV yet, right? It's been

1:38:39.200,1:38:47.280
more than 40 years since the first AIDS case 
was reported. We have loads of antivirals now,

1:38:47.840,1:38:53.360
working in different ways, but we don't have 
any vaccines. I think this would have been

1:38:53.360,1:38:59.280
really depressing if vaccines were the only 
things that were being worked on at the time.

1:38:59.280,1:39:04.880
But thankfully people were trying out random drug 
combinations, and that's why I think this first

1:39:04.880,1:39:10.800
step, of getting AZT, was so important. 
But my understanding is there aren't any

1:39:10.800,1:39:16.880
working vaccines that we know of yet... so 
I didn't actually read anything about them.

1:39:16.880,1:39:25.520
Well, I know a little smidgen. At 
Open Philanthropy where I work,

1:39:26.160,1:39:34.960
the team I work on supports a bunch of different 
biomedical research, and roughly a third of what

1:39:34.960,1:39:46.320
we fund in grants is, in some way or other, 
vaccinology or focused on vaccines. The area of

1:39:46.320,1:39:55.920
vaccines we've done may be the least in is HIV, 
and the reason for that is that it's very hard,

1:39:55.920,1:40:03.040
and now people have more knowledge of why 
it's hard. It also, happily, despite being

1:40:03.040,1:40:09.600
so underfunded at the beginning of the crisis in 
the US and elsewhere, now has attracted much more

1:40:09.600,1:40:14.560
attention and funding. So there's actually been 
tens of billions of dollars thrown at this problem

1:40:14.560,1:40:21.440
of 'How do you make an HIV vaccine?' The reason 
it's so difficult, I think, the clues to that

1:40:21.440,1:40:33.520
are located in what you've already said about the 
virus. Well, firstly, clue number one is that our

1:40:33.520,1:40:40.720
immune system doesn't control the virus naturally 
very well, once an infection is established.

1:40:42.400,1:40:49.120
Most people are not able to control an infection, 
once it's established, and that implies, okay,

1:40:49.120,1:40:56.160
well, what is a vaccine? A vaccine is trying 
to trigger your immune system to be prepared

1:40:56.160,1:41:03.280
for future invaders. If hardly anyone has a 
prepared successful immune response, what are

1:41:03.280,1:41:11.120
we even trying to mimic here? It's a tough problem 
statement, whereas something like COVID, plenty of

1:41:11.120,1:41:18.000
people do manage to control and their immune 
response is productive. You can pretty easily

1:41:18.000,1:41:22.400
see with COVID, especially at the beginning, well 
there's one protein on the outside of this virus,

1:41:22.400,1:41:27.360
the spike protein, that if you block it with 
antibodies, it is not getting into your cell.

1:41:27.360,1:41:35.120
So let's try and mimic that immune response.
And then another clue is about the rapid mutation.

1:41:35.120,1:41:43.680
So if your immune system is trying to prevent 
something that keeps changing, it's going to be

1:41:43.680,1:41:50.560
harder. And sure enough, if you create a vaccine 
which is less dynamic than your immune system,

1:41:50.560,1:42:02.080
and is only one thing, then you're not going to 
be clearing all of these different permutations of

1:42:02.080,1:42:08.880
the virus. These days, because the understanding 
of the immune system has progressed even outside

1:42:08.880,1:42:14.880
of HIV, over the last few decades, and because we 
have so much better tools, people are still going

1:42:14.880,1:42:22.800
at the problem, and have sort of ingenious and 
complicated ideas about how to make a HIV vaccine.

1:42:25.680,1:42:31.680
You may have heard the phrase "broadly 
neutralising antibodies" — that's all the rage

1:42:31.680,1:42:41.280
for what people are using to develop vaccines, 
and going after. But that's, importantly,

1:42:41.280,1:42:46.960
not what we're here to discuss today. I think you 
and I really are focused on medical impact and

1:42:46.960,1:42:54.880
this podcast is too, and it's so interesting 
that what we are here to discuss is a drug,

1:42:54.880,1:43:00.720
in the sense of, it's not prompting your immune 
system to respond in a certain way, like a vaccine

1:43:00.720,1:43:06.560
would. It's trying to avoid getting rejected 
by your immune system, and instead is trying to

1:43:06.560,1:43:12.160
just be a chemical that's hanging around, and the 
chemical's doing the work, not your immune system.

1:43:12.160,1:43:19.840
I was thinking about the broadly utilising 
antibodies. Just in case people are not aware,

1:43:21.680,1:43:28.960
I guess a fraction of people seem to be able 
to develop an immune response to a wider range

1:43:28.960,1:43:35.040
of HIV strains after it's diverged. 
So trying to find those antibodies,

1:43:35.040,1:43:42.480
that seem to be working against this broad range, 
that is what people are looking for. Right? The

1:43:42.480,1:43:49.600
other thing is there are some people who are 
still, we're working on combination vaccines,

1:43:49.600,1:43:58.320
I think. So vaccines that include multiple 
different components of the HIV virus. And this,

1:43:58.320,1:44:04.720
I know from direct experience, because 
I was once in an HIV vaccine trial.

1:44:04.720,1:44:08.560
Aha! So when was this?

1:44:08.560,1:44:11.280
This was in 2019.

1:44:11.280,1:44:11.780
2019.

1:44:12.640,1:44:16.560
You might be wondering, why am I getting 
an HIV vaccine? Why am I in this trial?

1:44:16.560,1:44:23.760
It was this phase one trial. So essentially, 
they're just testing the safety and some basic

1:44:24.800,1:44:30.160
reactions — immunological reactions 
— you have to a potential vaccine.

1:44:30.160,1:44:37.040
And I got contacted through Imperial where I 
think I was studying at the time or had been,

1:44:37.600,1:44:43.760
and I was like, I love science, 
I want to be part of this trial.

1:44:43.760,1:44:46.240
We need more Saloni's in this world.

1:44:46.240,1:44:52.080
Also I was thinking, well what if this 
candidate vaccine actually works? I'll

1:44:52.080,1:44:54.720
be immune to HIV. That would be so cool.

1:44:54.720,1:45:00.080
Well, oh yeah. So this was not controlled. It was 
a phase one, so you were definitely getting it?

1:45:00.080,1:45:06.240
Oh, I actually don't know. I mean, I could have 
been on placebo, but still you have a 50%-ish

1:45:06.240,1:45:14.320
chance, probably, get it for free getting an 
HIV vaccine for free if it works. It was a

1:45:14.320,1:45:22.320
really funny experience because... if you have 
met me or seen me in person, I'm quite small.

1:45:22.320,1:45:30.240
I thought that, well, I knew that this trial 
had this eligibility requirement that you had

1:45:30.240,1:45:37.280
to be within normal BMI. I am essentially 
on the cutoff of underweight and normal,

1:45:37.280,1:45:42.800
whenever I've checked. And that's just been true 
for years. So I was really worried that I would

1:45:42.800,1:45:47.680
just fall under the threshold, and I would 
not be allowed to participate in the trial.

1:45:47.680,1:45:53.600
I love that you are hustling to 
get into an HIV vaccine trial.

1:45:53.600,1:45:58.400
Like, eating more food to get in. Exactly.

1:45:58.400,1:46:01.760
Bulk season is on.

1:46:02.400,1:46:08.240
Okay. So I was trying, I was really hoping 
that I would get into this trial. I got to

1:46:08.240,1:46:16.160
the clinical trial site. They asked me a 
few questions, they asked for my consent,

1:46:16.160,1:46:22.000
et cetera. And then, they also wanted to measure 
me, to check that I met the requirements. So

1:46:22.000,1:46:29.120
they were measuring my weight and my height. They 
then put that into their computer and they said,

1:46:29.120,1:46:35.680
'Oh great, you've passed this threshold', and I 
saw this BMI value on their screen. I was like,

1:46:35.680,1:46:41.840
that's surprising. Great, but surprising. And 
then, I looked at the values that they had

1:46:41.840,1:46:49.760
entered, and it turned out that I was shorter 
than I thought I was... so my BMI was normal.

1:46:49.760,1:46:58.320
So it took medical development of HIV to 
get you to understand your height. I mean,

1:46:58.320,1:47:01.600
there's a lesson here, but I'm 
not quite sure what the lesson is.

1:47:01.600,1:47:05.280
Well, it's also, it's hard 
to measure your own height.

1:47:05.280,1:47:06.720
Great point.

1:47:06.720,1:47:14.720
I dunno. It was both exciting because I could 
now participate in this trial, but also there

1:47:14.720,1:47:22.640
was this sadness that I felt, realising that 
I was even shorter than I thought I was.

1:47:22.640,1:47:27.200
Oh God. Wait, so what happened? Are you protected?

1:47:27.200,1:47:35.600
I don't know. Well, they didn't unblind me from 
whether I was getting the vaccine or the placebo,

1:47:35.600,1:47:40.960
but I did go in; I think I was in for 
some eight sessions. They did a bunch of,

1:47:40.960,1:47:47.840
was it blood testing? They did some 
testing of uncomfortable parts of my body,

1:47:47.840,1:47:52.960
to see the effects of this vaccine. 
I don't think I had any side effects,

1:47:52.960,1:47:59.920
maybe a headache at some point, but that was all. 
It was pretty nice. It was a great experience.

1:47:59.920,1:48:01.280
Nice. Cool.

1:48:01.280,1:48:07.520
I mean, I would say I would recommend it, but 
really you should decide that for yourself.

1:48:07.520,1:48:18.560
Sounds good. I just recently was screening to 
sign up for a vaccine trial here in San Francisco,

1:48:19.120,1:48:23.760
and I did the 15 minute screen, and they 
signed me up to go in person. And then,

1:48:23.760,1:48:27.920
the day I was going to go in person, I had a 
meeting that clashed, and I haven't got around

1:48:27.920,1:48:35.920
to enrolling, so I'm feeling a lot of guilt. So 
now I have an extra incentive though. Maybe I'll

1:48:35.920,1:48:39.600
figure out I'm taller than I think I am. 
Maybe I'll figure out I'm shorter though.

1:48:42.160,1:48:48.320
Well, so the reason I brought this up was because 
it was a combination vaccine, but also it was a

1:48:48.320,1:48:56.160
funny story. But the vaccine that they were trying 
contained, I think, three different proteins of

1:48:56.160,1:49:05.440
the HIV. So I think it was one adenovirus that was 
modified to carry an HIV coat protein. There was

1:49:05.440,1:49:13.840
another that was a vaccinia virus, which is... 
is that the smallpox vaccine virus, I think?

1:49:13.840,1:49:15.760
Hmm, yeah, probably.

1:49:15.760,1:49:21.840
And then there was another, that was another 
coat protein. So they had tried out, I think,

1:49:21.840,1:49:24.720
one or two of these before in trials, and then

1:49:24.720,1:49:29.200
this was putting them together 
into this combination vaccine.

1:49:29.200,1:49:30.371
It's interesting-

1:49:30.371,1:49:31.520
And then I don't know How it worked out.

1:49:31.520,1:49:37.360
Yeah, it's interesting you mentioned a 
coat protein. It makes me think of the

1:49:37.360,1:49:41.040
design differences you are dealing with 
when you're trying to make therapeutics,

1:49:41.040,1:49:46.640
and when you're trying to make vaccines. 
With vaccines, stereotypically,

1:49:46.640,1:49:52.240
especially for antibody responses, you wanna 
look on what's on the outside of an invader,

1:49:52.240,1:49:57.760
what's sticking out that my antibodies can glue 
to, and maybe a coat protein is a good choice

1:49:57.760,1:50:02.960
because it might be sticking out? And you know 
what is not sticking out? Those strands of RNA,

1:50:02.960,1:50:08.400
that are not only inside the envelope, they're 
inside a capsid; your antibody is not getting in

1:50:08.400,1:50:15.200
there. However, a small molecule drug, which is 
a nice tiny little chemical, can diffuse to many

1:50:15.200,1:50:21.680
places very surreptitiously. So you really might 
be able to interfere with something that the virus

1:50:21.680,1:50:29.920
has tried to protect from your immune system, 
but has failed to protect from genius humans,

1:50:29.920,1:50:36.720
who are using good tools to make something that 
nature actually couldn't have really got to.

1:50:36.720,1:50:40.800
No, exactly. Yeah, that's a really good point.

1:50:42.240,1:50:49.760
Shall we talk a little bit about 
treatment for HIV and what that's like?

1:50:49.760,1:50:58.160
Sounds great. And maybe we should even skip 
to prevention! We've talked a bit about-

1:50:58.160,1:50:59.200
Let's do that.

1:50:59.200,1:51:05.200
Let's do it, because you've given us a good 
overview of how in the '90s, these new drugs

1:51:05.200,1:51:13.680
allowed patients who had HIV infections to have 
much longer life expectancy, and control their

1:51:13.680,1:51:19.440
infections. There's a lot more that we could say 
about the different improvements since then in

1:51:19.440,1:51:27.040
treatment. But the principle is somewhat similar, 
if you want to be on these combination therapies.

1:51:27.040,1:51:39.520
So let's skip to prevention because prevention 
has some overlapping path and some different path.

1:51:39.520,1:51:46.240
What were the first ways that you could try and 
prevent getting HIV, if you didn't have it yet?

1:51:46.240,1:51:53.760
Well, you could change the way you were having 
sex, the type of sex you were having. You could

1:51:53.760,1:52:01.440
have sex with fewer partners, and you could have 
sex with condoms, which provide a barrier. The

1:52:01.440,1:52:09.520
thing that really changed preventive strategies 
more recently though, was drug availability for

1:52:09.520,1:52:15.600
PrEP — pre-exposure prophylaxis. So that's 
different than post-exposure prophylaxis,

1:52:15.600,1:52:22.800
which is PEP. And the pre- means you're 
taking the drug before you have sex,

1:52:22.800,1:52:29.920
or before you get exposed in some other way. 
That means that, if any HIV particles enter

1:52:29.920,1:52:38.880
your system, the drug is going to help block 
an infection getting established. So PrEP as a

1:52:38.880,1:52:46.960
drug regimen first became available in 
the US at least, in 2012. So Truvada

1:52:46.960,1:52:55.040
is a combination of two drugs, tenofovir and 
emtricitabine. Do you know how to say that one?

1:52:55.040,1:52:55.600
No.

1:52:55.600,1:53:03.440
Okay. I mean, they put them 
in one daily oral pill. So,

1:53:03.440,1:53:08.720
more specifically, it's actually 
tenofovir disoproxil fumarate,

1:53:08.720,1:53:16.240
which I'm sure I'm also mispronouncing, or 
"TDF", in combination with emtricitabine. And

1:53:17.200,1:53:27.040
those were two separate drugs that had been 
approved for treatment of HIV in 2001 and 2003.

1:53:27.040,1:53:36.240
The combination of them was approved as a 
treatment called Truvada in 2004. Then by 2012,

1:53:36.240,1:53:43.280
the FDA approved Truvada as the first PrEP 
regimen, after a clinical trial showed that

1:53:43.280,1:53:47.674
it had high efficacy in preventing 
infection. And I, yeah, go ahead.

1:53:47.674,1:53:51.840
Yeah, it's so interesting that some of the 
same antivirals that are used in treatment

1:53:51.840,1:53:59.040
were also used in prevention. One thing that 
made me think about was, I was reading about

1:53:59.040,1:54:06.960
was azidothymidine — AZT — the first HIV drug, 
and I think there's a part of that story that

1:54:06.960,1:54:15.440
gave them a clue that antivirals could be used 
as prevention as well. That was that pregnant

1:54:15.440,1:54:24.800
women who had HIV who were taking AZT, were not 
passing it on to their babies at the same rate.

1:54:24.800,1:54:31.840
They started running this trial in the '90s, 
and in 1994, I think, the study was published.

1:54:31.840,1:54:39.120
There was this massive drop in the rates of 
transmission, from mother to child, of HIV.

1:54:39.120,1:54:45.440
And that is really interesting as well, because 
even though people were developing resistance

1:54:45.440,1:54:54.240
to different HIV drugs, if they were pregnant 
and taking it, the drug resistance was not as

1:54:54.240,1:54:59.360
much of a problem if they were taking it 
late enough, because you only need this

1:54:59.360,1:55:04.240
particular time span for it to be effective. 
It doesn't have to be effective for years.

1:55:04.960,1:55:06.400
Of course. I see.

1:55:07.280,1:55:09.600
But also I think that just gave people a hint

1:55:09.600,1:55:14.240
that this is something that 
could be used in prevention.

1:55:14.800,1:55:19.760
Yeah, that's such a neat real world 
proof-of-concept of what you can do

1:55:19.760,1:55:26.240
there. Truvada has been improved on since. 
So maybe I'll just go through a couple of

1:55:26.240,1:55:37.200
those improvements. The fundamental idea is 
similar for the main improvement drug. In 2019,

1:55:37.200,1:55:43.120
there was a new regimen called Descovy. 
And you might be wondering, is this from

1:55:43.120,1:55:50.320
a competitor who's trying to outdo Truvada? 
And it's from the same company, Gilead, who,

1:55:50.320,1:55:57.520
as a bit of a spoiler, developed Lenacapavir later 
in life. Descovy does have a longer patent though,

1:55:57.520,1:56:04.400
so it's a better variation for men, it's 
emtricitabine again, which I've probably

1:56:04.400,1:56:10.240
said three times in three different ways, and it's 
tenofovir again. It's the same dose actually of

1:56:10.240,1:56:17.360
emtricitabine. I think it's 200 milligrammes. The 
tenofovir is in a new form though. Instead of TDF,

1:56:17.360,1:56:27.600
it's TAF, which stands for tenofovir 
alafenamide. And both TDF and TAF are

1:56:27.600,1:56:34.240
"pro drugs". For tenofovir, that means that means 
your body is sort of doing some work once you

1:56:34.960,1:56:42.400
ingest them to enzymatically, convert them into 
tenofovir, and then into tenofovir diphosphate,

1:56:42.400,1:56:48.320
which is the active formula drug. My 
understanding of the difference is that,

1:56:48.320,1:56:56.000
for TDF, so the original one, that primarily 
happens in blood plasma and for TAF-

1:56:56.000,1:56:57.600
What happens? The, the change?

1:56:57.600,1:57:05.040
The conversion... into the active drug. And for 
TAF, that primarily happens in the immune cells.

1:57:05.040,1:57:08.960
You know, if you think about the difference there, 
well, getting the same thing out the other end,

1:57:08.960,1:57:12.800
why do I care? Well, if you're doing it in the 
blood, then your blood's circulating everywhere,

1:57:12.800,1:57:17.360
including your kidneys, and you can actually 
have more unwanted effects from that,

1:57:17.360,1:57:23.520
than if you're more secluded when you're 
making your active drugs. I think that's

1:57:23.520,1:57:27.120
why the safety profile of Descovy 
looks a little bit better. You have,

1:57:27.120,1:57:33.520
if you're on long-term daily use of the first 
one, it's got a pretty good safety profile,

1:57:33.520,1:57:43.280
but it can have negative effects on kidneys and 
bones — so bone density and kidney toxicity. And

1:57:43.280,1:57:50.320
so, here's where, if anyone's on the video, I'm 
going to do some show and tell. I don't know if,

1:57:50.320,1:57:55.680
are you the kind of person-? I keep all of my 
empty pill bottles into the future indefinitely?

1:57:55.680,1:57:57.440
Oh, I don't do that.

1:57:58.800,1:57:59.437
I do this.

1:57:59.437,1:58:00.640
So these are empty pill bottles.

1:58:00.640,1:58:05.760
These are empty pill bottles, which hopefully 
don't have private information on them. But

1:58:05.760,1:58:11.600
basically I think I do it because I have some 
vision of, I'm going to do some art project about

1:58:12.320,1:58:16.000
what it's like to be a modern human in the future, 
and you do that. But I think I actually stole that

1:58:16.000,1:58:19.920
from, I think I've seen an art project, which 
had loads of entry pill bottles. So I actually

1:58:19.920,1:58:24.240
don't have a plan for these pill bottles.
But basically, here's what you can learn. So

1:58:25.120,1:58:31.680
take one tablet by mouth every day. 
This is emtricitabine and tenofovir,

1:58:32.560,1:58:38.080
200 to 300 milligrams. Here's another 
thing you can learn. So this one is this

1:58:38.080,1:58:43.520
empty pill bottle says Laurus labs on it.
And you might be thinking Laurus labs,

1:58:43.520,1:58:50.240
that doesn't sound like Gilead true. And this 
other empty pill bottle says Amneal on it; doesn't

1:58:50.240,1:58:58.800
sound like Gilead either. The reason for that, is 
that the Truvada patent expired in 2020. So there

1:58:58.800,1:59:08.080
are now many generic drug manufacturers who make 
Truvada, which is why I am on Truvada, because

1:59:08.080,1:59:15.200
when I first asked to go on Descovy, my doctor at 
the time was like — I think that patent exposed

1:59:15.200,1:59:21.600
in 2031 — was like, uh, no, no, no, we're going 
to give you Truvada. And then I did the classic-

1:59:21.600,1:59:23.040
Because it's cheaper, or?

1:59:23.040,1:59:30.000
Yes, correct. So it's made by drug companies 
outside of the US, usually in middle income

1:59:30.000,1:59:34.960
countries or in lower income countries, a lot 
based in India, but other countries too. I

1:59:34.960,1:59:40.960
actually don't, I feel like one of these is an 
American company, but I might be wrong. And by

1:59:41.680,1:59:48.000
making these generic competitors, where they only 
have to prove to the FDA that it is similar enough

1:59:48.000,1:59:53.280
in terms of its pharmacodynamics and all that, 
they don't have to redo all the clinical trials.

1:59:54.080,2:00:00.800
They can sell for a cheaper price than Gilead 
might. I actually remember later on, the first

2:00:00.800,2:00:09.040
injectable PrEP came about, cabotegravir, and 
that is made by a different company called Viiv,

2:00:09.040,2:00:19.040
V-I-I-V. That, I believe, was approved in 
2021, if I'm remembering right. That is:

2:00:19.040,2:00:26.800
you only get one injection every two months.
So I had heard about it and asked a different

2:00:26.800,2:00:31.040
doctor I had at the time, could I, just 
asked about it, wasn't sure if I do it,

2:00:31.040,2:00:35.760
wasn't sure if I'd stick with what I 
was doing. And that was very expensive,

2:00:35.760,2:00:44.000
so that was a quick no. My insurance at that time 
was not enthused about that one. If memory serves,

2:00:44.000,2:00:51.520
it was roughly $4,000 a dose, and it's every 
two months, that's six doses a year. I think

2:00:51.520,2:00:58.640
it was roughly $20,000 a year. And the patient 
benefits sure enough was not, I shouldn't really

2:00:58.640,2:01:02.160
have been paying that much, because the 
drugs I was on were working perfectly fine.

2:01:02.160,2:01:08.320
I guess I'm interested about how 
PrEP works on a day-to-day basis.

2:01:08.320,2:01:14.560
Do you take it every day? Do you only 
take it sometimes? What's the pattern?

2:01:14.560,2:01:20.720
So it's the default is you take it every 
day; it's a daily oral pill, and that makes

2:01:20.720,2:01:27.520
sure that there's enough of the drug in your 
system that you are safe, whatever happens.

2:01:27.520,2:01:34.400
There is another regimen that men can take for 
Truvada, which is often referred to as 2-1-1,

2:01:34.400,2:01:44.800
so I do. Where instead of doing it every day, you 
take two doses, so two pills, the day of sexual

2:01:44.800,2:01:52.080
activity or some risk you're exposing yourself 
to, one pill the next day, one pill the day after,

2:01:52.080,2:01:58.000
and then, the rest of the time, you just don't 
take anything. That is easier for some people.

2:01:58.000,2:02:03.280
And that said, if I think about, if I do 
an informal poll in my head right now,

2:02:03.280,2:02:07.680
with my gay friends in San Francisco, I would 
say most probably do daily, just because it's

2:02:07.680,2:02:16.480
easy. Actually, recently, my doctor tried to move 
me onto Descovy, which is the second one, just

2:02:16.480,2:02:22.400
because it's better for kidneys and when I was 
a baby, I had some kidney issues. That I do not

2:02:22.400,2:02:29.920
believe doctors do recommend 2-1-1. And certainly 
my doctor and I asked him, well, I'm doing 2 1 1,

2:02:29.920,2:02:36.080
can I do that with Descovy? And he said, "Well, 
I'm not allowed to say yes to that." I said, "Oh,

2:02:36.080,2:02:40.400
what do you mean?" And he was like, "I don't 
think that they've studied it with Descovy."

2:02:40.400,2:02:46.320
And I was like, "Okay, so we're sort of going to 
move on from that. Are we? And you're going to

2:02:46.320,2:02:50.880
give me Descovy?" And he's like, "Yes, I'm going 
to give you Descovy." So I'm in a sort of grey

2:02:50.880,2:02:55.760
area on that, and anyone listening, don't treat 
me as a doctor and my recommendation on Descovy.

2:02:55.760,2:02:58.720
But so you've now moved to Descovy?

2:02:58.720,2:03:08.640
I'm in the process, which is... you know, I'm a 
fairly plugged in, good health-seeking-behaviour

2:03:08.640,2:03:14.620
type person in the statistics. And I still haven't 
got around to being on the best one. So I dunno,

2:03:16.240,2:03:21.520
if it's not top priority in a given month, 
I might not get around to changing to the

2:03:21.520,2:03:26.000
better drug. You know, I was talking to 
someone yesterday actually, because I said,

2:03:26.000,2:03:32.960
"I'm about to record this podcast, what are 
you on?" And he said, "Oh, I am on the daily

2:03:32.960,2:03:38.640
pills." Most people on the daily pills probably 
couldn't even tell you if they're on, they both

2:03:38.640,2:03:43.120
work really well, it doesn't matter that much.
But he said, "Oh, I was thinking of going onto

2:03:43.120,2:03:49.360
the injectable every two months. But then as I 
thought about it more, I have to travel for my

2:03:49.360,2:03:55.040
job. So I was worried, well, am I definitely 
going to be back in San Francisco at the time

2:03:55.040,2:04:00.640
I need to get the injection? Or is there going to 
be a two week delay where I'm somewhere else? And

2:04:00.640,2:04:06.080
then actually, I'm more at risk, and actually I 
think it's easier if I just do the daily pills."

2:04:06.080,2:04:13.360
And I think that actually gets to how these 
drugs can be. It can be complicated how they

2:04:13.360,2:04:18.800
interact with someone's life. And it's not just 
something you can read off a clinical trial, of

2:04:18.800,2:04:25.360
how useful they'll be. You have to think about — 
well, how is someone who needs this drug going to

2:04:25.360,2:04:35.920
use it in their real life? And what there might be 
counterintuitive kind of pros and cons of having a

2:04:35.920,2:04:42.800
big gap of two months between that sounds great 
on paper, but there you go, someone didn't want

2:04:42.800,2:04:47.520
it. They wanted to just do it every day. And I 
actually think about this with treatment as well.

2:04:48.960,2:04:55.600
I have a friend who started dating someone 
who was HIV-positive. He was HIV-negative,

2:04:55.600,2:05:02.560
he started to date someone HIV-positive. 
And now, with the great drugs that we have,

2:05:03.360,2:05:10.080
if you are positive and on treatment, you will 
be undetectable, you won't transmit the virus to

2:05:10.080,2:05:18.880
partners, which is incredible. But my friend, he 
knew that, sort of rationally, and he was still

2:05:18.880,2:05:25.520
anxious around sex. It was a scary topic, even 
if he could sort of rationally tell himself he

2:05:25.520,2:05:30.720
shouldn't be scared, and then that can be a tough 
thing for a partner to deal with. And one day,

2:05:30.720,2:05:41.360
his new boyfriend actually took his daily pills 
in front of my friend as a way to basically build

2:05:41.360,2:05:47.040
trust, and that was a kind of beautiful thing 
to do. I think, if I were in his position,

2:05:47.040,2:05:52.640
I probably would've been offended and annoyed, 
and he was so generous. And then sure enough,

2:05:52.640,2:05:58.880
my friend did build trust and fall in love and 
have, I think, they had a wonderful sex life.

2:05:58.880,2:06:01.920
But that is the kind of thing you wouldn't 
think about when designing a drug. It's like,

2:06:01.920,2:06:08.080
if you are on a treatment drug that was 
an injection, that was once every whatever

2:06:08.080,2:06:12.960
period. Well okay, that particular trust-building 
exercise would not be available to you. So oh,

2:06:12.960,2:06:22.240
it gets so complicated. And then the people most 
affected by HIV and AIDS these days — it affects

2:06:22.240,2:06:29.920
women in Africa more than it affects men now, 
especially in Southern Africa. And there the

2:06:29.920,2:06:37.280
complexities are very different and I won't be 
able to rightly summarise them. But for example,

2:06:37.280,2:06:43.760
if having to go to a clinic for a procedure or 
injection might be different than having to go

2:06:43.760,2:06:48.480
to a pharmacy to pick up pills, which might 
be different than having pills on your shelf

2:06:48.480,2:06:53.280
for many weeks, versus having to go more 
regularly and all of that matters a lot.

2:06:53.280,2:06:59.360
No, that's super interesting. I think one thing 
I was thinking about when you were talking about

2:06:59.360,2:07:09.760
this was, what are the different problems that 
people will face apart from- So we have this

2:07:09.760,2:07:13.840
sort of struggle to, I dunno, schedule some of 
these appointments, things like that. I think,

2:07:13.840,2:07:19.040
one thing on that front was, with some of 
these injectables, I think there's a kind

2:07:19.040,2:07:24.320
of leeway that you have for when you get the 
next dose. You don't have to get it exactly,

2:07:25.040,2:07:30.880
let's say, six months or exactly two months after, 
but there is a little grace period that you can

2:07:30.880,2:07:36.880
get it in. But that's still, probably sometimes, 
quite inconvenient for people if something happens

2:07:36.880,2:07:43.280
if they're in a different country or so.
But there's the issue of, I dunno, taking

2:07:43.280,2:07:50.960
a daily pill every day for many years, that might 
be hard for some people in terms of remembering,

2:07:50.960,2:07:56.560
especially if it's a preventive pill. It's not 
something that is necessarily super salient to

2:07:56.560,2:08:03.040
them as if it's a condition they already have. But 
also, it's this access, like, what if the clinic

2:08:03.040,2:08:09.600
is closed one day? What if something happens? 
What if, I don't know, someone takes your pills

2:08:09.600,2:08:16.400
or they drop out of your bag, or something like 
that. How are you going to get the next dose?

2:08:16.960,2:08:21.040
You know, I can answer that for me, 
and I pick 'em up once every month,

2:08:21.040,2:08:30.880
but I'm just one guy and I think we need 
to get a better answer for people who are

2:08:30.880,2:08:35.680
affected in other contexts. So I think 
it's time to phone a friend, Saloni.

2:08:35.680,2:08:37.920
We're going to phone a 
friend. This is so exciting.

2:08:37.920,2:08:38.800
We're going to phone a friend.

2:08:38.800,2:08:40.160
Who are we phoning?

2:08:40.160,2:08:43.200
We're going to phone my friend Douglas Chukwu,

2:08:43.200,2:08:50.480
who works at Open Philanthropy with me, and 
before that was a medical doctor in Nigeria,

2:08:50.480,2:08:58.240
and worked in public health on HIV treatment 
and prevention. So should we dial him up?

2:08:58.880,2:09:02.320
Hello Douglas. How are you 
doing? Thanks for joining.

2:09:02.320,2:09:04.640
Good, good. Great to be here.

2:09:04.640,2:09:09.280
Well, thanks for taking time out of 
your day. So, we worked together,

2:09:09.840,2:09:14.400
but before we worked together at Open 
Philanthropy, you were trained as a doctor

2:09:14.400,2:09:20.640
and worked on other things in public health, which 
is why we wanted to bring you on today. So yeah,

2:09:20.640,2:09:23.680
what's your background and what were you 
working on, before Open Philanthropy?

2:09:23.680,2:09:27.120
I trained as a medical doctor in Nigeria. 
So I had a couple of years of clinical

2:09:27.120,2:09:31.120
practice working as a medical officer in a 
government establishment and also a private

2:09:31.120,2:09:35.760
establishment. So had that dual experience 
and then piloted to work in public health.

2:09:35.760,2:09:39.680
Interestingly, most of my public health 
experience was in the field of HIV and AIDS.

2:09:39.680,2:09:46.080
Are people getting a weekly stock of treatment 
of PrEP, or will it last them months or

2:09:46.080,2:09:50.800
years? Like how long does- maybe this varies 
depending on the type of drug that they're using.

2:09:50.800,2:09:58.800
Yeah, so oral PrEP comes in, the commonest is 
the pack of 30 tablets, and oral PrEP is to be

2:09:58.800,2:10:05.040
administered daily. So the common, it varies 
from a range of one month to three months;

2:10:05.040,2:10:09.360
three months being the maximum, because for 
individuals on PrEP, they need to be tested every

2:10:09.360,2:10:15.280
three months, as per the national guidelines, 
so that's the touchpoint with the facility. The

2:10:15.280,2:10:19.760
treatment duration of prescription helps to 
make sure that they come for their refills,

2:10:19.760,2:10:25.280
they're assessed for adherence, they're tested for 
HIV and they're also monitored for side effects.

2:10:25.280,2:10:31.600
How important is public funding from 
donor countries like the US and the UK,

2:10:31.600,2:10:34.320
when it comes to HIV particularly?

2:10:34.320,2:10:37.280
Absolutely important, right? I'll give an example. 
There are various access- let's say for example,

2:10:37.280,2:10:41.600
there are a lot of individuals that have accessed 
treatment that wouldn't have accessed treatment

2:10:41.600,2:10:46.400
if the HIV programs in countries were 
entirely reliant on domestic funding.

2:10:46.400,2:10:51.520
And this varies across African countries, but in 
Nigeria, for example, over 80% of the funding for

2:10:51.520,2:10:59.440
HIV programs is via external funding. And then, 
there are some countries like South Africa,

2:10:59.440,2:11:04.160
where they've made some progress in terms 
of domestic financing for HIV programs,

2:11:04.160,2:11:13.760
I think as high as 70%. But overwhelmingly, in 
Africa, there are various country programs that

2:11:13.760,2:11:18.480
are hugely reliant on external funding 
for sustaining and delivering for HIV.

2:11:18.480,2:11:25.360
We are recording this at the beginning of 
April and I've still found it difficult to

2:11:25.360,2:11:35.680
get good reporting on quite what's happening with 
PEPFAR, the US- the main way that the US supports

2:11:35.680,2:11:42.480
HIV programming. And the answer of, what's 
happening may change in the coming months. But,

2:11:42.480,2:11:48.720
as you talk to your friends who work in 
public health, what are you hearing at

2:11:48.720,2:11:59.840
the moment? What has happened at facilities or on 
the ground, in reaction to the PEPFAR uncertainty?

2:12:00.560,2:12:04.880
The effects of this cut across not just the 
healthcare workers. But healthcare workers,

2:12:04.880,2:12:09.840
patient communities; there's a lot of uncertainty. 
There's a lot of unease and a lot of worry

2:12:09.840,2:12:16.720
about what the future holds. And a lot of these 
suspensions were abrupt. So people got stop work

2:12:16.720,2:12:20.880
orders. As I mentioned, there is a community 
component of healthcare service delivery.

2:12:20.880,2:12:25.040
There are community healthcare workers 
that are supported by the PEPFAR funding,

2:12:25.040,2:12:30.960
and having the stop-work orders meant people 
stopped getting tested in communities,

2:12:30.960,2:12:36.080
access to some medications were threatened, even 
though perhaps there were some stock to sustain

2:12:36.080,2:12:41.200
initial dispensing of ARVs (antiretrovirals). But 
clients were being told that if this continues,

2:12:41.200,2:12:45.840
you'd have to pay out of pocket for 
your medications and that's actually

2:12:46.560,2:12:51.760
troubling for the patient community. Additionally, 
I think about the broader implication of this,

2:12:51.760,2:12:58.880
which is knowing that the funding for HIV 
programs is actually threatened, that also

2:12:58.880,2:13:06.160
affects manufacturers thinking about maybe wanting 
to exit some markets. That kind of damages a lot

2:13:06.160,2:13:10.960
of the progress that has been made over the past 
couple of years in the field of HIV and AIDS.

2:13:10.960,2:13:17.600
There are country governments rallying up to 
cover some of those gaps. But those resources

2:13:17.600,2:13:25.360
pale in comparison to the amount of resources 
that the US government devotes to supporting this.

2:13:25.360,2:13:34.160
As a- if someone needed treatments, I guess 
in the last three months, or even now,

2:13:35.040,2:13:43.360
how would the cuts and the stop works order 
affect them? What would they be experiencing?

2:13:43.360,2:13:51.200
I would say to paint the picture, imagine a status 
quo where every day, community health care workers

2:13:51.200,2:13:54.720
report to the facility, gear up with their mobile 
testing kits, with their ARVs [antiretrovirals],

2:13:54.720,2:14:00.000
and they go out into hard to reach areas. They 
identify people who are positive for HIV, place

2:14:00.000,2:14:05.280
them on treatment. Some of these are pregnant 
mother who don't have the resources to come to the

2:14:05.280,2:14:11.840
facility. So that abrupt suspension means those 
individuals will not benefit from those services.

2:14:11.840,2:14:16.960
Now, beyond those who are yet to be identified, 
because that's the category I just talked about,

2:14:16.960,2:14:21.840
there are people who rely on these healthcare 
workers to reach them to receive their refills

2:14:21.840,2:14:27.360
for ARVs, right? So the suspension was abrupt, as 
you know. So people were just told to stop work,

2:14:27.360,2:14:32.720
and their clients who likely would be expecting 
their healthcare workers to deliver ARVs to them

2:14:32.720,2:14:41.840
and would have been affected by such stop work 
orders, so that's pretty much it. Because there

2:14:41.840,2:14:48.000
are those who still have drugs, but there are 
those who are actually suffering from these cuts.

2:14:48.000,2:14:52.880
My sort of understanding, I was 
reading a few articles about this

2:14:52.880,2:14:57.280
and the impression that I had was, the 
clinics might have some of the treatment,

2:14:57.280,2:15:05.120
but they're just shut and there's no one; the 
staff who are paid or supported by the US are

2:15:05.120,2:15:10.560
not allowed to go in, and people can't get 
treatment even if it's in the clinics there.

2:15:10.560,2:15:16.400
Absolutely, yes. And the staffing, as I 
mentioned, the staffing support, it's not limited

2:15:17.760,2:15:22.480
to the community setting. Even in healthcare 
facilities, there are one-stop shops that are

2:15:22.480,2:15:28.320
staffed by individuals that are supported by 
the performing, as you rightly pointed out,

2:15:28.320,2:15:31.840
that stop work affected those 
individuals and perhaps clients

2:15:31.840,2:15:38.640
would've presented to facilities and wouldn't 
have had, maybe, individuals to attend to them.

2:15:38.640,2:15:44.960
And I think we were talking about refills and how 
often people get refills, and if that's every 30

2:15:44.960,2:15:50.320
days or every three months or so, that probably 
adds up to quite a lot of people who have been

2:15:50.320,2:15:56.400
directly affected by these cuts over the last, 
almost, I guess, two and a half months, maybe.

2:15:56.400,2:16:00.000
Absolutely. Because people, I mean it's a 
three month cycle. It can be a six month

2:16:00.000,2:16:06.800
cycle for people who are stable, but every day 
marks someone's clinic appointment, right, so.

2:16:06.800,2:16:10.560
It's a scary time for a lot of people with HIV.

2:16:10.560,2:16:18.400
How excited are you for lenacapavir in the field? 
I think one reason, I think, it's going to be

2:16:18.400,2:16:23.840
quite important is because of this adherence 
issue that you mentioned, but also meaning

2:16:23.840,2:16:30.400
that people don't need to get refills as often, 
so there's a bit more stability for someone who

2:16:30.400,2:16:38.640
has had an injectable. Is that also your view? Are 
there other things that you see as part of this?

2:16:38.640,2:16:44.000
One of the challenges with oral PrEP is having to 
take it every day. And with suboptimal adherence,

2:16:44.000,2:16:50.160
there's the risk of resistance developing. So 
having a drug that's administered twice a year,

2:16:50.160,2:16:54.559
I mean, I won't say it's as good as a vaccine, 
but there are challenges with developing the

2:16:54.560,2:17:01.360
HIV vaccine. So this is as good as we are 
currently towards making sure that people

2:17:02.479,2:17:09.839
keep from getting infected with HIV. It's very 
exciting in the HIV prevention landscape, having

2:17:09.840,2:17:16.000
an injectable once a year, fingers crossed, but 
that would be amazing. That would be phenomenal.

2:17:16.000,2:17:18.160
We're almost there. We're almost there.

2:17:18.160,2:17:21.120
That was really helpful. Thank you so much

2:17:21.120,2:17:23.200
Thank you so much Douglas.

2:17:23.200,2:17:27.040
Thanks so much. Thanks so much. Happy to 
talk about this and very excited about

2:17:27.040,2:17:31.280
the development in the HIV prevention 
space. Hopefully these developments

2:17:31.280,2:17:37.104
continue and move the needle in terms of 
achieving epidemic control of HIV and AIDS.

2:17:37.104,2:17:41.359
That was great. This was so cool, to phone 
a friend and learn about what things are

2:17:41.359,2:17:47.040
like in treatment in Nigeria, the 
future of lenacapavir. But also,

2:17:47.040,2:17:55.439
all of the funding cuts and the disruption that's 
going on there right now. I think it really made

2:17:55.439,2:18:04.399
me think of how important some of these new drugs 
could be in terms of changing around the epidemic

2:18:05.120,2:18:14.000
in HIV, the possibility of using long-acting 
drugs, and by that we mean drugs that have an

2:18:14.000,2:18:20.720
effect for a really long period. Currently we 
have cabotegravir, which is a two monthly drug.

2:18:20.720,2:18:25.920
There's also lenacapavir, which is 
a six monthly drug. And potentially,

2:18:26.640,2:18:32.240
Gilead is also working on a once-yearly 
drug. And if that pans out, again,

2:18:32.240,2:18:38.479
I think it would completely change our ability 
to respond. Whether that is actually scaled up

2:18:38.479,2:18:46.639
is another question, and that's something 
we'll talk about later on. But it's really,

2:18:46.640,2:18:52.880
I think it's a change in what's 
possible in treating and preventing HIV.

2:18:52.880,2:18:59.439
Yes. And I feel like I have a lot to 
digest and we have a lot still to discuss.

2:18:59.439,2:19:10.639
So lemme go away for a second and think.
Okay, Saloni, we're back. How you doing?

2:19:10.640,2:19:16.560
I'm doing great. It's been five days. 
I've had a lot to think about. You look

2:19:16.560,2:19:19.600
like you're in a completely 
different place from before.

2:19:19.600,2:19:27.520
That's right. New shirt, new background. 
I am in New York City. I'm in the village,

2:19:27.520,2:19:34.720
the East Village — or at least I thought I was. 
I arrived and was informed by someone who lives

2:19:34.720,2:19:40.640
here that I'm actually in Stuy Town, which 
is not the same thing as the East Village,

2:19:40.640,2:19:47.359
but I've decided to kind of squint and it feels 
about the same and I'm having a good time. It

2:19:47.359,2:19:57.839
feels, I wish I could say I was here funded by our 
podcast to do some historical analysis of the AIDS

2:19:57.840,2:20:02.720
epidemic because I was in Castro before and now 
I'm in the Village. And those are two important

2:20:02.720,2:20:07.439
parts to the story. However, I am actually just 
here visiting a friend. Totally unrelatedly.

2:20:07.439,2:20:13.040
I really enjoyed thinking about 
what Douglas told us about how

2:20:13.040,2:20:19.280
treatment works in the field in a clinic 
in Nigeria. But also just thinking about

2:20:19.280,2:20:25.439
the different approaches that people 
have to prevention, whether that's

2:20:25.439,2:20:34.160
with condoms or behavioural changes or 
PrEP, this amazing breakthrough in 2012,

2:20:34.160,2:20:42.240
of multiple drugs in a combination that 
can reduce the chances of infection. And

2:20:42.240,2:20:48.719
it's really interesting first to think about the 
behavioural aspects that lead to, basically, how

2:20:48.720,2:20:54.960
do people actually take these drugs in practice 
and how does that inform drug development? How

2:20:54.960,2:21:02.399
does that inform the kinds of new treatments 
that we need and whether they're effective.

2:21:02.399,2:21:10.000
I think that's ultimately the key breakthrough 
of this new drug that we're going to talk about,

2:21:10.000,2:21:17.600
lenacapavir, that instead of being a daily pill 
that people would take as they do with PrEP,

2:21:17.600,2:21:24.479
it's this long-acting injectable. So it's an 
injection that you would take. So there's,

2:21:24.479,2:21:30.559
I guess, two injections every six 
months, and this massively reduces

2:21:30.560,2:21:38.640
the chances of infection. It's also been used as 
a treatment for people with drug-resistant HIV,

2:21:38.640,2:21:43.120
and there could be other purposes for it as well. 
So I think that's really the key breakthrough and

2:21:43.120,2:21:51.439
I think I really started to understand exactly 
how that would matter for someone who has HIV,

2:21:51.439,2:21:58.559
thinking about how do they get their next supply 
of the drug; how this makes a difference to them.

2:21:58.560,2:22:06.640
Yeah. It also got me thinking about the costs; 
how can we get the costs low for new drugs so

2:22:06.640,2:22:13.280
that they can get used more. And that's something 
that I'm interested to talk about with lenacapavir

2:22:13.280,2:22:19.359
too. But what is lenacapavir, Saloni? It's 
time for you to teach me something new.

2:22:20.399,2:22:27.120
I think to teach you about what lenacapavir is, 
we have to go back a little bit and talk about

2:22:27.120,2:22:35.840
the capsid. So if you remember from before, the 
capsid is this thimble-like structure within the

2:22:35.840,2:22:44.399
HIV virus that contains the RNA molecule, and it 
contains a bunch of other enzymes. It's the core

2:22:44.399,2:22:53.519
that stays intact when HIV enters cell, and this 
capsid takes that information, and those enzymes,

2:22:53.520,2:23:03.120
into the cell's nucleus. They then allow the 
RNA molecule — the genetic code of HIV — to

2:23:03.120,2:23:12.479
turn into DNA to integrate into our own cellular 
DNA, and then to proliferate into many more HIV

2:23:12.479,2:23:21.040
particles. So it's this key structure that's kept 
intact throughout this process. Once it gets to

2:23:21.040,2:23:29.840
the nucleus, it then starts to dissolve, letting 
the RNA molecule out, letting the enzymes out to

2:23:29.840,2:23:38.399
do their jobs. And then, later on, when the new 
HIV particles start to be formed, it then starts

2:23:38.399,2:23:44.639
to form; the capsid starts to form again. 
This process is actually really interesting;

2:23:45.760,2:23:51.439
and quite interesting, both as a process 
but also in terms of what it looks like.

2:23:51.439,2:23:51.600
Yes!

2:23:51.600,2:24:03.439
This HIV capsid is made up of, I think, more 
than 1,500 or so proteins. But each of these

2:24:03.439,2:24:09.679
proteins comes in groups. So some of the 
groups of this protein are in groups of six,

2:24:09.680,2:24:15.439
which are called hexamers. So hex- is 
six, and some of them are in pentamers;

2:24:15.439,2:24:26.160
penta- means five, but there are 250 hexamers or 
so, it kind of varies. And exactly 12 pentamers.

2:24:26.160,2:24:30.160
I'm looking at it now, and it looks 
like flower petals that are falling

2:24:30.160,2:24:35.760
into place. How on earth is it so exact? 
Why are there 12? Why are there 12? And

2:24:35.760,2:24:43.760
the 12 are not exactly... they're sort 
of dotted around in a pretty pattern,

2:24:43.760,2:24:49.359
but not necessarily how I would've 
designed it, if I were thermodynamic.

2:24:49.359,2:24:55.120
Yeah, I mean if you're listening, it's 
this thimble-shaped structure. Most of

2:24:55.120,2:25:02.399
that is made of hexamers of this protein. So, 
imagine some six-shaped thing, maybe like a

2:25:02.399,2:25:11.345
star-shaped cereal that I used to have called 
Honey Stars. Oh yeah. And they were very tasty.

2:25:11.345,2:25:13.040
Or Shreddies, which I'll 
never say a bad word against.

2:25:14.560,2:25:21.360
And then, in a few places in this capsid, 
there are other structures that are only five

2:25:22.080,2:25:28.800
that only have five proteins. So they're 
like five-star shapes. And these look,

2:25:28.800,2:25:37.359
I mean, where these are organised in the whole 
capsid doesn't look very symmetrical to me.

2:25:37.359,2:25:37.920
No.

2:25:38.880,2:25:46.240
And it's quite strange, but what I was reading 
was that: the placement or the number of these

2:25:46.240,2:25:52.960
pentamers within the whole structure changes 
the whole shape of it. It's sort of like,

2:25:52.960,2:25:58.080
this structure, that's where the hexamer 
are tesselating. So they're all fitting

2:25:58.080,2:26:04.000
together in this very symmetrical way 
between them, but then these five-shaped

2:26:04.000,2:26:11.040
pentamers determine the curve, I think, 
of the capsid. That's very interesting.

2:26:11.040,2:26:15.920
Yeah. I feel like I want... if I redo 
the tiles in my bathroom or something,

2:26:15.920,2:26:21.600
I want to do this. But now that you've 
told me the shape, the 3D structure,

2:26:21.600,2:26:27.359
is an important property, that wouldn't be so 
good on tiles. So I'm going to have to rethink.

2:26:27.359,2:26:30.559
You would have... Maybe it would 
be more like a flower vase?

2:26:30.560,2:26:34.880
Yeah. Oh, great idea actually. Yeah, 
it's very floral, it's just so,

2:26:34.880,2:26:41.920
it is very beautiful. I recommend people 
listening Google it. And I think recently,

2:26:41.920,2:26:45.040
it's only the last 10 years or something 
where we've really known what these

2:26:45.040,2:26:48.960
different polymers look like and how they 
assemble and all that. Is that right?

2:26:48.960,2:26:58.880
I think that's right. Only since the 2010s have 
we had advances in microscopy that allow us to see

2:26:58.880,2:27:03.040
some of these particles, with enough 
resolution to see what they actually

2:27:03.040,2:27:09.760
would look like in this coherent 
structure. That is super interesting.

2:27:09.760,2:27:15.359
As I think about the drugs you've described 
previously, what is interesting about this one:

2:27:15.359,2:27:19.920
if we're talking about the capsid and we're 
going to try and go after the capsid with a drug,

2:27:22.000,2:27:29.040
this is a structural part of the 
virus, but it's not as functional

2:27:29.040,2:27:33.680
in the direct sense. I'm not imagining 
something getting integrated into my DNA,

2:27:33.680,2:27:40.880
so I'm going after the integrase; and I'm not 
imagining I'm chopping up the large string of

2:27:40.880,2:27:46.640
proteins into smaller proteins, and going after 
the protease. This is more like just a package,

2:27:46.640,2:27:51.680
a thimble, a bullet; that it must be 
required, because why would it still

2:27:51.680,2:27:58.560
be there? But the function is not as direct. 
I don't know. Does that ring through to you?

2:27:58.560,2:28:04.160
I think you're kind of right, but I think 
it is important. Because to make sure all

2:28:04.160,2:28:09.040
of this stuff doesn't fall out, I guess, 
in some other part of the cell. Like,

2:28:09.040,2:28:17.120
we need to carry or transport the RNA molecule 
and the other enzymes to the nucleus. But it

2:28:17.120,2:28:24.960
also has these functions where, based on 
the structure, based on the shape of this

2:28:24.960,2:28:32.319
capsid — that allows it to enter the nuclear 
pores and the nucleus, the little holes and

2:28:32.319,2:28:39.759
this shape allows it to wiggle through. And I 
think it might also be involved in stimulating

2:28:39.760,2:28:47.680
the reverse transcriptase step. So the capsid 
is somehow involved in making that start,

2:28:47.680,2:28:54.080
and I think that is also quite a new discovery 
that people have had within the last five years.

2:28:54.080,2:28:58.800
Got it. Anyway, keep going. 
Thanks for pausing for me.

2:28:58.800,2:29:04.880
Not at all. Okay. So we've talked about what the 
capsid looked like. What does lenacapavir have

2:29:04.880,2:29:13.920
to do with this? This also, it really helps to 
have a visual, but I'm going to try to explain it

2:29:14.640,2:29:20.560
in words as well. So we have this structure 
where, the capsid is made out of these proteins.

2:29:20.560,2:29:26.080
The proteins are sometimes in hexamers, 
sometimes in pentamers; I guess if you

2:29:26.080,2:29:31.359
imagine that these hexamers or pentamers 
are fitting together, they're a bit like,

2:29:31.359,2:29:39.280
putting your hands together with your knuckles. 
They're kind of fitting together between your

2:29:39.280,2:29:47.359
fingers. And imagine doing that lightly; 
you're not fitting them too strongly. So

2:29:47.359,2:29:53.200
there's a little bit of space and flexibility 
between your two hands. But then, lenacapavir,

2:29:54.479,2:30:02.879
it wedges itself into those gaps between 
both of your hands' knuckles. And that

2:30:02.880,2:30:10.080
means that it now becomes very stiff. Now 
you don't have this much ability to move,

2:30:10.080,2:30:18.640
to move the structure around. This stiffness 
becomes a problem in several ways during the

2:30:18.640,2:30:27.920
HIV's life cycle. I think that's quite cool just 
to think about how the overall shape of this

2:30:27.920,2:30:35.520
capsid changes, based on lenacapavir fitting 
into these little gaps between the proteins.

2:30:35.520,2:30:38.560
That is good. It's quite subtle as 
well. 'Cause it's not, I think of a

2:30:38.560,2:30:42.319
drug coming in and trying to nuke some 
structure, you know, blow it up. But

2:30:42.319,2:30:49.439
actually you're saying no, we're just going 
to change properties of how squidgy it is.

2:30:49.439,2:30:55.839
No, exactly. And yeah, I mean, I think it's 
fascinating. So we have all of these little

2:30:55.840,2:31:02.000
lenacapavir molecules. Lenacapavir itself 
is quite a small molecule, so it fits into

2:31:02.000,2:31:08.479
these little gaps. And if we go back to this 
lifecycle of where the capsid is imported,

2:31:08.479,2:31:16.080
so we have: the HIV virus has entered the cell, 
has released some of its contents, which includes

2:31:16.080,2:31:23.040
the capsid. The capsid is then trying to enter the 
nucleus. In order to get into the nucleus, it has

2:31:23.040,2:31:30.880
to fit through these holes, the nuclear pores. 
And to do that, it binds to certain proteins on

2:31:30.880,2:31:39.200
the nuclear pore. And it turns out that those 
binding spots are blocked by lenacapavir.

2:31:39.200,2:31:39.600
Okay.

2:31:39.600,2:31:46.160
So it binds in exactly the same spots where those 
proteins would attach and let it squeeze through

2:31:46.160,2:31:54.479
those gaps. And the second thing is, the capsid 
is now too stiff, because of lenacapavir blocking,

2:31:54.479,2:31:57.599
so it's less flexible.
Knuckles are engaged.

2:31:57.600,2:32:05.760
Exactly. Your knuckles are engaged. You can't 
squeeze through the little holes of the nucleus.

2:32:05.760,2:32:11.840
So that's another important thing.
Then, the next step, and this is super

2:32:11.840,2:32:17.280
interesting, so it's not just that one step that 
lenacapavir disrupts, but it's actually multiple

2:32:17.280,2:32:25.519
steps during this life cycle. Imagine that some of 
the capsids have still somehow made it through to

2:32:25.520,2:32:37.520
the nucleus. Now the capsid needs to dissolve and 
allow the RNA molecule out, and allow the reverse

2:32:37.520,2:32:45.439
transcriptase to turn the RNA into DNA. It needs 
integrase to turn that DNA into a part of our

2:32:45.439,2:32:54.479
own cell's DNA. But now it can't break; it can't 
dissolve. It can't uncoat anymore, because of this

2:32:54.479,2:33:06.639
rigidness. It's just too rigid. But also sometimes 
it's so rigid that it cracks too early, and that

2:33:06.640,2:33:15.280
early cracking makes it hard for the capsid 
stimulating the reverse transcriptase, to start

2:33:15.280,2:33:26.880
doing that process. So that is super interesting. 
We've now blocked it from entering the nucleus. If

2:33:26.880,2:33:33.120
some gets through, it's now not able to release 
its contents, or it breaks too early. And then,

2:33:33.120,2:33:41.040
there's a third part that lenacapavir disrupts 
as well. So imagine that you've now, somehow,

2:33:41.920,2:33:46.800
some of the virus particles or some of the capsids 
have still made it through, or maybe you're in a

2:33:46.800,2:33:54.800
different part of the HIV virus lifecycle. You're 
now trying to create these new virus particles,

2:33:54.800,2:34:04.399
the descendants of the initial one. In order to 
do that, we have this, the immature HIV virus, and

2:34:04.399,2:34:05.359
I remember, yeah,

2:34:05.359,2:34:11.439
This now has protease involved, right? So 
we have protease cutting up these giant

2:34:11.439,2:34:19.839
polyproteins into their proper form, and 
we have this new capsid trying to form,

2:34:19.840,2:34:27.920
to surround all of the RNA and the other enzymes. 
But what happens with lenacapavir is that, because

2:34:27.920,2:34:41.200
it's in this too stiff kind of formation, it's 
unable to form in the correct shape and it just

2:34:41.200,2:34:50.399
doesn't fully form. So there's this image here, 
where you can see that the normal way that all

2:34:50.399,2:34:58.719
of these proteins in the capsid form is that- We, 
okay, so we have all these hexamers and pentamers;

2:34:58.720,2:35:04.479
they create these little clusters, they 
somehow self-assemble. Maybe this is just

2:35:04.479,2:35:09.520
something that we don't understand yet, but 
somehow they self-assemble into a bigger.

2:35:09.520,2:35:14.160
Which blows my mind just from, if I'm 
visualising the cytoplasm of a cell, I'm like,

2:35:14.160,2:35:20.240
there's so much going on there. How do these all 
stay together and not get distracted? But anyway.

2:35:20.240,2:35:28.719
Exactly. So we're getting these clusters of 
multiple hexamers and pentamers. That happens,

2:35:28.720,2:35:33.840
but now because of the stiffness, 
the shape isn't forming correctly,

2:35:33.840,2:35:41.600
and the full shape just doesn't work anymore. 
So I mean I think this whole process to me was

2:35:41.600,2:35:48.960
really interesting to learn about. I knew that 
lenacapavir somehow had this amazing effect,

2:35:48.960,2:35:54.479
at very low concentrations. It somehow has 
a really long effect lasting for at least

2:35:54.479,2:36:01.280
six months. But I didn't really know about the 
mechanics of that works. And the other thing I

2:36:01.280,2:36:08.160
didn't realize, until reading for this episode, 
was multiple steps are inhibited. So it disrupts

2:36:08.160,2:36:14.479
this process at multiple steps. I think that... 
Now, I would say this doesn't mean that it's

2:36:14.479,2:36:21.679
impossible to develop resistance against it, but 
it does probably explain why it's so effective,

2:36:21.680,2:36:25.200
that it's targeting these multiple 
steps. It's reducing the probability

2:36:25.200,2:36:31.599
that an infection can and multiply, and 
so I think that was very interesting.

2:36:31.600,2:36:34.960
Yeah, I was going to say, it sounds almost 
like a combination drug itself. If you've

2:36:34.960,2:36:41.600
got three different stages it's acting at. But I 
don't know if the mutations that would generate

2:36:41.600,2:36:46.479
resistance to are correlated there or not. 
It makes sense to me they'd have a fail safe,

2:36:46.479,2:36:51.040
for mopping up the capsids that 
try and form out the other end.

2:36:51.040,2:36:56.560
I think what I've learned is that there 
is still... drug resistance can form,

2:36:56.560,2:37:03.439
but it's something that forms if someone is 
on long-term treatments with lenacapavir;

2:37:03.439,2:37:07.599
it's not something that they would 
have before the infection. And so,

2:37:07.600,2:37:16.080
it is still useful as a preventive, sort 
of, anti-infection tool. But as a treatment,

2:37:16.080,2:37:21.200
there have been cases of people developing 
resistance and that would be through exactly

2:37:21.200,2:37:25.276
how lenacapavir fits into the little 
gaps between the capsid proteins.

2:37:25.276,2:37:26.000
The gaps.

2:37:27.600,2:37:29.840
If you change the shape of that, yeah.

2:37:29.840,2:37:36.000
It makes just from a selection pressure point of 
view. If you've got many more viral particles,

2:37:36.000,2:37:39.359
you're infected already, then 
you've got higher probability,

2:37:39.359,2:37:47.280
probably, of a mutation that's good 
for HIV. Neat. I like it. How did

2:37:47.280,2:37:52.719
it come about from a- was this the 
first capsid inhibitor? Lenacapavir?

2:37:52.720,2:37:57.200
This wasn't the first capsid inhibitor, 
I think people had been trying to target

2:37:57.200,2:38:08.160
it for a while. And in 2010, Pfizer had 
developed another molecule called PF-74,

2:38:08.160,2:38:17.519
and that seemed to bind to this little gap, 
this pocket as well, and it could block and

2:38:17.520,2:38:27.280
hyper stabilize the whole capsid shape. But 
it didn't work so much in the human body;

2:38:27.280,2:38:32.639
this drug, it was taken orally, and it 
just didn't stick around in your body.

2:38:32.640,2:38:41.120
And so they gave up on the drug and they 
started working on other drugs instead. Instead,

2:38:41.120,2:38:50.240
what happened was Gilead tried to build 
on this PF-74. It seemed quite promising,

2:38:50.240,2:38:58.399
because you do have this, you have a potential 
way of targeting the capsid protein. But we're

2:38:58.399,2:39:06.719
just missing out on making it more available 
in the body and long lasting. And so they did

2:39:06.720,2:39:12.399
something called "parallel synthesis". They just 
tried creating lots of very similar compounds

2:39:12.399,2:39:18.160
to that — using the same molecule, but then 
adjusting it in lots of different ways. And

2:39:18.160,2:39:24.319
the way that happens in the lab is, you have lots 
of test tubes or plate wells, where you have the

2:39:24.319,2:39:33.359
initial molecule PF-74, and then you run lots of 
different reactions in those different test tubes,

2:39:33.359,2:39:39.200
under the same conditions — so you have maybe 
three or four, with some reaction going on and

2:39:39.200,2:39:50.479
so on. And they did a lot of iteration based 
on that and eventually resulted in lenacapavir.

2:39:50.479,2:39:56.799
Got it. Well, thank you PF-74 for trying 
first. It's interesting. It sounds like

2:39:56.800,2:40:03.840
we had the idea and there was some binding 
going on and it was working okay. But then,

2:40:03.840,2:40:09.120
the practicality of the human body — you've got 
a lot more steps. You don't just have to bind

2:40:09.120,2:40:16.080
the part of the HIV virus, you also have to 
survive the machinations of the human organs.

2:40:16.080,2:40:22.319
And this is the really interesting thing 
that we'll talk about later. In terms of,

2:40:22.319,2:40:28.479
how does this actually stick around for so long 
in the body? How does this have an effect for over

2:40:28.479,2:40:35.359
six months? That itself is really impressive to 
me. I thought they could maybe just briefly talk

2:40:35.359,2:40:40.960
about some of the different kind of iterations 
that people do and how important they are

2:40:40.960,2:40:43.120
To lenacapavir, or?

2:40:43.120,2:40:52.240
Yeah, exactly. So we have PF-74. Actually 
don't you know someone who worked on this?

2:40:52.240,2:40:59.040
I do, I do. I was wondering whether to say, 
but I think I may have mentioned her earlier

2:40:59.040,2:41:07.600
and she in her PhD worked on PF-74, so in an 
academic setting, not at one of the companies,

2:41:07.600,2:41:11.600
but I think she knows a bunch of the 
iterations that you're about to tell me,

2:41:11.600,2:41:16.319
but I don't know; I should have 
asked her more questions first.

2:41:16.319,2:41:23.120
Yeah, I mean it's so fascinating to 
know that there are people surrounding

2:41:23.120,2:41:28.720
us that have been responsible for these huge 
breakthroughs, and they're just normal people-

2:41:28.720,2:41:29.220
I know.

2:41:30.240,2:41:31.679
Sometimes friends.

2:41:31.680,2:41:36.640
I had to tell her I was about to talk 
about lenacapavir, before I realised

2:41:36.640,2:41:43.680
that she had done all the- well, so much of 
the work, that led up to it. It's wild. Yeah.

2:41:43.680,2:41:50.319
I think, yeah, it's incredible. Okay, so 
we have PF-74, what happens now? So we

2:41:50.319,2:41:54.880
have all of these little reaction 
test tubes going on. One of them,

2:41:54.880,2:42:00.960
they introduced a hydroxyl group — that's 
basically an oxygen and a hydrogen,

2:42:00.960,2:42:07.600
and then they added an indole ring — that's 
a fused ring structure. So there's multiple

2:42:07.600,2:42:16.000
atoms in a ring with a nitrogen group. This 
massively increased the effect of the drug,

2:42:16.000,2:42:20.960
but it didn't work in the body, because it 
was broken down by enzymes in the liver.

2:42:20.960,2:42:21.460
Okay.

2:42:21.840,2:42:29.040
Okay, we've got, this one, it didn't really work 
out. Then they tried another type of ring. This

2:42:29.040,2:42:37.120
is a six structured ring with one nitrogen atom 
instead. This then improved how stable it was

2:42:37.120,2:42:43.840
in the body. It was not broken down very quickly 
anymore, but now, the effectiveness was reduced.

2:42:44.640,2:42:49.120
I love how we're making atomic 
level differences here. Anyway,

2:42:49.120,2:42:51.840
keep going. Okay, so we tried that one. No luck.

2:42:51.840,2:42:59.680
They did a bunch of other changes. There 
were some seven or eight compounds they made,

2:42:59.680,2:43:05.520
before getting to this breakthrough. This actually 
comes from this really interesting book that I was

2:43:05.520,2:43:14.160
reading called Drug Development Stories, and 
these researchers just put together how all of

2:43:14.160,2:43:20.240
these different types of breakthrough drugs in the 
last few years were developed. And so this last-

2:43:20.240,2:43:24.960
I have few books like that that I'm trying 
to, I can't remember if I own that one,

2:43:24.960,2:43:28.399
but they're fun to flick through some. 
I'm going to have to check my bookshelf

2:43:28.399,2:43:33.759
when I get back to San Francisco. If 
not, it's going on the order list.

2:43:33.760,2:43:40.160
It's often hard to find the exact stories 
about- behind how these drugs are developed,

2:43:40.160,2:43:45.439
and I assume part of that is because it might be 
some kind of trade secret or something like that.

2:43:46.319,2:43:51.599
But when I do come across someone writing a 
retrospective or, you know, someone giving

2:43:51.600,2:43:56.880
me the details, it's just a completely 
different picture. It really helps you

2:43:56.880,2:44:02.720
understand exactly what they struggled with, how 
they were thinking about the process, and so on.

2:44:02.720,2:44:04.800
So if you're listening and you're working on

2:44:04.800,2:44:09.520
drug development in some form and you're 
wondering, oh, if I write up my process,

2:44:09.520,2:44:15.600
is anyone even going to read it? We're gonna 
to read it. Write it up! Saloni needs it!

2:44:15.600,2:44:22.319
I need it. I love it. Okay, so now 
we have this breakthrough compound,

2:44:22.319,2:44:28.080
I think, which might've been the ninth 
one that they made as an adjustment.

2:44:28.080,2:44:34.720
They now replaced the amide group with an 
amino indisole group. And that is a ring

2:44:35.600,2:44:44.240
with two nitrogen atoms. This improved both the 
potency, how well it fit, and also reduced the

2:44:44.240,2:44:51.679
level of breakdown in the body. This felt 
like, okay, we're making something that

2:44:52.640,2:44:57.359
effectively tracks both of these key things 
that are important in drug development.

2:44:57.359,2:44:58.319
Got it.

2:44:58.319,2:45:04.880
Then it was just slight adjustments, slight 
tweakings from there. They added another

2:45:05.600,2:45:10.479
amino group, they changed the placement of 
the amino group. They introduced a sulfone

2:45:10.479,2:45:17.040
group — that's a sulphur with two oxygens, 
and that was what resulted in this very

2:45:17.040,2:45:23.799
highly potent molecule that was very stable 
in the body, and that became lenacapavir.

2:45:23.799,2:45:29.920
Wow! We made it. Highly potent, 
highly stable. And it's funny,

2:45:29.920,2:45:35.439
I think when you say, "The body's not breaking 
it down well", to me it sounds like, "Oh God,

2:45:35.439,2:45:40.160
the body's not breaking it down well?" And 
in fact, we've been seeking a molecule the

2:45:40.160,2:45:45.920
body doesn't break down well, because it can 
last for longer and protect you for longer.

2:45:47.120,2:45:52.800
It's different purposes, right? If you're trying 
to reduce maybe the side effects of some drug,

2:45:52.800,2:45:59.760
you want it to get broken down very quickly, 
just have its action; disappear. But if you're

2:45:59.760,2:46:04.640
trying to develop some drug that has a 
very long lasting effect, then you want

2:46:04.640,2:46:13.600
it to stick around for a long time. So, I mean, 
finding something that is both very safe, very

2:46:14.240,2:46:24.080
highly effective — potent — and also very stable 
in the body that makes a great long lasting drug.

2:46:24.080,2:46:27.920
That's the trio. And we got there.

2:46:27.920,2:46:34.240
The other interesting thing I learned about 
lenacapavir, and I think you might maybe come

2:46:34.240,2:46:40.719
to this later on, but it's the drug with the 
most fluorine atoms in it that's approved by the

2:46:40.720,2:46:49.680
FDA in the US. So fluorines are often used to 
increase the stability, I think, in the body,

2:46:49.680,2:46:57.920
and lenacapavir has 10 of these atoms in the 
whole molecule. Usually, that has led to drugs

2:46:57.920,2:47:04.319
that are unsafe in some way, but in this case 
it has a very high safety profile as well.

2:47:04.960,2:47:11.279
I remember seeing that, and I'm not enough of 
a chemist to tell you why there aren't more,

2:47:12.880,2:47:14.640
but I do like the idea that there's fluorine

2:47:14.640,2:47:20.080
in the aqueous solution. There's 
fluorine in the water, you know.

2:47:20.080,2:47:25.279
What did we- we talked to your 
friend, Sanela, is that right?

2:47:25.279,2:47:25.920
Yeah.

2:47:27.040,2:47:34.319
Was there stuff that you learned from her on 
lenacapavir and capsid inhibitors as well,

2:47:34.319,2:47:36.719
beyond what we've talked about?

2:47:36.720,2:47:44.000
The thing that she emphasized to me was really in 
line with what you just said, about how iterative

2:47:44.000,2:47:56.240
the process was. That starting with PF-74, as a 
great binder, there was a lot of tries for just...

2:47:56.240,2:48:01.359
getting the other properties that could make this 
into a useful drug. So I think the thing she was

2:48:01.359,2:48:09.120
really struck by was the "PK", as drug developers 
say, looking really good for lenacapavir.

2:48:09.120,2:48:10.479
What's the PK?

2:48:10.479,2:48:19.040
The pharmacokinetics, I want to say, and there's 
again: this will betray that I'm not a medicinal

2:48:19.040,2:48:25.760
chemist. But pharmacokinetics and pharmacodynamics 
are the two things that you'll hear people talk

2:48:25.760,2:48:32.319
about all the time, about how the body processes 
drugs. But I mean also, she was really emphasising

2:48:32.319,2:48:37.920
that previous molecules' stability was an 
issue. And then seeing lenacapavir being so

2:48:37.920,2:48:44.560
stable and so low toxicity. You know, you have 
such a small amount and it can stick around for

2:48:44.560,2:48:52.640
six months; and it's getting to the 20th or 15th 
iteration of the initial principle. Then getting

2:48:52.640,2:48:59.439
those great properties, it was wonderful 
for her to see as someone in the field. And

2:48:59.439,2:49:06.000
as someone outside of the field, I didn't realise. 
I'd heard of lenacapavir as this miracle drug,

2:49:06.000,2:49:11.279
and I didn't realise, of course, what came 
before. There's a lot of steps to get there.

2:49:11.279,2:49:16.719
It's not that you suddenly come out swinging 
and suddenly discover the exact perfect thing.

2:49:18.160,2:49:23.680
Right. I mean, the thing that I remember, 
I did my undergraduate degree in biomedical

2:49:23.680,2:49:30.640
science like 10 years ago, I think? And I 
remember a little bit about pharmacology

2:49:30.640,2:49:39.360
on PK as, it's more, you're measuring 
both how fast is the drug broken down,

2:49:40.240,2:49:47.679
how much of the drug is broken down, how quickly, 
but also how available is it? How much does it

2:49:47.680,2:49:54.800
actually get to the organs that you need it to 
get to? Is it able to have its effect there?

2:49:54.800,2:50:01.600
And I think with lenacapavir, it seems to be very 
effective even at small doses. So if you imagine

2:50:01.600,2:50:07.920
that over time, so you have this injection 
first, of lenacapavir, and then, over time,

2:50:07.920,2:50:15.359
there's this decaying exponential curve. So it 
just quickly starts to break down, and then that

2:50:15.359,2:50:25.279
break down process slows down. But even at the 
very low levels, it's still very effective, and

2:50:25.279,2:50:37.200
able to block this little site within the capsid 
proteins. And that is what makes it so effective.

2:50:39.439,2:50:48.559
Well, let's do a quick detour, if you'll let 
me, on long-acting drugs other than lenacapavir.

2:50:48.560,2:50:58.720
Because the principle we're talking about 
applies not just to this molecule. The nature

2:50:58.720,2:51:03.840
of a long-acting injectable or a long-acting 
drug, well, what makes it long lasting? What

2:51:03.840,2:51:09.520
makes it long-acting? It's relative to what we're 
used to; relative to immediate release drugs.

2:51:09.520,2:51:16.640
You know, if you think about the PrEP drugs that 
we talked about earlier in the episode — those you

2:51:16.640,2:51:24.000
take daily, and that's because you want to have 
enough of the active ingredient in circulation,

2:51:24.000,2:51:32.960
in case HIV enters your system during that period. 
But pretty quickly, most drugs get metabolised and

2:51:32.960,2:51:39.920
get filtered out and leave your body in urine, 
and sometimes other ways, but that's the main one.

2:51:40.800,2:51:48.319
So an issue that you mentioned is: if you have to 
take a drug every day, you might forget to take

2:51:48.319,2:51:52.799
it. You might run out of the drug and not have 
time to get a refill for a few days, or a week,

2:51:52.800,2:52:00.560
or I actually am behind on refills, because I'm in 
New York right now. Or you might not want people

2:52:00.560,2:52:09.440
who you live with to see that you take pills 
regularly, so keeping them around is not ideal.

2:52:10.240,2:52:16.719
This is important with HIV as well, 
because some of the preventive drugs

2:52:16.720,2:52:21.680
might also look like treatment drugs. 
I think you mentioned this earlier. And

2:52:21.680,2:52:27.840
so there's also the stigma around people 
thinking that you have an HIV infection,

2:52:27.840,2:52:31.920
and then they're worried about that, even 
though these drugs are very effective and

2:52:31.920,2:52:40.560
it's very unlikely to transmit, if you're using 
these drugs. But I think this whole thing of:

2:52:40.560,2:52:47.920
"how do people actually take it in their daily 
life?" is so relevant through drug development.

2:52:47.920,2:52:53.840
Absolutely. And the long lasting prevention 
that people might be most familiar with is birth

2:52:53.840,2:53:04.720
control. Where a lot of, some similar issues are 
a big part of what can drive different women to

2:53:04.720,2:53:11.200
want to make different choices with birth control. 
So the pill was first approved by the FDA in 1960,

2:53:11.200,2:53:17.040
after development in the '50s, as daily oral 
contraception. And then from the '60s onwards,

2:53:17.040,2:53:25.680
there was a lot of work to see if you could 
make the different options for birth control

2:53:25.680,2:53:32.800
long lasting. So can you have hormones that are 
inside a silicone tube of some form, that you can

2:53:32.800,2:53:40.720
control the diffusion over time, so that you don't 
have to take daily pills or take a daily hormones.

2:53:41.920,2:53:48.800
And now there are multiple options for that. 
So you can have hormonal IUDs that release

2:53:48.800,2:53:56.880
progestin slowly over time. You can have arm 
implants. The approach there, is to take an

2:53:56.880,2:54:03.520
existing biological molecule, or synthesise 
an alternate of a hormone, and then have a

2:54:03.520,2:54:08.479
different packet that controls the diffusion.
So it's a bit different than what we just talked

2:54:08.479,2:54:19.599
about with lenacapavir, where the drug itself 
is so insoluble and stable and it itself sticks

2:54:19.600,2:54:24.800
around. Whereas, let's say you have a drug 
that doesn't naturally stick around, well,

2:54:24.800,2:54:34.399
maybe you can design a delivery mode — a polymer, 
a liposome, that you can keep it inside of, and

2:54:34.399,2:54:40.080
slowly diffuse, and you can achieve the same goal. 
Maybe you could just put it in a drop of oil,

2:54:40.080,2:54:45.120
or you could suspend it in some other 
way, maybe you could put it in a device.

2:54:45.920,2:54:52.240
And sure enough, for HIV one new mode of 
prevention that we haven't talked much about

2:54:52.240,2:54:59.920
yet is vaginal rings — where you can have a slow 
release, you can insert a vaginal ring monthly,

2:54:59.920,2:55:06.560
say, and have a slow release that a 
woman, who wants to not be a risk of HIV,

2:55:06.560,2:55:13.600
has more control over if you have to get in 
some circumstances, than if she had to get

2:55:14.479,2:55:19.839
pills that were more visible to people in her 
household or her husband or something like that.

2:55:19.840,2:55:30.160
So there are a lot of different strategies here.
When it comes to HIV, what's interesting to me is

2:55:30.160,2:55:37.519
that there are a few long lasting drugs that are 
being tested now, or are near the finish line now,

2:55:37.520,2:55:46.880
that use different approaches and yet achieve 
this kind of similar goal. The three that come

2:55:46.880,2:55:53.200
to mind for me are lenacapavir, which we've been 
talking about a lot, but then also cabotegravir,

2:55:53.200,2:55:59.840
which is already approved in some countries — 
in the US — which is this injection every two

2:55:59.840,2:56:06.479
months. And then also islatravir, and other 
follower drugs, that are being made by Merck,

2:56:06.479,2:56:12.160
that are oral drugs that you might be 
able to take once a month for prevention;

2:56:12.160,2:56:15.840
or for treatment, maybe you take once a 
week. There's a few different regimens

2:56:15.840,2:56:24.560
being tested. And my understanding of why 
islatravir sticks around for a month is-

2:56:24.560,2:56:26.080
Which one is islatravir?

2:56:26.080,2:56:29.024
Islatravir is a drug-

2:56:29.024,2:56:29.040
Is that Merck's?

2:56:29.040,2:56:37.439
Yes. Made by Merck. You know, imagine instead of 
taking daily oral PrEP, you take one pill once a

2:56:37.439,2:56:47.279
month say, and they've tested that regimen. 
What happens after you ingest that pill is,

2:56:49.680,2:56:56.800
the active ingredient sticks around 
intracellularly. So it's not- when you're

2:56:56.800,2:57:02.080
imagining lenacapavir, what you should imagine is: 
you've got a drug substance dissolved, in a liquid

2:57:02.080,2:57:09.439
that's 40% water; you are getting injected 
with that. The liquid's kind of dispersing,

2:57:09.439,2:57:14.879
and the drug is sticking around and forming 
a solid on the video just pointed at my arm,

2:57:14.880,2:57:20.880
but it's not in your arm, it's your stomach 
or in your butt. Then that active ingredient

2:57:20.880,2:57:28.000
is slowly dissolving, over the course of 
many months. Whereas, with islatravir,

2:57:28.000,2:57:34.479
you should imagine that the active ingredient is, 
in some way — and I wish I knew more about this,

2:57:34.479,2:57:40.319
I don't — is in some ways sticking around in 
your cell. So that if the HIV virus is entering

2:57:40.319,2:57:46.399
your cell, it's going to do its work, which is 
wonderful, but it's not some big depot or some

2:57:46.399,2:57:52.399
big lump. It's totally different and it might 
stick around for different biological reasons.

2:57:52.399,2:58:00.960
That is so interesting. I had no idea I was going 
to ask about this container or the package. Does

2:58:00.960,2:58:07.359
the package dissolve? Is the package, is that 
actually a separate molecule, or something

2:58:07.359,2:58:12.399
like that? But you answered my question. 
I guess that also varies with other drugs.

2:58:12.399,2:58:17.439
Yes, it does vary with other drugs, and 
the answer will differ for other drugs,

2:58:17.439,2:58:26.240
because there are different ways to achieve the 
goal. If you don't get a drug like lenacapavir

2:58:26.240,2:58:35.279
that has properties of slow degradation, then 
you might want to achieve the goal via packaging

2:58:35.279,2:58:40.319
your active ingredient differently.
One thing that I think is, in a way,

2:58:40.319,2:58:48.399
fortunate about both lenacapavir and islatravir 
and cabotegravir — which is also an injection

2:58:48.399,2:58:56.240
that's a suspended solid that slowly degrades — 
is that you don't need a very complicated package.

2:58:56.240,2:59:05.920
Because a complicated package adds complexity, 
manufacturing cost, and makes me less optimistic

2:59:05.920,2:59:13.600
that in the near-term a drug will get used in 
lower resource settings. Whereas the simpler you

2:59:13.600,2:59:21.920
can make the package, the more likely you can use 
a drug in many settings. In this case, it's just

2:59:21.920,2:59:29.520
the drug itself — well, simplifying a little — but 
it's mostly the drug itself in a syrup solution.

2:59:29.520,2:59:34.241
And you know what? I could take a look at that. I 
might be able to one day inject myself with that.

2:59:34.800,2:59:38.880
This is so interesting. I mean, the other 
thing that reminded me was, I think what

2:59:38.880,2:59:45.680
I read about the fluorine atoms, is that those 
help it stick around in your fat. So you have

2:59:45.680,2:59:53.120
the injection either in your abdomen or your 
butt, and then I think the fluorine atoms keep

2:59:53.120,3:00:04.080
it around the fatty areas, but also they create 
this little lump underneath your skin, right?

3:00:04.080,3:00:10.160
Yes. Well, I mean, absolutely. 
And as I think about the lump,

3:00:10.160,3:00:13.200
I was just thinking about it spatially, 
and I would love someone who's worked on

3:00:13.200,3:00:18.319
injectables and on lenacapavir to correct 
me in the comments, if I'm thinking about

3:00:18.319,3:00:25.920
it spatially wrong, but. Is lenacapavir a one or 
two millilitre injection? Do you happen to recall?

3:00:25.920,3:00:28.000
Yes. Yeah, it is one to two.

3:00:28.000,3:00:35.760
Okay. So my memory of how millilitres work is that 
that's one cubic centimetre. Is that right? Okay,

3:00:35.760,3:00:41.359
so now I'm visualising we've got some liquid 
that's one by one by one centimetre. That's quite

3:00:41.359,3:00:48.799
a lot. And you get injected somewhere, and the 
liquid does disperse it. You've only got some of

3:00:48.800,3:00:54.080
it left, but a good amount of what you're getting 
injected with is this high concentration drug that

3:00:54.080,3:01:01.040
becomes solid. So sure enough, you should expect 
that to be a lump. You just got injected with a

3:01:01.040,3:01:05.920
decent volume of stuff and fair enough, some 
of it precipitates and becomes solid. If there

3:01:05.920,3:01:10.560
wasn't a lump, I'd almost be confused. I'd be 
like, where did it go? I thought it was meant

3:01:10.560,3:01:16.480
to stick around for six months, and anyway, so I 
think lumps form, definitely in different people.

3:01:17.120,3:01:24.559
I think what I read was also that the first 
injection typically leads to this lump forming,

3:01:24.560,3:01:32.880
but the subsequent doses don't. So that makes me 
think that maybe there's some bodily reaction to

3:01:32.880,3:01:39.200
the drug substance that creates the lump, and that 
the second or third time, your body gets used to

3:01:39.200,3:01:44.639
it in some way, or it just doesn't have the same 
reaction. But that was also really interesting.

3:01:45.920,3:01:54.720
It reminds me a little bit about when people 
had smallpox vaccines, you could see that from

3:01:55.439,3:02:05.200
the little mark on their shoulder. With this, 
you have this tangible little bump in your butt

3:02:05.200,3:02:13.519
or your stomach that typically shows you 
if the drug has formed this little depot.

3:02:13.520,3:02:19.520
I'm curious about these other 
long-acting injectable drugs

3:02:19.520,3:02:27.279
that you've read about. So how else do they 
differ from regular drugs that only last a

3:02:27.279,3:02:33.759
short amount of time? Are they more expensive? 
What's their safety like? What's all of that?

3:02:33.760,3:02:39.680
Great questions. And they, again, are going to 
have variable answers rather than an easy one.

3:02:40.479,3:02:47.519
It depends on what package you're including. 
On the safety front, you really want to have,

3:02:47.520,3:02:52.320
if anything, higher safety standards when 
you're doing something long lasting. Because

3:02:53.200,3:02:59.040
with a drug that washes out of your system 
within a day, there's only so bad it can be.

3:02:59.040,3:03:03.840
But with something that sticks around for a 
long time, you want to make sure it's harder

3:03:03.840,3:03:08.240
to get rid of, so you want to make sure that, 
before you get injected or before you take it,

3:03:08.240,3:03:15.200
there's not going to be toxicity or longer issues.
I think that that is part of why, with lenacapavir

3:03:15.200,3:03:22.399
and with other drugs, you get an oral lead-in 
for two days before you get long injections,

3:03:22.399,3:03:30.639
just to check tolerability. It's interesting 
though, on efficacy as well, that you can

3:03:30.640,3:03:40.240
sometimes have the same underlying chemical, 
that is trying to achieve some medical goal,

3:03:40.240,3:03:47.200
but it can be higher efficacy in a long 
lasting injectable for a couple reasons.

3:03:47.200,3:03:55.840
One that we've mentioned is the real world and 
having- it's much more reliable if you just have

3:03:55.840,3:04:02.479
to get one injection or swallow one pill, than if 
you have to remember to do it all the time. But

3:04:02.479,3:04:08.559
another is more chemical or more to do with the 
body — which is that when you take daily pills,

3:04:08.560,3:04:16.560
you get this spike in how much of the drug 
you have, relatively soon after you take it,

3:04:16.560,3:04:23.520
operating in the bloodstream or elsewhere, that 
then decays relatively quickly. So you're kind of

3:04:23.520,3:04:29.920
doing this: spike and drop, spike and drop, spike 
and drop, for a daily drug. And that's not ideal.

3:04:30.880,3:04:37.279
Often you want to be in a therapeutic window 
that is not so spiky. With these long lasting

3:04:37.279,3:04:44.880
injectables, you have a lot more control, and you 
can tune that a lot more easily. You can predict,

3:04:44.880,3:04:49.680
okay, on a given day, how much am 
I going to have in my bloodstream,

3:04:49.680,3:04:55.520
based on how much I was initially injected with, 
and how much time it's been since then. You can

3:04:55.520,3:05:01.200
really hit that therapeutic window perfectly, so 
you can end up with a chemically more effective

3:05:01.200,3:05:07.439
drug. And you know, as I said, you want 
to check the safety a bit more though.

3:05:07.439,3:05:12.399
That totally makes sense. So the 
therapeutic window — that's the range of,

3:05:13.439,3:05:20.240
I guess, the volume or something of that drug in 
your system, and having that in the ideal range,

3:05:20.240,3:05:25.439
right? Yeah. And I guess the other 
thing I was thinking about was:

3:05:25.439,3:05:30.799
maybe it's not just about the predictability, 
it's also that there's less of a fluctuation.

3:05:30.800,3:05:40.399
Maybe some people react badly to these spikes, 
or the lack of- if the dosage suddenly drops,

3:05:40.399,3:05:46.559
does that mean the effectiveness is now not high 
enough? But if you can manage to get a stable

3:05:47.520,3:05:53.760
level of this drug for really long time, 
that is probably better in some respects.

3:05:53.760,3:06:00.640
Yep, absolutely. And my mind goes to: 
what other diseases for prevention or

3:06:00.640,3:06:10.960
for treatment — other than HIV — would you want 
those properties most? That's beyond the bounds

3:06:10.960,3:06:17.920
of this short podcast. But the ones that I 
can't wait to learn more about myself are...

3:06:17.920,3:06:25.920
So, malaria is one, where if you are 
under five and live in West Africa,

3:06:25.920,3:06:31.520
if you're a kid in West Africa, you will get 
preventive malaria drugs during the rainy season,

3:06:31.520,3:06:36.960
where you're most likely to get malaria; 
that's about four months long. Currently,

3:06:36.960,3:06:44.080
kids will get three-days-in-a-row worth of drugs, 
each month. So that's four times three. And the

3:06:44.080,3:06:51.439
drugs, they don't taste great, or kids don't 
always like 'em, and don't always take them all,

3:06:51.439,3:06:56.559
all three days in a row. If you're a busy 
parent and your kid is making a scene,

3:06:56.560,3:07:04.240
then you might not make sure to force all 
12 of those doses. But if you could get a

3:07:04.240,3:07:13.040
long lasting injectable for a season, that could 
cover that season and make sure you had the right

3:07:13.040,3:07:17.120
amount of preventive drug in your system, 
if you got bit by a mosquito. My goodness,

3:07:17.120,3:07:24.240
that could be an enormous deal. I don't 
think, based on my knowledge of malaria drugs,

3:07:24.240,3:07:28.080
that we're close to rolling out something like 
that. But I do know that people are working on

3:07:28.080,3:07:34.479
that problem, and it's very interesting.
I think that tuberculosis is another area;

3:07:34.479,3:07:39.759
Hepatitis C is another area, rheumatic heart 
disease is another area. So a few infectious

3:07:39.760,3:07:48.080
disease areas that I'm sort of — ooh, I'm getting 
excited about, but don't know in super depth.

3:07:48.080,3:07:53.680
And then, beyond infectious disease too, 
there's a bunch of potential applications.

3:07:54.479,3:07:59.679
I think there's a psychosis drug; there's 
a schizophrenia or bipolar disorder drug,

3:07:59.680,3:08:05.200
I think, that's also long-acting. What I 
was thinking about was: maybe it's useful

3:08:05.200,3:08:10.559
for things that are hard to predict when you're 
going to be infected by them, so you would prefer

3:08:10.560,3:08:17.279
to take something that lasts a long time, and 
so for various infectious diseases, that seems

3:08:17.279,3:08:23.120
useful. Maybe also for chronic diseases, where 
you have the condition for a really long time,

3:08:23.120,3:08:29.279
therefore it's useful to have this long 
acting treatment, instead of doing something

3:08:29.279,3:08:35.759
where you have to take it every day. But, I 
mean, that kind of covers most diseases...

3:08:36.319,3:08:41.359
There could be lot that fit that description 
and that you should also think about:

3:08:41.359,3:08:47.679
when do you want something to end? And 
I think, for a lot of those diseases,

3:08:47.680,3:08:53.920
you would not want a drug necessarily that lasted 
for life. If you could have one stick around-

3:08:53.920,3:08:54.640
That is very true.

3:08:55.840,3:09:01.760
For some drugs, for example, if you think, "Oh, 
there's a chance I want to get pregnant in the

3:09:01.760,3:09:07.279
next few years," then you might want more control 
over when the drug's out of your system, in case

3:09:07.279,3:09:13.840
that drug hasn't been tested as much in pregnancy, 
or has been tested and isn't as safe in pregnancy.

3:09:15.040,3:09:22.800
There's one thing that in the literature that I've 
found interesting to read is: the limitations, in

3:09:22.800,3:09:27.439
children, of long lasting drugs. I was wondering, 
oh, that's such a shame because of, for example,

3:09:27.439,3:09:35.200
the malaria thing. I was just talking about why 
is there a limitation? One reason is that the

3:09:35.200,3:09:43.519
dose that you get given, of different drugs, is 
relative to how big you are. And the correct dose

3:09:43.520,3:09:49.760
is often bigger, the bigger you are. The trouble 
is, with a child, that in six months, their size

3:09:49.760,3:09:54.479
is going to change a lot, and that means you 
might want to be dealing with a different

3:09:54.479,3:09:59.759
drug dose in six months time. So there's reasons 
like that, that you don't want to last forever.

3:09:59.760,3:10:08.240
That contrasts, usually, with vaccines where 
you actually wouldn't mind having as long a

3:10:08.240,3:10:18.080
lasting memory response as you can. And there 
are some vaccines that are known to be pretty

3:10:18.080,3:10:24.160
poor at generating long-lasting memories: 
flu vaccines are sort of a famous example;

3:10:24.160,3:10:30.639
the malaria vaccines so far aren't that 
durable, but are getting better. And then,

3:10:30.640,3:10:36.960
really, the best vaccine immunologists talk about 
is yellow fever vaccine, where you could live for

3:10:36.960,3:10:40.399
10,000 years and you wouldn't be getting 
yellow fever after you got that vaccine.

3:10:40.399,3:10:42.960
Wow, I didn't know that.

3:10:42.960,3:10:43.920
Yes. Well, that will-

3:10:43.920,3:10:46.800
I can live for 10,000 years?

3:10:46.800,3:10:49.200
That's what I've heard, but I don't know how they

3:10:49.200,3:10:53.120
extrapolated that. I think you can 
live for 10,000 years, and I think-

3:10:53.120,3:10:56.960
I mean, I kind of want to 
live for 10,000 years now,

3:10:56.960,3:11:01.359
just to see this vaccine 
lasting that long against it.

3:11:01.359,3:11:07.040
Imagine if you got a challenge with yellow 
fever at the end of 10,000 years to prove it,

3:11:07.040,3:11:12.399
and then it actually got you. Throwing 
up black bile and everything else.

3:11:12.399,3:11:18.319
The other thing I was thinking about, when 
you mentioned that on the size of the body,

3:11:18.319,3:11:25.840
was with lenacapavir: so, what I understood was, 
okay, the fluorine atoms keep lenacapavir in

3:11:25.840,3:11:30.399
fat tissue, but again, this is a problem for 
the same reason that you mentioned. If your

3:11:30.399,3:11:35.279
body size changes, that could affect 
how quickly it dissolves, I think,

3:11:35.279,3:11:44.160
or how much is remaining in this little depot. So 
imagine if you lost weight, for whatever reason,

3:11:44.160,3:11:49.760
you're on a diet or something, and somehow, 
suddenly, this drug becomes more potent or

3:11:49.760,3:11:55.200
effective in your body. I just thought that 
was a funny thing that I hadn't thought about.

3:11:55.200,3:11:59.599
No, totally. Another one is, that's 
hard to predict ahead of time is,

3:11:59.600,3:12:05.840
are you going to want to go on another drug, 
for another reason, that might have a negative

3:12:05.840,3:12:12.479
interaction with a long-lasting drug you're 
already on? My understanding of lenacapavir is,

3:12:12.479,3:12:16.319
they haven't discovered many drug 
interactions that are that concerning.

3:12:16.319,3:12:21.279
But that might apply to other long lasting 
drugs, so you got to think about that.

3:12:21.279,3:12:27.120
That's a really good point. The thing 
that reminds me of is grapefruit juice.

3:12:27.120,3:12:33.840
Have you read about this thing with- 
so, grapefruit juice has this chemical

3:12:33.840,3:12:40.800
that interferes with your liver's metabolism 
of lots of different drugs. And if you were-

3:12:40.800,3:12:43.200
I had grapefruit juice 
three days ago. I'm nervous.

3:12:43.200,3:12:44.720
Oh, did you? Yeah.

3:12:44.720,3:12:45.439
Well that explains-

3:12:45.439,3:12:50.799
Well, I mean, it's not all of the drugs, but 
it seems like it's quite a number of drugs

3:12:50.800,3:12:57.439
that are affected by this. And so, if you're 
drinking grapefruit juice, for whatever reason,

3:12:57.439,3:13:02.080
it sometimes makes various 
drugs last longer in your body,

3:13:02.080,3:13:05.680
because it interferes with their 
breakdown. This is true for, I think,

3:13:05.680,3:13:15.359
some other chemicals and drinks as well, but 
that's the most commonly-known in medicine.

3:13:15.359,3:13:18.719
So if you're getting towards the tail-end of

3:13:18.720,3:13:23.279
your six months of lenacapavir, just 
start doing shots of grapefruit juice?

3:13:23.279,3:13:29.120
I don't know if it interferes with lenacapavir 
specifically, but it seems to be a bunch of other

3:13:29.120,3:13:36.160
drugs. But I mean, it's just an interesting thing 
to think about, because now you're on this- okay,

3:13:36.160,3:13:42.080
you're on this six-monthly drug. What are the 
things that you now have to be thinking about,

3:13:42.080,3:13:48.880
to make sure that this drug is still working as 
expected? It's great to hear that it doesn't have

3:13:49.439,3:13:52.879
many drug interactions, but I 
think that's the other thing that,

3:13:53.600,3:13:56.960
if I was a drug developer, 
I would be thinking about.

3:13:56.960,3:14:00.479
And I'm sure there's more still 
to learn about drug interactions,

3:14:00.479,3:14:11.439
so we'll find out. The reversibility of some other 
preventive tools is a key part of why people want

3:14:11.439,3:14:18.559
to use them. For IUDs, for example, you can get an 
IUD removed and it will not affect your long-term

3:14:18.560,3:14:27.520
fertility. Whereas for lenacapavir, you got to 
wait it out. Once it's in there, it's in there.

3:14:28.080,3:14:34.399
I would find that exciting. But yeah, you're 
right. If there's something that could go wrong,

3:14:34.399,3:14:38.399
that is something that, this 
is why we need these long

3:14:38.960,3:14:44.640
clinical trials to make sure that this drug 
is safe, in the way that people take them,

3:14:44.640,3:14:53.342
in their daily lives. Maybe we should talk 
about the clinical trials and what they should-

3:14:53.342,3:14:57.439
I would love to, yes. I was wondering 
about the clinical trials. How did we

3:14:57.439,3:15:05.200
confirm that lenacapavir does actually work 
in the way we hoped, and that it is safe?

3:15:06.800,3:15:10.880
Where did we go once the scientists 
had done the experiments in the lab?

3:15:13.200,3:15:21.599
The period when lenacapavir was developed 
was the late 2010s. They did some molecular

3:15:21.600,3:15:28.479
studies to develop lenacapavir. They then did 
a phase one study — which is you're testing

3:15:28.479,3:15:35.919
the safety in a small number of participants, and 
you're checking basic things about the properties,

3:15:35.920,3:15:43.600
the pharmacokinetics, how long does it stay in the 
body, how effective is it in very specific ways,

3:15:43.600,3:15:49.360
in a small number of people — I 
think there's usually dozens or so.

3:15:50.160,3:15:57.760
Then after that, they moved on to phase two 
trials; this is a second part where it's a larger

3:15:57.760,3:16:03.120
number of people. Now, you're testing a little 
more about the safety, because now you have a

3:16:03.120,3:16:09.519
wider range of people with different backgrounds; 
they might be taking other drugs at the same time;

3:16:09.520,3:16:15.760
they have different behaviours and so on. So you 
can find out a little bit more about the safety.

3:16:15.760,3:16:23.360
But also, now with a larger sample, you can see 
how effective the drug is. They did this with

3:16:23.920,3:16:31.359
people who already had HIV and were 
taking other antiretroviral drugs,

3:16:32.000,3:16:37.840
and they saw how this combination of 
lenacapavir plus those other drugs worked.

3:16:37.840,3:16:47.439
Then, the breakthrough that really got me to 
notice this drug was their phase three trial; and

3:16:47.439,3:17:02.639
their phase three trial was- I think it started in 
2021. They had two different trials. One was with,

3:17:02.640,3:17:10.319
I think, one was with men; the other one was with 
cisgender women — specifically adolescent girls

3:17:10.319,3:17:16.880
and young women. This quite important because 
the transmission, and the effect of these drugs,

3:17:16.880,3:17:23.680
can vary for women who are trans because the 
route of infection, their sexual activity,

3:17:24.560,3:17:30.720
how that actually works, is different.
So they're focusing specifically on cisgender

3:17:30.720,3:17:40.239
women aged between 16 and 25; girls who were not 
using PrEP; they hadn't done HIV testing, or they

3:17:40.239,3:17:48.160
hadn't done it in the last three months. I think 
it was several thousand- it was around 8,000 women

3:17:48.160,3:17:58.559
in this trial across Uganda and South Africa. This 
is important because, if you're trying to test how

3:17:58.560,3:18:06.800
effective a drug is, you need enough, well- there 
need to be enough people, at least on the placebo,

3:18:06.800,3:18:14.080
who are getting infected, so that you can see 
what the difference would be with lenacapavir.

3:18:14.080,3:18:21.519
So they focused on these areas where HIV 
incidence rates were relatively high,

3:18:21.520,3:18:30.000
meaning that more than three or four people per 
hundred people were being infected per year. So

3:18:30.000,3:18:36.880
imagine 3% of the population of this age is 
being infected with HIV per year. That's,

3:18:37.520,3:18:43.840
to me, that's really high. This is quite 
common in some areas of South Africa and

3:18:43.840,3:18:52.000
Uganda. That means that it's much easier 
to tell if lenacapavir has an effect.

3:18:52.000,3:19:00.560
Because you'll detect in the other arm of the 
trial that there were HIV infections occurring.

3:19:00.560,3:19:05.920
Right. If you imagine, okay, if you were 
doing a trial in the UK or in the US,

3:19:05.920,3:19:14.080
where people are already taking PrEP or 
the rate of HIV is just so low to begin

3:19:14.800,3:19:21.279
with, then imagine no one in the 
placebo group gets HIV. How are

3:19:21.279,3:19:22.145
you going to tell if lenacapavir is better 
than that? There's nothing lower than zero.

3:19:22.145,3:19:28.880
You have to be able to distinguish, 
because happily, its transmission is

3:19:28.880,3:19:34.800
lower than it was twenty years ago or thirty.
Yeah, it's interesting. It's a very clear case,

3:19:34.800,3:19:43.920
I guess, of trials in- well, I'm in the US, you're 
in the UK, but I may benefit from this drug,

3:19:43.920,3:19:51.279
living in the US, based on trials that occurred 
in other countries — because those trials were

3:19:53.359,3:20:00.559
in higher transmission settings. So you can 
get a statistical answer to the question;

3:20:00.560,3:20:08.160
that would've been harder if you were just doing 
the trials in the US. So that's a kind of selfish

3:20:08.160,3:20:16.479
benefit that the US, and people like me, get 
from the global nature of clinical trials.

3:20:16.479,3:20:22.080
That's what I think makes it really important to- 
we'll come to talk about this later on, but this

3:20:22.080,3:20:27.519
is what makes it really important to think about 
how to actually get it to be accessible. How to

3:20:27.520,3:20:35.200
scale up this drug, in the future, to the people 
who need it. This wasn't developed without the

3:20:35.200,3:20:41.599
help of all of these participants who agreed to be 
in this trial, who are living thousands of miles

3:20:41.600,3:20:49.439
away from us, and who are responsible for this 
breakthrough being tested, and the fact that we

3:20:49.439,3:20:54.399
know that it works, and so on. One of the other 
things that I found quite interesting about this

3:20:54.399,3:21:01.679
trial is: how trials actually work in terms 
of the healthcare and the screening involved.

3:21:02.560,3:21:10.720
So I think when people think of a trial, maybe 
they don't realise that you're not just receiving

3:21:10.720,3:21:17.712
the treatment itself, but people are doing 
other types of screening to monitor how you

3:21:17.712,3:21:24.960
are responding to the drug. They're also, in this 
case, doing tests and screening for other types of

3:21:24.960,3:21:33.520
related diseases. What they did in this trial was: 
they provided individual counselling to people;

3:21:33.520,3:21:39.439
they provided condoms, lubricants; they'd 
have support for reproductive health in

3:21:39.439,3:21:44.879
general. They also provided treatments 
for other sexually transmitted infections,

3:21:44.880,3:21:51.760
and they did routine tests for some common 
sexually transmitted infections, like chlamydia,

3:21:51.760,3:21:59.439
gonorrhoea, and syphilis there. Which means that, 
by participating in this trial, not only do you

3:21:59.439,3:22:05.200
get the potential of this drug and also the side 
potential side effects, or risks, of participating

3:22:05.200,3:22:12.080
in trial — but you also get the actual 
healthcare because of the clinical trial setting.

3:22:12.080,3:22:19.040
I think that's something that people might not 
realise: we have this system where we have these

3:22:19.040,3:22:23.279
clinics, or these hospitals, that are running 
these clinical trials; they're also providing

3:22:23.840,3:22:30.239
care to the people in the trial. When we think 
about how to actually set up these clinical

3:22:30.239,3:22:37.359
trials in countries in Africa, it's not just about 
doing the testing. The researchers, sometimes,

3:22:37.359,3:22:44.719
are involved in setting up new clinics, they're 
involved in recruiting staff to work on all of

3:22:44.720,3:22:51.760
this testing, and stuff like that, and that has 
various other benefits for the people in the area.

3:22:51.760,3:22:57.920
One thing that we've worked on at Open 
Philanthropy is — an area that's not HIV

3:22:57.920,3:23:08.239
but relates — is congenital syphilis. 
Syphilis is- nowhere is it incredibly

3:23:08.239,3:23:13.200
prevalent, but in some places, maybe one 
to three percent of people have syphilis.

3:23:15.279,3:23:21.200
That is not something you want for yourself, 
but it also is really bad if you're pregnant,

3:23:21.200,3:23:27.200
because you may have a birth complication 
basically, or a miscarriage, or being born

3:23:27.200,3:23:36.880
with congenital syphilis is very dangerous. 
One thing that HIV care has brought in many

3:23:36.880,3:23:45.760
countries is better antenatal screening, prenatal 
screening for HIV for many women. You'll get at

3:23:45.760,3:23:50.640
least one visit with a doctor, whereas in many 
areas before, if you were pregnant, you might

3:23:50.640,3:23:56.640
not have even had one visit with a doctor. 
So that infrastructure set up to screen for,

3:23:56.640,3:24:04.160
or to give you a touch point and some care while 
you're pregnant — mostly funded by the HIV world,

3:24:04.160,3:24:12.399
and by PEPFAR, and other donor countries — 
has enabled screening for syphilis as well.

3:24:12.399,3:24:18.960
There's now a dual test where — it's a rapid test; 
it costs just under a dollar — you can screen for

3:24:18.960,3:24:24.800
HIV and syphilis at the same time. And then, 
if you are positive for syphilis, you can get

3:24:24.800,3:24:32.560
relatively easy treatment on penicillin. That's 
such a clear example of the benefits of this

3:24:32.560,3:24:40.560
infrastructure, that were not initially planned 
from when HIV donors made those investments,

3:24:40.560,3:24:47.840
but that spill over. So it makes sense to me 
that, not only are there benefits from the

3:24:47.840,3:24:52.640
knowledge gained, of these lenacapavir trials, 
but people in the trials got better care too,

3:24:52.640,3:24:59.120
and there's now probably better trained doctors 
in the area and that kind of thing, yeah.

3:24:59.120,3:25:06.720
We have thousands of young women in this trial. 
First, they were tested for whether they already

3:25:06.720,3:25:13.279
had HIV at that point. This was useful, as I'll 
come back to later, this is useful to know: What

3:25:13.279,3:25:20.719
is the rate of HIV in the population? It turned 
out that it was about 2.5 per hundred people per

3:25:20.720,3:25:30.399
year, which is, so 2.5% of people in this age 
group are infected by HIV per year — which is,

3:25:30.399,3:25:38.559
to me, really difficult to think about. Those 
women would not necessarily benefit from this

3:25:38.560,3:25:42.800
treatment; that was not the purpose of this 
trial — it was to find out whether we could

3:25:42.800,3:25:50.479
prevent new infections. Those participants were 
not included in the rest of the study, I think.

3:25:50.479,3:25:53.200
So, you screen at the beginning for HIV;

3:25:53.200,3:25:56.960
if you're already positive, then 
this is not the trial for you.

3:25:56.960,3:26:05.840
Right. And the women who were HIV-negative were 
then randomly assigned to getting either this

3:26:05.840,3:26:12.880
lenacapavir injection, or- I think this 
is Descovy — it's the oral PrEP pill,

3:26:12.880,3:26:22.160
which is emtricitabine and tenofovir 
alafenamide. Or the third option was F/TDF,

3:26:22.160,3:26:24.369
I think there's another name 
for this, is that Truvada?

3:26:24.369,3:26:24.399
Truvada? That's Truvada, yeah.

3:26:24.399,3:26:32.799
That's Truvada; that's also an oral pill that 
people take daily; that is also emtricitabine,

3:26:32.800,3:26:40.479
but this time it's tenofovir disoproxil fumarate. 
And I think what they did was — because one of

3:26:40.479,3:26:47.919
them is an injection; the other two are oral 
pills — they actually gave fake version of the

3:26:47.920,3:26:51.520
opposites to all the participants, so that 
they don't know which one they're getting.

3:26:51.520,3:26:52.560
Oh, okay. Nice.

3:26:52.560,3:26:57.840
They're all getting an oral pill and 
they're also all getting an injection.

3:26:57.840,3:27:01.520
But some of the people who are getting 
the injection are getting lenacapavir,

3:27:01.520,3:27:05.840
the others are just getting a placebo — which 
is just some water or something like that.

3:27:05.840,3:27:11.359
Which, you know, I'm already curious about. 
Three things from what you said: number one is,

3:27:11.359,3:27:18.399
if I get the fake lenacapavir injection, does it 
form a depot? And can I tell that it's actually

3:27:19.040,3:27:25.680
fake, because it doesn't form a lump? But my 
other two reactions are maybe more fundamental.

3:27:25.680,3:27:30.720
So it sounds like there's no placebo here — in the 
sense of, there's no one who's getting no drugs,

3:27:30.720,3:27:36.720
because that would be unethical. We already 
have drugs that we know will reduce your

3:27:36.720,3:27:42.319
chance of acquiring HIV if you're on 
them. So the arms that you described

3:27:42.319,3:27:48.639
are the lenacapavir arm, descovy arm, 
truvada arm, and there's no zero arm.

3:27:48.640,3:27:55.600
Right, and in other clinical trials, you're 
not necessarily preventing people from getting

3:27:55.600,3:27:58.800
other treatments; they could be taking 
other treatments for the same disease,

3:27:58.800,3:28:04.479
at the same time. But, in this case, they 
wanted to see: How effective was this as

3:28:04.479,3:28:11.919
a prevention? And to get enough statistical 
power, you need everyone to be in the trial,

3:28:11.920,3:28:17.279
and you're having this situation where you 
don't want any of them to be taking PrEP

3:28:17.279,3:28:25.759
that you can't analyse in a consistent way. 
It's both ethical in the sense that they're

3:28:25.760,3:28:30.399
providing all of the participants with 
one of the three PrEP drugs, but it also

3:28:30.399,3:28:36.239
helps because it makes these comparisons 
simpler. They're not taking an additional

3:28:36.239,3:28:43.040
PrEP- some of them are not taking an additional 
PrEP drug that could complicate the analysis.

3:28:43.040,3:28:47.359
Okay. Got it. Makes sense. And then my third 
reaction was, it's interesting that Descovy

3:28:47.359,3:28:53.839
was in the mix because I thought that Descovy 
was not approved for use in cisgender women.

3:28:53.840,3:29:02.960
Yes, you're right, it hadn't been tested before in 
women. The company, Gilead had been criticised for

3:29:02.960,3:29:08.640
this, having this approved but not testing 
it beforehand. So this also functioned as,

3:29:08.640,3:29:14.960
not just the trial for lenacapavir, but it also 
tested how effective Truvada and Descovy are-

3:29:14.960,3:29:15.279
Great, okay.

3:29:15.279,3:29:19.040
-in the same population. So you get 
to have the answer to three questions.

3:29:19.040,3:29:25.040
You have the answer to how effective 
lenacapavir is, descovy, and truvada.

3:29:25.040,3:29:29.600
And then because of what you said, about the 
screening on the way in, can they compare it

3:29:29.600,3:29:34.319
to what they think the background rate, 
if you're on nothing, probably would be?

3:29:34.319,3:29:38.799
Yes, yes. So actually we have 
four things you can find out.

3:29:38.800,3:29:43.920
Well, that one's not measured. 
I guess that one's interpolated.

3:29:43.920,3:29:50.880
So you can now compare these three drugs, but you 
can also compare them all to not taking any drugs.

3:29:50.880,3:29:56.800
Okay. Well, I feel- I'm on the edge 
of my seat. So, what were the results?

3:29:56.800,3:30:04.160
What were the results? I'm going to show this 
chart. So this chart compares the outcomes in

3:30:04.160,3:30:08.319
each of these groups. The first bar 
is showing the background incidence;

3:30:08.319,3:30:14.319
these are the women who tested positive at 
the start of the trial, and around 2.4% of

3:30:14.319,3:30:27.587
them got infected with HIV per year in FTAF, 
which is Descovy? That had a rates of 2.02.

3:30:27.588,3:30:28.633
That's very similar to the background, right?

3:30:28.633,3:30:36.720
And in Truvada, it's around 1.69. These numbers 
are the point estimates, and that's our best

3:30:36.720,3:30:44.399
guess. But there's uncertainty around just 
what the number is; they all roughly fit

3:30:44.399,3:30:55.679
into the same range. So without drugs, Descovy, 
and Truvada have similar rates of HIV infection.

3:30:55.680,3:31:03.200
And I think the reason for this is because it's 
hard to take these on a consistent basis over

3:31:03.200,3:31:10.960
time — these are daily pills where the problem 
is: one, maybe remembering to take it every day;

3:31:10.960,3:31:16.720
second, having enough supplies with you 
every day; the stigma that we talked about;

3:31:16.720,3:31:21.760
maybe these issues around getting a refill 
on time. So there are all of these issues

3:31:21.760,3:31:29.120
that make it difficult to take these drugs in the 
long term, for women in Uganda and South Africa,

3:31:29.120,3:31:33.519
where this trial was done. What was 
interesting about the study is that

3:31:33.520,3:31:40.160
they could actually measure how regularly people 
were taking these drugs through blood testing.

3:31:40.160,3:31:42.479
Is that because they're taking samples? Wow.

3:31:42.479,3:31:50.160
They're taking dried blood spot samples from 
people and then they're testing the level of

3:31:50.160,3:31:56.479
tenofovir in their red blood cells. So this 
directly tells them what is a concentration

3:31:56.479,3:32:03.200
of this drug in this participant? 
With this, you can see how, over time,

3:32:03.200,3:32:11.120
the expected- how frequently people are taking 
them, that reduces over months of the study.

3:32:11.120,3:32:18.080
People, on average, are taking them quite often, 
but over time that adherence gets much lower,

3:32:18.080,3:32:24.800
so they're mostly taking them two or- one or 
two times per week, by the end of the trial.

3:32:26.319,3:32:35.439
And this gets back to this previous chart. So 
we've seen what happens with Descovy and Truvada,

3:32:35.439,3:32:46.639
what happens with lenacapavir? Zero women 
out of 2,134 get infected with HIV. That is

3:32:46.640,3:32:54.560
just an incredible result. And there is some 
uncertainty around that. The efficacy — so

3:32:54.560,3:33:01.760
how much lower the rate of HIV infection 
is, compared to the background rates — is

3:33:01.760,3:33:08.319
96 to a hundred percent. So it's somewhere- 
it's not completely effective, necessarily,

3:33:08.319,3:33:14.541
because there isn't a large enough sample to say 
that this is a hundred percent efficacy rate-

3:33:14.542,3:33:17.600
We can't rule out-
-but it is more than 96% of a reduction.

3:33:17.600,3:33:24.880
That is so incredible. So incredible. 
That is so incredible. Zero cases.

3:33:24.880,3:33:32.239
Zero cases. I was reading about this, I 
think, on STAT news — the health and medicine

3:33:32.239,3:33:37.679
magazine — and they mentioned how these results 
were presented at this conference, and they just

3:33:37.680,3:33:44.080
got this standing ovation where people, I mean, 
unsurprisingly, this is just an incredible result.

3:33:44.080,3:33:45.439
Unbelievable.

3:33:45.439,3:33:51.599
One last thing about lenacapavir, 
truvada, and descovy is the side

3:33:51.600,3:33:55.680
effects. We talked about- okay, 
we have this long-lasting drug,

3:33:55.680,3:34:02.640
that means there's a risk of long-lasting side 
effects as well. What happened in this trial was,

3:34:02.640,3:34:07.840
they didn't find that much of a difference between 
the different groups. Most of the side effects

3:34:07.840,3:34:13.279
were similarly seen in the different groups, 
and those are mostly things like headaches,

3:34:13.279,3:34:20.880
fevers — but again, when we record side effects 
in a trial, we can't necessarily, conclusively,

3:34:20.880,3:34:25.120
say that these are because of drugs. People 
have headaches, fevers just anyway in a typical-

3:34:25.120,3:34:30.239
We don't have a clean placebo here, I guess. 
I mean it's interesting that, when I think

3:34:30.239,3:34:35.679
about what friends report as side effects of 
oral PrEP — because it's oral, there's often

3:34:35.680,3:34:42.399
digestive issues or stomach problems. I wonder 
if they tested for that, because I would guess

3:34:42.399,3:34:47.599
intuitively that lenacapavir would've fewer of 
those ones. But I dunno if they tested for that.

3:34:47.600,3:34:55.120
And you would be right! They did find lower 
rates of nausea and vomiting with lenacapavir,

3:34:55.120,3:35:01.040
and I guess this is because of the difference — 
where it's not an oral drug, it's an injectable,

3:35:01.040,3:35:05.279
so it's not passing through your 
digestive tract and your stomach.

3:35:05.279,3:35:09.840
We don't need to do trials. You 
can just quiz me on my guesses.

3:35:09.840,3:35:18.479
The other thing was the little bumps on people's 
skin, the depots of lenacapavir — so that was

3:35:18.479,3:35:24.639
quite common. About 70% of the people 
who got lenacapavir develop these little

3:35:24.640,3:35:34.960
bumps — nodules — under their skin, and those 
typically shrunk down to normal after a while,

3:35:34.960,3:35:41.120
but also, the next doses tended to 
not- you wouldn't tend to see those.

3:35:41.120,3:35:46.800
Yeah, I'd love to learn more about that. To me, 
it sounds like success: I want a little nodule,

3:35:46.800,3:35:51.760
I want to know that drug is there, and, sure 
enough, in six months, I want the nodule to be

3:35:51.760,3:35:58.880
gone because the drug is gone. But I guess there's 
more going on in the body than I'm projecting.

3:35:58.880,3:36:06.720
I was also surprised that 30% of people who get 
lenacapavir don't develop these nodules. What is

3:36:06.720,3:36:12.000
causing this difference? And I, sadly, don't know 
the answer to that, but it's quite interesting.

3:36:14.479,3:36:20.399
I was about to ask about other trials, 
outside of women, but is there anything else,

3:36:20.399,3:36:27.439
on this trial, that I should know first?
I think, maybe, we should talk a little bit

3:36:27.439,3:36:35.759
about why was there almost no difference between 
Descovy and Truvada and the background rates,

3:36:35.760,3:36:41.279
and why is it that lenacapavir is so 
effective in these trials, or like,

3:36:41.279,3:36:48.719
in the real world? I think there's- so I would 
say that it's not entirely clear how effective

3:36:49.359,3:36:56.559
Descovy and Truvada are, compared to not taking 
any drugs. That's just because the uncertainty on

3:36:56.560,3:37:05.279
those is fairly moderate, so there isn't a very 
precise figure that we would have; it seems like

3:37:05.279,3:37:10.160
they're roughly similar, but there could still 
be some meaningful reduction that these are

3:37:10.160,3:37:16.720
providing. I guess the other reason is, if people 
are not taking it regularly — these oral pills

3:37:16.720,3:37:25.279
regularly — for whatever reason, in the long term, 
that reduces the efficacy. So even if someone

3:37:25.279,3:37:31.120
was taking it every day, it would be a higher 
level of effectiveness than in the real world,

3:37:31.120,3:37:38.000
where people are taking it less often. And I think 
this is why lenacapavir is so much more effective:

3:37:38.000,3:37:45.760
it's not just that it's highly effective on its 
own, but it's also really long lasting, and that

3:37:45.760,3:37:53.840
both of those contrast with Descovy and Truvada.
Yeah, I guess it's proving the hypothesis with

3:37:53.840,3:37:59.359
data; that's what we were wondering, 
and we were hopeful that lenacapavir's

3:37:59.359,3:38:04.319
long-lasting properties would pay 
off, and it looks like they did.

3:38:04.319,3:38:12.960
They did. And they did another trial with men and 
gender-diverse people — this was in six countries:

3:38:12.960,3:38:18.640
the US, I think, some South American and 
central American countries as well. Again,

3:38:18.640,3:38:22.640
the reduction you would see — with 
lenacapavir, on how likely it is for

3:38:22.640,3:38:31.600
an infection — was massive. It was two people 
out of thousands who contracted HIV versus,

3:38:31.600,3:38:36.000
I think, more than a dozen, or a 
dozen, in the other groups. So again,

3:38:37.359,3:38:46.880
this time their estimate was that there was 
82 to 99% efficacy for this drug. And again,

3:38:46.880,3:38:54.960
they saw that Descovy and Truvada had a very 
little impact compared to the background rate.

3:38:54.960,3:39:00.880
It's a big win. Okay, so there were 
two cases; so it wasn't zero. Two out

3:39:00.880,3:39:06.080
of a couple thousand, but the 
reduction in risk is enormous.

3:39:06.080,3:39:11.200
Is huge. I would say that, 
you shouldn't go away from

3:39:11.200,3:39:15.200
this thinking this completely prevents infections;

3:39:15.200,3:39:23.279
there is still a chance. But the reduction is so 
large that it's a really important breakthrough.

3:39:24.160,3:39:31.920
My hope, and I hate to be hopeful, but you can get 
non-linear population effects with transmission

3:39:31.920,3:39:41.120
reductions, where, if a transmission per event 
drops 90% at a background rate, the background

3:39:41.120,3:39:46.559
rate might start dropping too. I mean, it depends 
on interactions with treatment drugs and a lot of

3:39:46.560,3:39:52.800
other factors — but if you can imagine that, 
per event, your risk is going down and then,

3:39:52.800,3:39:59.680
over time, the background rate going down, 
that's actually a very large effect together.

3:40:00.560,3:40:05.520
That reminds me of this concept of the 
reproductive number, that a lot of people

3:40:05.520,3:40:12.960
would've heard about during the COVID pandemic- 
R nought. The R nought. So this is the number of

3:40:12.960,3:40:20.880
people, on average, that someone infects, if 
they've been infected. So if I was infected,

3:40:20.880,3:40:26.640
maybe I would infect three other 
people on average with the coronavirus,

3:40:26.640,3:40:32.479
in this case. The higher the number, 
the harder it is to control the disease,

3:40:32.479,3:40:40.000
but also the faster it spreads in the population. 
And if it gets below one — if I'm spreading it

3:40:40.000,3:40:47.760
to less than one person on average — 
eventually that disease will die out.

3:40:47.760,3:40:52.160
It's a dream that we can head towards 
now, maybe. Okay, that's PrEP.

3:40:52.880,3:41:00.479
Did lenacapavir get tested in trials for 
treatment as well, not just prevention?

3:41:00.479,3:41:07.759
Yes! It was actually tested and approved as a 
treatment drug before these preventive trials

3:41:07.760,3:41:14.800
and results. The first trial was as a treatment 
for drug-resistant HIV, where they're testing

3:41:14.800,3:41:21.840
lenacapavir plus the standard regimen that 
people are having, in people who have tried

3:41:21.840,3:41:30.880
many different treatments so far, and this is plan 
C or D. So it was effective in those trials; it

3:41:30.880,3:41:38.720
was approved as a treatment for drug-resistant HIV 
based on that. They also did a phase two trial,

3:41:38.720,3:41:46.000
where they tested it as a first line treatment — 
so that means that would be: How effective is it,

3:41:46.000,3:41:50.880
as the first treatment that someone receives 
if they have been diagnosed with HIV? So they

3:41:50.880,3:41:58.319
compared lenacapavir with other existing drugs. 
There are a bunch of other ongoing trials,

3:41:58.319,3:42:05.519
still. So I think there are more 
long-acting treatments, where they're

3:42:05.520,3:42:11.040
testing a combination of lenacapavir 
and islatravir, which you mentioned.

3:42:11.040,3:42:11.680
The Merck one.

3:42:11.680,3:42:17.279
That was the Merck oral pill, 
which are both long-acting, right?

3:42:17.279,3:42:18.479
Yeah.

3:42:18.479,3:42:22.799
And then, there were a bunch of other 
types of treatments: so they're doing

3:42:23.840,3:42:30.880
testing in children and adolescents; 
they're also testing whether it can

3:42:30.880,3:42:39.040
be used in people who have been receiving other 
types of antiretroviral drugs, but they still

3:42:39.040,3:42:46.720
have some HIV that's suppressed in their body. 
As we talked about, hours ago, one of the things

3:42:46.720,3:42:55.760
that you see with HIV is that these drugs can 
block the multiplication of HIV in your body. But

3:42:55.760,3:43:02.960
there are also particles that would stay in some 
parts of your body, hidden in silence, and these

3:43:02.960,3:43:10.720
reservoirs of HIV are difficult to get rid of. 
So is it possible to use lenacapavir to disrupt

3:43:10.720,3:43:16.080
these reservoirs? That is one of the questions 
that they're looking at in this other trial.

3:43:16.080,3:43:21.040
And then there are a bunch of others, so they 
think there's another one with lenacapavir plus

3:43:21.040,3:43:27.920
cabotegravir in people who have taken lots 
of other treatments. And these trials are

3:43:27.920,3:43:32.880
being conducted, essentially, almost like, 
all over the world — it's North America,

3:43:32.880,3:43:41.680
Europe, and Southern Africa; many different 
countries. It's basically quite a big process:

3:43:41.680,3:43:47.439
to do this lab testing, to develop these 
drugs, to then test them in some places,

3:43:47.439,3:43:53.519
in certain countries like the US or like Southern 
Africa, and then to scale it up, to these massive

3:43:53.520,3:44:03.600
trials — is quite interesting, but also, to me, 
quite impressive how fast this has happened. That

3:44:03.600,3:44:14.080
drug was only developed in 2018 and it's been — 
well, okay, it has been seven years since then.

3:44:14.080,3:44:20.160
I'm used to seeing timelines that are so long, 
and this was approved for the first time in 2022,

3:44:20.160,3:44:27.279
I think, as a treatment for drug-resistant HIV, 
which is only four years in trials. I think that,

3:44:27.279,3:44:35.439
I mean, on a personal level, I think that could 
still be sped up, but that is an impressive speed.

3:44:35.439,3:44:43.359
Yeah, it's a four year starting clock to finish 
line, not so bad. But we're not at the finish line

3:44:43.359,3:44:51.759
yet. The finish line is, is this really going to 
impact people's lives who are at risk of HIV? So

3:44:51.760,3:45:00.560
just summing up what we just covered: we have 
now looked at long-lasting drugs as a concept,

3:45:00.560,3:45:09.439
and other HIV long-lasting drugs, and long-lasting 
drugs in other areas that we're excited about,

3:45:09.439,3:45:14.399
but are still in development.
And then, you just outlined with lenacapavir,

3:45:14.399,3:45:23.359
what was the clinical story to get here and who 
can benefit from lenacapavir? And it sounds like,

3:45:23.359,3:45:30.239
the people who've called this a miracle drug, to 
me, are basically right. In the clinical trial for

3:45:30.239,3:45:38.800
cisgender women, there were zero HIV infections 
among the 2,000 women who got this injection, and

3:45:38.800,3:45:46.880
there were tens of infections in the other arms, 
for women on other forms of PrEP; so this is a

3:45:46.880,3:45:54.479
totally different situation. And for men who have 
sex with men, and other trans people who are in

3:45:54.479,3:46:02.319
the other trial, the other phase three — there was 
also a massive drop of, say, 80, 90 more percent

3:46:02.319,3:46:11.040
in transmission. So, how do we get this drug 
to people, Saloni? That's what I want to know.

3:46:11.040,3:46:18.880
Maybe, also, just to think back to the whole 
timeline of drug development in this field. In

3:46:18.880,3:46:24.640
the 1980s, in the early 1980s, when the first 
case was reported — no drugs; people thought

3:46:24.640,3:46:31.199
this was an untreatable disease, or they wanted 
to treat it, but they had no idea how. In 1987,

3:46:31.199,3:46:39.519
the first HIV drug, azidothymidine. '95, the first 
protease inhibitor, and the start of combination

3:46:39.520,3:46:46.640
therapy that completely changed the survival 
for people with HIV. In 2012, is that right?

3:46:46.640,3:46:54.640
We have PrEP — truvada — introduced, and then, in 
2022, we have lenacapavir, as this drug-resistant

3:46:54.640,3:47:03.760
treatment. Then, now, we have lenacapavir as 
a preventive drug, that is so long-lasting,

3:47:03.760,3:47:10.080
and both a breakthrough in terms of the 
effectiveness, in terms of how you take it,

3:47:10.080,3:47:15.600
how long it lasts in the body, but also, 
because it was a completely new type of

3:47:15.600,3:47:25.279
treatment. It works- it inhibits the capsid of the 
HIV virus; it's not just tweaking existing drugs,

3:47:25.279,3:47:30.239
it's this whole new type of treatment 
that now opens up the field of research

3:47:30.239,3:47:39.199
to developing more capsid drugs, I think, as 
well as more long-lasting drugs in the body.

3:47:39.199,3:47:42.160
All of that building on each other to get us to

3:47:42.160,3:47:46.880
this moment. The decades of 
science. I feel so grateful.

3:47:46.880,3:47:55.359
Decades of science. Let's talk about where we are 
now, in terms of, how are we going to scale this?

3:47:55.359,3:48:01.905
I mean, not us, specifically, but how are people 
going to scale up this drug? — you and me, back

3:48:01.905,3:48:07.040
of the van — getting them to everyone who- Like, 
just driving this van around in small villages.

3:48:07.040,3:48:10.720
I'm ready. Road trip?

3:48:10.720,3:48:20.560
So how is this going to be rolled out to 
people who need it, across the world? I think,

3:48:20.560,3:48:26.239
the most important continent here is Africa, 
and Southern Africa. Maybe we should talk a

3:48:26.239,3:48:33.120
little bit about: What has the situation 
been like until now? How does that process

3:48:33.120,3:48:39.279
work? How are people getting treatments 
across Africa, and how has that happened?

3:48:39.279,3:48:40.479
I love it.

3:48:40.479,3:48:48.719
Let's talk about HIV treatment and prevention 
around the world, how that's worked so far.

3:48:48.720,3:48:56.720
Where we are now, where we could go from 
here. I didn't know, until a few years ago,

3:48:56.720,3:49:06.960
how big the HIV treatment and prevention programs 
were worldwide. The biggest progra is PEPFAR, the

3:49:06.960,3:49:14.960
President's Emergency Plan for AIDS Relief, which 
was launched in 2003 by the Bush administration.

3:49:14.960,3:49:21.439
That was, at that point, the largest ever 
US global health initiative for a single

3:49:21.439,3:49:31.599
disease. It was 15 billion, as a commitment 
over five years, to fight HIV and AIDS in

3:49:31.600,3:49:38.319
affected countries — mostly in Southern Africa, 
but also other countries. I remember reading

3:49:38.319,3:49:45.840
about how this was formed, how the whole program 
came together, and it was super interesting and

3:49:45.840,3:49:52.960
inspiring — this idea that you could actually 
set up this huge program to treat millions of

3:49:52.960,3:50:02.800
people in the poorest parts of the world against 
this really deadly, scary disease. At that point,

3:50:02.800,3:50:08.720
there were effective combination drugs available, 
to people in the US and other richer countries,

3:50:08.720,3:50:15.040
but people in Africa were not, you 
know, they weren't able to access them,

3:50:16.560,3:50:22.960
which is quite scary. You have such a hugely 
unequal outcomes just based on where you live,

3:50:22.960,3:50:29.520
but also there is this drug that feels just 
out of reach that wasn't getting to people.

3:50:29.520,3:50:38.800
And what I read was that Bush wanted to do 
something big on HIV and AIDS and he asked

3:50:38.800,3:50:47.680
several people working with him on health in 
the US including Anthony Fauci and Dr. Mark

3:50:47.680,3:50:56.880
Dybul to figure out what was possible, and they 
looked at what was already being done in Africa.

3:50:57.840,3:51:02.720
Was there anyone who was receiving treatment 
at this time? How were they getting it? And

3:51:02.720,3:51:09.120
the main source that they found at the time was 
TASO, The AIDS Support Organisation and Doctors

3:51:09.120,3:51:16.640
Without Borders or Medicins Sans Frontieres — 
and they had been providing treatments to people,

3:51:18.479,3:51:24.479
I think it was in one- two countries, South 
Africa and Malawi. They were providing these

3:51:24.479,3:51:34.080
generic versions of antiretrovirals to them, on a 
voluntary basis. But what really stuck in my mind

3:51:34.080,3:51:39.439
was, when I was reading this interview of Mark 
Dybul — one of the people who worked on setting

3:51:39.439,3:51:46.160
up this program and planning it out — was that 
he mentioned that, at that point, TASO, the AIDS

3:51:46.160,3:51:52.639
support organisation, were actually transporting 
this in little, I think, fridges on their bags-

3:51:52.640,3:51:53.140
Wow.

3:51:54.239,3:51:59.920
-that they were carrying around on 
motorbikes around to remote villages,

3:51:59.920,3:52:07.760
to get these drugs to people who needed them. 
That was very inspiring. But the success of

3:52:07.760,3:52:14.239
Doctors Without Borders, in small scale- 
in providing treatment at a small scale,

3:52:14.239,3:52:19.279
showed that this was possible. Could it be scaled-

3:52:19.279,3:52:20.800
It's a proof of concept?

3:52:20.800,3:52:26.000
It's a proof of concept. So now what's needed is 
to set up these supply chains to do this at a much

3:52:26.000,3:52:34.239
bigger scale — to set up the drug development and 
manufacturing; set up the networks of clinics,

3:52:34.239,3:52:40.000
and the people who would be providing 
treatment to people in remote villages,

3:52:40.000,3:52:49.520
and so on. So Mark Dybul and Anthony Fauci put 
together these plans of: how much this would cost,

3:52:49.520,3:52:58.160
what it would look like to operate this, how it 
might look like at scale. The other reason this

3:52:58.160,3:53:04.479
is really interesting is because, at the time, 
people didn't think it was possible to do this.

3:53:04.479,3:53:11.279
They thought this was just some pipe dream. 
Several reasons: one is this is a really poor

3:53:11.279,3:53:18.000
region. Trying to set up something like this 
at scale requires a lot of work; you have to

3:53:18.000,3:53:24.960
work with community leaders, you need to hire 
people, train people to provide this treatment.

3:53:24.960,3:53:31.279
I think there was also this perception, 
that some people had, was that poor people

3:53:31.279,3:53:39.519
in Africa couldn't take daily pills; they 
couldn't follow these regimens. And I mean,

3:53:40.080,3:53:45.600
the fact that this program is so effective 
has, I think, has shown that that's not the

3:53:45.600,3:53:52.800
case. But it's also this idea that, just because 
something is difficult to take regularly doesn't

3:53:52.800,3:53:58.560
necessarily mean that we should stop there, and 
accept that as status quo. You could eventually

3:53:58.560,3:54:04.560
develop long-lasting drugs, you could find some 
way to make it easier for people to access these

3:54:04.560,3:54:11.360
treatments on a regular basis. And I think that 
kind of attitude shift is really important here.

3:54:14.560,3:54:18.720
You can do big things and sometimes they 
work, and at the turn of the century,

3:54:18.720,3:54:27.279
I feel like there was more optimism around big 
global health improvements and projects. And at

3:54:27.279,3:54:33.519
the same time as PEPFAR was getting started, or a 
similar time, the Global Fund was getting started,

3:54:33.520,3:54:40.800
which was not just a US program, but 
was a multilateral program that involved

3:54:40.800,3:54:47.279
many different donor countries. So higher income 
countries contributing into a pooled fund,

3:54:47.279,3:54:52.639
which would focus on HIV/AIDS, and 
malaria, and tuberculosis. Three

3:54:52.640,3:54:58.319
of the biggest infectious disease killers 
around the world, at the time and still now.

3:54:58.319,3:55:06.960
And since then, we have as a species, as a global 
society, made a huge amount of progress on all

3:55:06.960,3:55:16.479
three of those diseases. We should probably show a 
graph of HIV/AIDS incidence, but also AIDS deaths,

3:55:16.479,3:55:23.199
because the curve really bent down with these 
commitments from the Global Fund and from PEPFAR.

3:55:23.199,3:55:29.760
It was a problem that seemed to be spiralling 
out of control, and then lots of energy, focus,

3:55:29.760,3:55:36.319
attention, resources were put into it, on 
really scaling up proofs-of-concepts, and

3:55:36.319,3:55:42.319
really committing to it. Lo and behold, there were 
results. And that's inspiring to look back on,

3:55:42.319,3:55:47.359
and I wonder what it would've been like to be in 
the room when people thought it was impossible,

3:55:47.359,3:55:52.880
and you really thought, like Mark Dybul 
and others, but we gotta go for it.

3:55:52.880,3:56:02.720
It's also incredible to think about, from the 
perspective of people in Africa, how common HIV

3:56:02.720,3:56:09.920
was, at the time, is probably not obvious to 
some of us. But, in the year 2000, there were

3:56:09.920,3:56:20.479
several countries in Southern Africa where, some 
twenty, 15 to 30% of the adult population had HIV.

3:56:20.479,3:56:29.599
That is scary to think about, for such a deadly 
disease: how it affects the people themselves,

3:56:29.600,3:56:38.960
their families, the society as a whole. This was 
really effective and successful, both at changing

3:56:38.960,3:56:48.319
what life was like for people with HIV, but just 
also the culture around it, and it's continued

3:56:48.319,3:56:55.840
since 2003, so it's just an incredible program. 
It's one of the biggest global health programs.

3:56:55.840,3:57:03.840
But at the same time, it costs a very small 
fraction of our incomes here, or in the US,

3:57:03.840,3:57:11.439
to contribute to PEPFAR or the Global Fund, and it 
makes a massive impact on people around the world.

3:57:11.439,3:57:17.839
Well, it reminds me of the graph that 
you showed earlier in this episode,

3:57:17.840,3:57:23.040
when combination treatment first came out 
in the nineties, and you saw this totally

3:57:23.040,3:57:34.399
discontinuous drop in mortality rates. This was 
scaling that drop up to people who did not have

3:57:34.399,3:57:41.120
access to the drugs, until there was a global 
commitment behind them. That means that there

3:57:41.120,3:57:47.920
are many people — not just millions, but tens 
of millions of people — who are alive right now,

3:57:47.920,3:57:55.840
who are on drugs that control their infection, 
who would not have been alive. And it's so heady,

3:57:55.840,3:58:04.319
it's impossible to, at least for me, to 
get my head around that, but friends, so

3:58:04.319,3:58:13.199
many families. So it's mind blowing to think how 
different the world would be for so many people.

3:58:13.199,3:58:22.479
The scale of this is also incredible to me. The 
estimates are that there've been 25 million people

3:58:24.479,3:58:31.599
whose early deaths were prevented because of 
PEPFAR as a program. 25 million is such a,

3:58:32.560,3:58:37.920
in terms of the number of lives saved, it's 
just so huge to think about, it's like London's

3:58:37.920,3:58:47.920
population is what, 11, 12 million? That's two, 
more than two of the entire... Wow. It's just,

3:58:47.920,3:58:56.160
imagine that not existing; all those people 
not being alive. It's just a huge impact.

3:58:56.160,3:58:58.479
20 San Franciscos.

3:58:58.479,3:59:01.519
20 San Franciscos.

3:59:01.520,3:59:07.439
So what's happening now, then? It's 
all good news, by the sounds of things.

3:59:07.439,3:59:15.439
It's not good news, sadly. It's April when 
we're recording this and there's- in the

3:59:15.439,3:59:21.839
last few months, the picture around PEPFAR 
and various other global health programs has

3:59:21.840,3:59:30.720
completely changed. At the end of January, 
there was a foreign aid spending freeze. So,

3:59:30.720,3:59:37.199
all funding to various global health, 
humanitarian programs was frozen. Also,

3:59:37.199,3:59:47.120
staff who were working at the US aid agency, 
USAID, — some thousands of them were laid off.

3:59:47.120,3:59:52.880
The remaining ones were asked not to talk to 
the public, but also, they weren't able to keep

3:59:52.880,3:59:58.960
in contact with the programs in the field; 
they were just banned from communicating.

3:59:59.920,4:00:06.160
There was also this — because of all the layoffs — 
that also meant running the programs on the field

4:00:06.160,4:00:15.680
was difficult, for people who were working 
overseas. This freeze was meant to be — this

4:00:15.680,4:00:23.439
is the stated intention — which is that it was 
for a three-month-long review of these programs,

4:00:23.439,4:00:30.319
of all foreign aid programs, to see if they 
aligned with the Trump administration's interests,

4:00:30.319,4:00:36.719
such as national security and things like that.
But a lot of them were, I mean, this just froze

4:00:36.720,4:00:43.279
all of them at once, rather than doing a review at 
the same time while they were ongoing. But also,

4:00:43.279,4:00:50.080
even after this freeze was ended, which 
was after- I think it was 39 or so,

4:00:50.080,4:01:01.279
35 business days, instead of three months, and 
some 80 to 90% of programs were just cancelled.

4:01:01.279,4:01:08.399
Because programs refers to specific recipients 
of funds, I think for specific purposes, that

4:01:08.399,4:01:18.319
doesn't mean that that PEPFAR was cut by 90% or 
so. For HIV and AIDS, I think it was around 23%,

4:01:18.319,4:01:24.239
if you roughly estimate based on the amount of 
funding those specific council programs received,

4:01:24.239,4:01:33.199
that were cut. And that seems like maybe 
just a fraction of this worst-case scenario,

4:01:34.000,4:01:39.359
of the whole thing being paused, but I think 
because so many thousands of staff were laid

4:01:39.359,4:01:44.960
off — and because of this uncertainty 
and this freeze — that actually had,

4:01:44.960,4:01:51.520
from what I can understand, quite a large impact.
One reason for that is: clinics, or programs,

4:01:51.520,4:01:56.960
receiving the funding, some of them weren't 
able to survive more than a few months without

4:01:56.960,4:02:01.680
continued funding. Some of the funding was 
for stuff that they had already completed,

4:02:01.680,4:02:07.279
and they were just waiting to receive the payment 
for it, and that was cancelled as well. Some of

4:02:07.279,4:02:13.920
them were for continued work, and this was 
cancelled. If clinics are not able to survive

4:02:13.920,4:02:18.960
for more than a few weeks, or a few months, 
without this funding that they're expecting, they

4:02:18.960,4:02:28.880
might just shut down. So in the field, what people 
might have seen would be a physical clinic that's

4:02:28.880,4:02:34.160
there — sometimes the treatment is actually inside 
the clinic — but they can't access it, because the

4:02:34.160,4:02:42.239
clinic is shut down, and/or the staff is just not 
around. There aren't people working the clinics.

4:02:42.239,4:02:50.399
And thirdly, trying to get new supplies to these 
clinics was also disrupted, so they weren't able

4:02:50.399,4:02:56.559
to restock on a lot of the treatments.
At the point where we are now, I think

4:02:56.560,4:03:03.920
there's been a lot of disruption, but there's 
a rough plan to merge all of these foreign aid

4:03:03.920,4:03:10.560
programs under USAID into the State Department 
of the US. I think this is still a little bit

4:03:10.560,4:03:16.560
up in the air, in terms of how that will 
actually work, how will it affect PEPFAR?

4:03:16.560,4:03:24.239
One thing I'm slightly worried about is 
what will they actually cut? What aligns

4:03:24.239,4:03:29.199
with the Trump administration's interests, 
and what doesn't? And I think you can get

4:03:29.199,4:03:37.120
a little bit of a clue from how they responded 
when this funding aid freeze began in January.

4:03:37.120,4:03:42.319
They initially said that there were certain 
parts of the program that would continue,

4:03:42.880,4:03:48.960
such as treatments for pregnant women. But 
from what I understand, that actually didn't

4:03:48.960,4:03:55.120
happen in practice, because of all of these 
layoffs, and the funding cuts, and so on.

4:03:55.120,4:04:02.880
But what I'm worried about is, what about 
prevention as a whole? We think about the critical

4:04:02.880,4:04:10.640
part of PEPFAR as being preventing mother-to-child 
transmission, but it's also the broader thing of:

4:04:10.640,4:04:17.600
how do we reduce the spread of HIV? How do we 
treat everyone who has HIV? This began as such

4:04:17.600,4:04:29.920
an ambitious and effective program that managed 
to treat some 20.5 million people last year,

4:04:29.920,4:04:37.359
and now that's massively been frozen and we 
don't know how much of it's going to remain.

4:04:37.359,4:04:45.839
It's so frustrating. It's so frustrating 
to see such a good use of money that,

4:04:45.840,4:04:52.160
in the grand scheme of things, is not that much 
money from the US government's perspective.

4:04:52.160,4:04:55.840
I mean, it's frustrating in the short term. I'm 
more nervous about treatment than prevention;

4:04:55.840,4:05:04.160
of what you're describing of clinics being closed. 
If you have HIV, you need to be on daily drugs,

4:05:04.160,4:05:12.559
and if you run out and can't get a refill, oh God, 
it must just be so scary. It must be so scary. And

4:05:12.560,4:05:19.840
then I totally agree. We're at this present, we 
at this wonderful moment of prevention and driving

4:05:19.840,4:05:26.640
down transmission is just becoming more and more 
possible with lenacapavir and with other drugs.

4:05:27.760,4:05:34.640
It's... I find it really heartbreaking because, 
until this point, until this year, I was so

4:05:34.640,4:05:41.520
excited about how would lenacapavir or other 
long-acting treatments change the picture? Like,

4:05:41.520,4:05:46.880
would we be able to effectively eliminate 
the transmission of HIV, in some of these

4:05:46.880,4:05:52.080
countries? And I think that's possible, and I 
think that's a little bit ambitious, just like

4:05:52.080,4:05:58.000
PEPFAR is. I think it's possible. But instead 
of going ahead, and kind of going big on this,

4:05:58.000,4:06:04.560
trying to really cut it down, and actually, then, 
not needing such a large program because you've

4:06:04.560,4:06:10.880
managed to reduce the number of people who 
are affected by this. Instead of doing that,

4:06:10.880,4:06:18.800
were on this road of a lot of uncertainty and 
disruption, and that it's come so suddenly that

4:06:18.800,4:06:26.479
people couldn't plan for it. They didn't expect 
that their treatment would suddenly disappear. I

4:06:26.479,4:06:33.839
think what the figures were showing, that I was 
reading, was- the program, PEPFAR is so big,

4:06:33.840,4:06:42.000
there's 20 and a half million people receiving 
treatment from PEPFAR per year. On a daily level,

4:06:42.000,4:06:50.800
that's 200,000 people who are getting their 
refills; 200,000 people who are realising

4:06:50.800,4:06:58.080
the clinics are shut, and they don't know 
how they're going to get their next supply.

4:06:58.080,4:07:05.519
Well, and that's 200,000 cases of higher 
risk of mutated viruses as well, presumably.

4:07:05.520,4:07:13.920
There's a risk of the resurgence of HIV, if 
people stop taking treatment for, I think,

4:07:13.920,4:07:20.319
it's a few months or so, and that has some 
quite nasty side effects in the initial period;

4:07:20.319,4:07:28.799
and then there's also of the complications that 
people would have with HIV as a disease itself.

4:07:28.800,4:07:33.600
Okay, so to summarise: the launch of 
PEPFAR, the launch of the Global Fund

4:07:33.600,4:07:40.399
were a particularly ambitious moment, and 
there were people around who thought maybe

4:07:40.399,4:07:43.679
they would not achieve their aims, 
because it never been done before.

4:07:43.680,4:07:49.359
And lo and behold, they achieved great 
things for tens of millions of people.

4:07:49.359,4:07:51.759
For over two decades.

4:07:51.760,4:07:59.439
Over two decades. And now we're not only, 
we've been talking, in this episode,

4:07:59.439,4:08:04.479
about great technical achievement. That could be a 
launch of a new ambitious program and really drive

4:08:04.479,4:08:10.959
down transmission, but really, we're at a state 
where, not only are the ambition levels lower,

4:08:10.960,4:08:20.560
but the basics are not, as we record, 
being provided for everyone who needs them.

4:08:20.560,4:08:25.920
Right. It brings me to this point that I 
often think about. So when I write this

4:08:25.920,4:08:33.120
Substack newsletter on medical innovation, and 
what is the use of new treatments if there's

4:08:33.120,4:08:41.840
no one to distribute them, or if there's no one 
that can access them. This is the whole point of

4:08:41.840,4:08:48.560
medical- or, these breakthroughs don't matter, if 
they're not getting to the people who need them.

4:08:48.560,4:08:54.479
Okay, so do we have any hope for the Global 
Fund? So listeners who are in the US, think about

4:08:54.479,4:09:02.719
PEPFAR. Listeners in other countries, think about 
your own health systems and think about the Global

4:09:02.720,4:09:08.640
Fund, which, I assume, is up for replenishment 
pretty soon, and is doing a lot of this work too.

4:09:08.640,4:09:15.439
Replenishment is when they get funding 
replenished from various countries.

4:09:15.439,4:09:16.000
Yes, exactly.

4:09:16.000,4:09:19.600
And those are decided by 
their foreign aid budgets.

4:09:19.600,4:09:30.160
Correct. So foreign aid in many higher-income 
and middle-income countries will contribute

4:09:30.160,4:09:37.359
some amount of tax revenues, or of government 
revenues, to foreign aid to other countries.

4:09:37.359,4:09:46.239
A good slice of that is global health; and HIV 
treatment and prevention is a reasonable portion

4:09:46.239,4:09:56.559
of that global health contribution, often via the 
Global Fund. And the UK is not in a particularly

4:09:56.560,4:10:03.600
ambitious moment either. At the time of recording, 
we are a month in, or possibly two months now,

4:10:03.600,4:10:13.040
into an announcement under the Labour government 
that, instead of returning to 0.7% of GNI,

4:10:13.040,4:10:20.800
basically GDP, being contributed to foreign aid — 
as was the case under the last Labour government

4:10:20.800,4:10:24.560
and the beginning of the last Conservative 
government — we are going to drop down

4:10:24.560,4:10:32.800
to 0.3. So under COVID, we dropped from 0.7 to 
0.5 in what was termed a temporary measure for

4:10:32.800,4:10:42.714
COVID. And now instead of returning to 0.7, we're 
dropping down to 0.3 and we're going through-

4:10:42.714,4:10:43.760
0.3%, wow.

4:10:43.760,4:10:44.319
Yes.

4:10:44.319,4:10:50.399
So small. I mean, to me it seems so small 
as well, because I have read about some of

4:10:50.399,4:10:55.120
these other programs that our foreign aid has 
funded, and some of them are really impactful.

4:10:55.840,4:11:02.160
It's stuff like — oh my gosh — it's vaccination 
of millions of children against these deadly

4:11:02.160,4:11:10.239
diseases that don't really affect us very much 
in wealthier countries. It's stuff like trachoma,

4:11:10.239,4:11:16.559
which is this bacterial eye infection. The US 
and the UK, and some other philanthropic donors,

4:11:16.560,4:11:23.439
came together to fund this really ambitious 
program to supply antibiotic treatments,

4:11:23.439,4:11:28.879
better sanitation measures, and so on, to 
hundreds of millions of people across Africa,

4:11:28.880,4:11:34.319
and they've massively reduced this... really 
painful bacterial eye infection that can lead

4:11:34.319,4:11:41.679
to blindness in children. These are some huge 
successes that many people don't even know about.

4:11:41.680,4:11:47.600
Tuberculosis — way down from 20 years ago. 
Malaria, recently stalling, but has been

4:11:47.600,4:11:56.479
driven down a lot since 2000; mostly affects 
children. Progress is possible. Okay, well,

4:11:56.479,4:12:04.080
now that we've depressed ourselves sufficiently, 
there's both the financing that is not looking

4:12:04.080,4:12:10.800
so hot right now. The ambition not looking so 
hot right now. Another thing that you can do,

4:12:10.800,4:12:19.199
to try and get more people access to treatment 
and prevention, is to drive the cost down of

4:12:19.199,4:12:26.960
the actual drugs themselves. And maybe it's time 
we talk about that with respect to lenacapavir.

4:12:26.960,4:12:32.319
What do we know about the 
cost of lenacapavir right now?

4:12:32.319,4:12:39.199
Well, there's the cost of production and then 
there's the price that the company selling the

4:12:39.199,4:12:49.679
drug charges. The price that Gilead is charging, 
in the US, is $42,250 — is the last number I saw.

4:12:49.680,4:12:50.800
Per person?

4:12:50.800,4:12:57.199
Per person, per year. So that covers all 
the injections for the first year. That

4:12:57.199,4:13:05.359
is about double what cabotegravir — the 
Viiv injectable — is priced at in the US.

4:13:05.359,4:13:12.399
I was priced out of Cabotegravir; I assume I will 
be priced out of lenacapavir in the short term.

4:13:12.399,4:13:18.879
Why do they charge that much money? Well, a lot 
of the development that we talked about earlier

4:13:18.880,4:13:26.239
in the episode is done as a pure expense. So 
all of the scientists who are working at Gilead,

4:13:26.239,4:13:31.199
trying to iterate on the drugs to make them 
better, the funding for all the, or at least

4:13:31.199,4:13:36.080
most of the, clinical trials we discussed — 
that's all done as an expense. So they want to

4:13:36.080,4:13:42.319
charge more out the other end, to some patients, 
so that they can recoup some of that money. And

4:13:42.319,4:13:48.000
if they make any profit, hopefully some of it 
will get reinvested in more drug development,

4:13:48.000,4:13:53.520
not just distributed to shareholders.
That, of course, raises questions:

4:13:53.520,4:13:58.640
well, do you think that someone in 
Botswana is going to pay $42,000?

4:13:59.600,4:14:05.920
My guess is no, and I'm not going to pay $42,000, 
so I'm with them. The price versus cost is very

4:14:05.920,4:14:11.680
different for patented medicines, often. 
There are estimates for how much it will

4:14:11.680,4:14:19.520
cost to produce a generic version of lenacapavir 
that are under a hundred dollars; I've seen as

4:14:19.520,4:14:24.479
low as $40. I haven't looked into the 
drivers of those estimates, or how much

4:14:24.479,4:14:30.959
Gilead has really revealed about the production 
methods in public — which would give you better

4:14:30.960,4:14:38.239
methods of coming up with estimates there.
The good news, though, is that Gilead has

4:14:38.239,4:14:48.319
already signed agreements with six generic 
suppliers of lenacapavir for 120 countries;

4:14:48.319,4:14:54.080
so, plenty of low-income and middle-income 
countries in that mix. And you may recall there

4:14:54.080,4:14:59.680
are about 200 countries in the world, so there's a 
lot that are not covered there, but for those 120

4:14:59.680,4:15:08.640
countries, these suppliers will be able to provide 
versions of lenacapavir that have been shown to

4:15:08.640,4:15:17.199
be therapeutically-equivalent to the initial 
drug that Gilead used, in the clinical trials.

4:15:17.199,4:15:26.319
Though these new production runs will be made 
by different companies, that have been given the

4:15:26.319,4:15:36.080
rights and taught a bit by Gilead how to do it. 
Those generic companies will be manufacturing,

4:15:36.080,4:15:42.559
hopefully, eventually, enough supply for 
use in those 120 countries. What Gilead,

4:15:42.560,4:15:49.760
in their press release in October, has 
committed to is that they plan to provide

4:15:49.760,4:15:57.040
Gilead-supplied product at no profit to Gilead, 
until generic manufacturers are able to fully

4:15:57.040,4:16:05.279
support demand in high-incidence, resource-limited 
countries. So that's a great start, to be honest.

4:16:05.279,4:16:15.920
The questions that leaves me are, well, how 
quickly- what is the supply? How quickly could

4:16:15.920,4:16:21.199
demand be met? And, are we sure that they're going 
to ramp up quickly enough? I mean, number one. But

4:16:21.199,4:16:27.840
then, secondly, what about the other 80 countries? 
So there are plenty of countries in South America,

4:16:27.840,4:16:35.760
for example, that are not in the 120 covered by 
this generic agreement, and that have relatively

4:16:35.760,4:16:44.880
have medium HIV incidence — where a lot of people 
could be protected by this drug that now exists.

4:16:44.880,4:16:49.920
Some of them were part of the clinical 
trials for lenacapavir as well,

4:16:50.479,4:16:57.519
which is quite depressing. But, I think, what 
I've read is that they plan to provide it

4:16:58.399,4:17:05.120
to people in the trial. I'm not sure that 
that extends to the whole country. I mean,

4:17:05.120,4:17:09.840
could someone just make this drug themselves? 
Or, I mean, not like an individual.

4:17:15.840,4:17:21.120
There's something beyond that, which are: 
global intellectual property, and patents,

4:17:21.120,4:17:25.680
and enforcing those; then, there may 
also be some technical barriers. So

4:17:25.680,4:17:30.560
I'll talk about the patents and then the 
technical barriers. On the patents front,

4:17:30.560,4:17:38.960
there's a long history — with HIV specifically — 
of the tension of global pharmaceutical companies,

4:17:38.960,4:17:46.640
who want to enforce patents and high prices, and 
patients and activists and advocates, who want

4:17:46.640,4:17:54.319
medicines to be available for more people sooner.
And there's two broad solutions to this problem.

4:17:54.319,4:18:01.120
One is, what I just described, with Gilead, 
which is voluntary licensing, where Gilead

4:18:02.319,4:18:08.000
arranges with other generic companies: 'Okay, 
you can do this and we will not sue you for

4:18:08.000,4:18:13.760
selling these drugs in these 120 countries. But 
if you sell them elsewhere, maybe we will sue

4:18:13.760,4:18:20.159
you.' That's voluntary [licensing].
Then there's compulsory licensing,

4:18:20.159,4:18:29.199
where a country may determine that they have a 
public health crisis, to the extent that they

4:18:29.199,4:18:35.519
are not going to- the normal patent rules are out 
the window. So this almost happened with HIV in

4:18:35.520,4:18:41.359
South Africa in the late '90s, early 2000s, 
and more recently, I believe, in Colombia,

4:18:41.359,4:18:48.319
where the government's like, 'Make drugs, that's 
okay, so long as people get the drugs.' I think

4:18:48.319,4:18:55.599
that is appropriate in some medical emergencies.
It was a big topic of dispute and debate in COVID

4:18:55.600,4:19:00.560
as well, where I think the debate actually 
goes a bit of a different direction — probably

4:19:00.560,4:19:08.319
not worth getting into now, but vaccine 
manufacturing is quite different to generic

4:19:08.319,4:19:14.799
small molecule manufacturing. Small molecules are, 
in general, pretty commoditized: there are many,

4:19:14.800,4:19:19.600
many companies who can make them in 
many, many countries. And vaccines,

4:19:19.600,4:19:25.600
the product is the process to some degree — 
how you actually manufacture a given vaccine,

4:19:25.600,4:19:30.560
in a particular bioreactor, with particular cells, 
with particular growth medium, you got to kind

4:19:30.560,4:19:35.120
of get taught by the original manufacturer. 
It's harder to just scale up, and it's harder

4:19:35.120,4:19:41.760
to even infringe on a patent, if you wanted to.
That brings me to the technical blockades here,

4:19:41.760,4:19:49.199
where, for a traditional generic drug, a small 
molecule, there are not many technical blockades;

4:19:49.199,4:19:55.199
you can just- even if a company has not revealed 
all the secrets of how they made something,

4:19:55.199,4:20:00.960
they have to file with a regulator, and give some 
information — some of which is then made public,

4:20:00.960,4:20:08.479
once a drug approval is given. They have to list 
certain information on a label for patients,

4:20:08.479,4:20:12.799
of what the heck is in this drug. 
And also, if you're a competitor,

4:20:12.800,4:20:18.399
you can simply buy their drug once it's on the 
market, and analyse what's in it. So you have

4:20:18.399,4:20:27.439
a lot of tools where you can basically enter as a 
competitor, from a technical point of view. Then,

4:20:27.439,4:20:36.479
the difficulty with some long-acting drugs 
is that they are more complicated to copy. If

4:20:36.479,4:20:44.799
you were dealing with a long-acting drug that 
had a liposome, or had a particular polymer-

4:20:44.800,4:20:46.159
What's a liposome?

4:20:46.159,4:20:50.159
You know, I don't want to answer that 
question because I'll get it wrong.

4:20:50.159,4:20:51.760
It's a fatty blob, right?

4:20:51.760,4:20:52.800
Say it again?

4:20:52.800,4:20:55.323
It's a fatty blob.

4:20:55.323,4:20:55.359
Exactly, thank you!
"Lipo" means fat.

4:20:55.359,4:20:59.199
It's a fatty blob, I think it's probably 
a bilayer. I think, basically imagine a

4:20:59.199,4:21:04.239
fatty blob that encapsulates the thing you 
care about, but there are different fatty

4:21:04.239,4:21:08.800
blobs you might want to make, and there are 
companies trying to improve their fatty blobs,

4:21:08.800,4:21:13.840
and it's harder to copy the fatty blobs that are 
right at the frontier of fatty blob technology

4:21:13.840,4:21:18.720
than it is to copy the small molecule.
Now, the good news about lenacapavir,

4:21:18.720,4:21:24.159
and the good news about cabotegravir, and the good 
news about islatravir — three of the long-lasting

4:21:24.159,4:21:30.239
drugs we've talked about for HIV — is that they 
don't seem to be right at the hardest end. We

4:21:30.239,4:21:36.080
don't have a liposome, for example, involved. 
We do, potentially, have some things that make

4:21:36.080,4:21:43.840
it a little harder than usual and — I'm a little 
bit beyond my knowledge about how it applies to

4:21:43.840,4:21:51.199
lenacapavir, and I would love to see how it goes — 
but, for example, I think you need a nano miller,

4:21:51.199,4:21:58.000
where you grind up your drugs! This is true for 
Cabotegravir and, I think, probably not true for

4:21:58.000,4:22:04.640
lenacapavir. You grind up your drug crystal, so 
they're tiny, tiny, tiny, tiny, so that when you

4:22:04.640,4:22:10.880
disperse them in a liquid, and then when you 
inject that liquid with the solids, you've got

4:22:10.880,4:22:20.319
solids that are high-surface-area-to-volume-ratio. 
So the question that comes to mind for me is:

4:22:20.319,4:22:26.799
Which generic companies own a nano miller, for 
example, and does that machine cost $8 million,

4:22:26.800,4:22:32.319
you know, how much? Those questions start rearing 
their head, and the next set of questions for me

4:22:32.319,4:22:41.359
is... there's a tried-and-true regulatory pathway 
at the FDA and other regulators, for generic

4:22:41.359,4:22:48.719
equivalents to small molecules: you have to prove 
only that they are equivalent in certain respects;

4:22:48.720,4:22:56.720
you do not have to redo everything else.
Are we sure that those tests are going to

4:22:56.720,4:23:03.279
be good enough, for long-acting injectables, or 
for long-acting drugs in general? Or is there

4:23:03.279,4:23:08.880
some other reason why, towards the tail end 
of many months, there may be more deviation

4:23:08.880,4:23:13.520
than you got from just testing the batch 
chemically, and well- should we actually,

4:23:13.520,4:23:17.600
therefore, rerun a big clinical trial? 
And that would be cost-prohibitive;

4:23:17.600,4:23:22.319
then you really would not see much generic entry. 
I don't think that that's the way lenacapavir

4:23:22.319,4:23:29.759
will go. It certainly would get my hair up 
if people started worrying about it though.

4:23:29.760,4:23:36.159
I mean, I guess this also makes me think about 
Gilead producing it, probably, with multiple

4:23:36.159,4:23:42.720
manufacturing plants or so, and they would have to 
do this internal testing, presumably. Hopefully,

4:23:42.720,4:23:46.720
there's a way to do that in the same way — or 
they're producing it in a very similar way,

4:23:46.720,4:23:52.239
that they would be able to know early on, is this 
the equivalent; are these molecules equivalent

4:23:52.239,4:24:00.479
across these different sites? The other question 
that I had was, I had heard of this thing called

4:24:00.479,4:24:05.839
the Medicines Patent Pool? Is that something- 
have we already covered that, or what is that?

4:24:05.840,4:24:12.319
Well, conceptually, a little bit, but actually 
no. They are kind of an intermediary. They're

4:24:12.319,4:24:19.920
a UN-backed non-profit that tries to help match 
up originator companies — that are filing patents

4:24:19.920,4:24:24.399
on new medicines, taking those medicines 
through clinical trials, marketing them

4:24:24.399,4:24:33.279
in some countries — trying to match them up with 
generic companies, who might want to market the

4:24:33.279,4:24:40.559
drug in countries the originator company doesn't 
focus on as much, or doesn't care about as much,

4:24:40.560,4:24:47.920
in terms of making a profit, for example. 
Often, there's a match to be made there,

4:24:47.920,4:24:51.600
that both companies are very happy with.
You know, especially if you're a smaller

4:24:51.600,4:24:58.239
originator company, who doesn't have any 
experience selling into Bangladesh, then

4:24:58.239,4:25:03.920
you might very well want your drug to be used by 
people in Bangladesh who want your drug... but you

4:25:03.920,4:25:09.359
just don't have the resources to get up to speed 
with the drug regulator in Bangladesh. You're

4:25:09.359,4:25:16.000
going to be on the hook if anyone sues you for 
side effects in Bangladesh; it's a big proposal.

4:25:16.000,4:25:21.760
But the Medicine Patents Pool, MPP, will sit in 
the middle and say, 'Look, we have relationships

4:25:21.760,4:25:27.760
with many generic manufacturers, many of whom 
have lots of experience in Bangladesh. If you just

4:25:28.479,4:25:33.359
sign on the dotted line here and say, you're not, 
basically, you're not going to sue them if they

4:25:33.359,4:25:40.719
sell in Bangladesh, then you can both be happy 
and patients get to benefit.' And they've had some

4:25:40.720,4:25:48.399
successes, in particular with HIV over the years. 
I think they were set up in 2010 so, for the more

4:25:48.399,4:25:55.439
recent round of voluntary licensing. But they 
worked on Cabotegravir, the initial long-lasting

4:25:55.439,4:26:02.719
injectable drug, and sort of sat between 
Viiv, the originator company, and, I believe,

4:26:02.720,4:26:08.960
three generic companies there, to help transition 
that- help get that out into more countries.

4:26:08.960,4:26:15.120
I didn't know much of that at all. That's 
really cool. I'm kind of thinking about,

4:26:15.120,4:26:20.800
okay, we heard a little bit about how PEPFAR 
was formed and that was set up, and yeah,

4:26:20.800,4:26:27.040
I'd very curious about that. The other thing 
I was thinking was, okay, given that we have

4:26:27.040,4:26:34.560
all of this — we have the Medicines Patent Pool, 
we have these licenses with generic manufacturers.

4:26:34.560,4:26:41.680
Also, Gilead, I think, they've said that 
they have the capacity to manufacture upto

4:26:41.680,4:26:49.199
10 million doses this year. I initially thought 
that was a big number, but if you think about it,

4:26:49.199,4:26:55.920
it's two doses per six months, so divide 
by four. Divide 10 million by four. That's

4:26:55.920,4:27:04.640
roughly 2.5 million people who would get this. 
And it's both a treatment for drug-resistant HIV,

4:27:04.640,4:27:11.920
and it's going to be used for prevention, 
so that's actually a small fraction. So

4:27:11.920,4:27:17.279
hopefully the generic manufacturers 
help to scale that up to some degree.

4:27:17.279,4:27:22.559
But aside from that, okay, if PEPFAR has 
this uncertain future right now, I think

4:27:22.560,4:27:29.920
we do know that the Global Fund is going to try 
and roll it out, but trying to scale that up is,

4:27:29.920,4:27:36.319
I think, maybe the next big thing to try to 
focus on — if people are listening and have

4:27:36.319,4:27:42.159
some way to convince their government that 
this is a really important thing to work on.

4:27:42.159,4:27:48.159
I think the Global Fund seems like it's going to 
be doing lots of the work, in terms of funding

4:27:48.159,4:27:55.760
these programs to roll it out and administering 
it worldwide. But I'm curious if there are

4:27:55.760,4:28:01.760
other things that come to mind, in terms of how 
you're thinking about the potential future here.

4:28:01.760,4:28:09.199
Those are the main ones to me. I think the time is 
now and the opportunity is here, and if you really

4:28:09.199,4:28:20.239
just take a step back, a whole 'nother level — 
what is happening in many countries most affected

4:28:20.239,4:28:33.359
by HIV? Well, a happy piece of news is that- 
I am cautious saying this in April 2025, where

4:28:33.359,4:28:38.239
trade relations are also a little bit up in the 
air, but the happy news from the last few decades

4:28:38.239,4:28:47.760
is that most lower- and middle-income countries 
have been growing economically, and that was not

4:28:47.760,4:28:57.279
true in the 1960s, for example. We, in fact, in 
most countries, have been seeing income growth for

4:28:57.279,4:29:04.559
people, and seeing the tax base of those countries 
also grow — meaning those countries can do more

4:29:04.560,4:29:15.279
public health interventions, too. And you know, 
that is really the future that will make a lot of

4:29:15.279,4:29:21.920
people with HIV have more sustainable healthcare- 
is that if you are in Nigeria, you should not have

4:29:21.920,4:29:27.920
to rely on the whims of the American public, 
and in the future you hopefully will not.

4:29:27.920,4:29:34.319
But that, unfortunately, is decades away. The 
tax base of many countries is not high enough

4:29:34.319,4:29:42.399
to provide lenacapavir to people who need it, and 
people are not individually rich enough to pay for

4:29:42.399,4:29:51.759
lenacapavir, if they need it. So now is a moment 
for people, especially voters, in richer countries

4:29:51.760,4:30:01.199
to kind of do our part. It's not a permanent 
humanitarian effort, I think it's a decadal- the

4:30:01.199,4:30:07.840
next few decades really matter. And we have these 
amazing new technologies that actually enable a

4:30:07.840,4:30:15.680
dent to be made. That would be where I would leave 
the topic of scale up is... now's the moment,

4:30:15.680,4:30:23.920
and I hope, I hope, that when you and I are 30 
years older, this kind of topic doesn't exist

4:30:23.920,4:30:34.479
in the same way, because we don't have to think 
about external sources of financing quite as much.

4:30:34.479,4:30:40.159
No, I really hope so as well. I mean, I 
think, right now, where we are, this still

4:30:40.159,4:30:46.080
seems quite out of reach for a lot of people in 
the most-affected countries, in Southern Africa.

4:30:46.080,4:30:54.479
I think I was reading... the average spending 
on healthcare some $80 per person per year in

4:30:54.479,4:31:02.399
Southern Africa, whereas it's, what is it, like 80 
times that or something in the US? And meanwhile,

4:31:02.399,4:31:08.960
I mean, that's not just for lenacapavir, which 
would be some $40 or so at a generic price,

4:31:08.960,4:31:16.000
it's also all the other treatments, 
preventions, testing, and so on. Right now,

4:31:16.000,4:31:25.199
it still feels quite out of reach, and I'm just- 
The Global Fund is going to be really important;

4:31:25.199,4:31:34.880
trying to keep PEPFAR running; but also trying 
to build up more capacity within these countries

4:31:34.880,4:31:43.680
to protect the people who are affected by HIV 
right now. That seems like a difficult problem.

4:31:43.680,4:31:50.319
It sounds like there's some uncertainty in the 
next decade then. So are there other tools that

4:31:50.319,4:31:58.080
are maybe not scientific, but more economic or 
financial, that we can apply to be more ambitious?

4:31:58.080,4:32:06.239
I think there are a few. I mean, it's not 
just lenacapavir. I think this is one really

4:32:06.239,4:32:11.279
important area that we want to scale up, 
but there are also these other potential

4:32:11.279,4:32:17.439
future drugs like islatravir, any other 
potential long-acting drugs that might work.

4:32:17.439,4:32:25.679
How do people get the funding model so that 
we're not having to be in the situation where

4:32:25.680,4:32:33.760
a pharmaceutical company is trying to recoup their 
costs at a very high price in richer countries,

4:32:33.760,4:32:42.080
and then hopefully, voluntarily, agreeing to these 
agreements, which may or may not be scaled up. One

4:32:42.080,4:32:50.319
idea that comes to mind for me is this idea of 
an Advance Market Commitments, or an AMC. And

4:32:50.319,4:32:59.040
I love the idea of AMCs. I've written about them 
a bunch, read a lot of stuff about them. I think,

4:32:59.040,4:33:07.199
the way to think about this is to contrast it with 
the regular approach to funding drugs or vaccines.

4:33:07.199,4:33:13.599
So we usually have this situation where a 
pharmaceutical company, or philanthropic funder,

4:33:13.600,4:33:20.160
or some government is trying to decide upfront 
which bets might work out — which companies

4:33:20.160,4:33:27.359
or which researchers might develop an effective 
drug — and they're funding those groups directly,

4:33:27.359,4:33:32.480
and then some of them will work out, some of them 
won't. In drug development, the success rate is

4:33:32.480,4:33:38.480
very low, and that means most of these bets 
are going to fail. There's going to be a lot of

4:33:38.480,4:33:47.759
wasted money on the funder's side. And secondly, 
there's this huge expense covered by individual

4:33:47.760,4:33:51.920
pharmaceutical companies, in developing 
the drug, that they now want to recoup,

4:33:51.920,4:33:59.039
so they charge these very high prices. It takes 
a while, usually, for a drug to go off-patent,

4:33:59.039,4:34:08.959
or for them to agree to these generic-licensing 
approaches, and that is a lot of time wasted;

4:34:08.959,4:34:13.520
it's a lot of people who are not getting 
the drugs or the vaccines that they need.

4:34:14.080,4:34:19.039
Is there another way to do it? And 
I think AMCs are one answer to that.

4:34:20.240,4:34:29.119
An AMC is kind of an inversion of this — 
where, instead of funding the groups directly,

4:34:29.119,4:34:36.160
you're funding the potential successful products 
at the end. So, you're setting up this pool of

4:34:36.160,4:34:43.680
funding — which might be some billions of dollars 
or so — if a company or a research group can

4:34:43.680,4:34:53.199
develop a drug or vaccine that meets certain 
standards. The amount that's given to these

4:34:53.199,4:35:00.719
companies, or manufacturers, depends on how much 
they're manufacturing. It's usually on a per-dose

4:35:00.719,4:35:07.439
basis — essentially, how many doses have you 
administered? You get more funding based on that.

4:35:07.439,4:35:13.919
I think this is a really cool idea 
for two reasons. One is, as a funder,

4:35:13.920,4:35:20.959
you don't have to know who is going to succeed. 
You have this pool of funding if something

4:35:20.959,4:35:27.439
succeeds. If nothing succeeds, you don't pay that 
money. So you are saving on that. And secondly,

4:35:28.240,4:35:34.719
you're also rewarding companies that scale up 
the drug faster and get it out to people who

4:35:34.719,4:35:42.080
need it. At the same time, you're only doing 
this for the successful drug. So you're giving

4:35:42.080,4:35:51.199
this stable potential future market to companies; 
they're going to have this commitment in advance,

4:35:51.199,4:35:55.439
often years in advance, of what the 
price is going to be, and they can plan

4:35:55.439,4:36:00.878
much more effectively based on that.
This has been tried for pneumococcal

4:36:00.879,4:36:06.799
vaccines in the past. I think this was in the late 
2000s, there was this advance market commitment

4:36:06.799,4:36:14.080
set up to try to speed up the production of 
pneumococcal vaccines. Pneumococcal disease

4:36:14.080,4:36:21.840
is a respiratory lung infection that affects 
people worldwide, and we already had effective

4:36:21.840,4:36:27.520
vaccines for it in richer countries, but in 
Africa, there were different strains of the

4:36:27.520,4:36:36.000
bacteria that weren't targeted for the vaccines.
So this AMC was set up, knowing that it was

4:36:36.000,4:36:43.199
possible to develop a vaccine for these 
other strains. There was this pool of 1.5

4:36:43.199,4:36:49.920
billion dollars that was there for companies to 
receive, depending on how much they produced,

4:36:49.920,4:36:57.600
if they managed to get a vaccine through clinical 
trials to show safety and efficacy. And it was

4:36:57.600,4:37:02.000
very successful — so the scale-up of this 
pneumococcal vaccine in African countries was

4:37:02.000,4:37:09.840
very fast. I think three or four companies managed 
to produce effective vaccines including the Serum

4:37:09.840,4:37:15.680
institute and I think Pfizer was another one of 
them. It just shows this model, of how this can

4:37:15.680,4:37:22.000
work, and you don't even to have it for- you don't 
even need to believe that it's possible to develop

4:37:22.000,4:37:28.400
a drug or vaccine for it, because if it doesn't 
work out, you don't have to pay that funding out.

4:37:28.400,4:37:35.039
The people who do have to pay are, like, 
the pharmaceutical company themselves in the

4:37:35.039,4:37:40.799
early stages — they will still have to make the 
decision on whether this is a good bet for them.

4:37:40.799,4:37:47.600
Yeah, I think in that case, in the pneumococcal 
case, the Pfizer vaccine did — people reviewing

4:37:47.600,4:37:54.879
what effect did this really have — they think that 
did get rolled out quicker, maybe scale up years

4:37:54.879,4:37:59.439
quicker than it would've otherwise. I think 
the Serum one ended up coming through later,

4:37:59.439,4:38:05.919
and maybe being less affected. But the scale 
up is so important for actually getting drugs

4:38:05.920,4:38:11.199
to people who need them, not just inventing 
cool stuff. And when I think of applying this

4:38:11.199,4:38:20.560
across to lenacapavir, I think to this great 
piece that Kamal Nahas wrote in Asimov Press

4:38:20.561,4:38:27.279
about lenacapavir, and where he touched a bit 
on the voluntary agreements in 120 countries,

4:38:27.279,4:38:35.840
and what's happening outside of those countries.
Maybe this is the shape of problem that,

4:38:35.840,4:38:44.000
for those 80 countries, where there's not in 
each country, enough demand, or there's too

4:38:44.000,4:38:53.520
much uncertainty around demand, for a company — 
Gilead or generic company — to enter that market,

4:38:53.520,4:38:59.920
and try and start selling to the public 
healthcare systems. If there were an AMC

4:38:59.920,4:39:08.959
that aggregated across those countries, and made 
the demand clearer, and had a price that was fair,

4:39:08.959,4:39:15.199
but also enough that money could be made to 
make it sustainable for the companies entering,

4:39:16.080,4:39:19.439
maybe that's a place for an AMC, I don't know.
What do you think of that?

4:39:19.439,4:39:25.199
That's a great example of where it can be used. 
I actually also think it could be used in scaling

4:39:25.199,4:39:31.359
up a drug even once it's been approved, because 
this second part of what an AMC is used for, in

4:39:31.359,4:39:40.400
the scale up — having the amounts that companies 
receive be based on the amount they manufacture

4:39:40.400,4:39:47.279
means that you're incentivizing this large-scale 
manufacturing, and actually administering it to

4:39:47.279,4:39:56.000
people. That is something that could still be 
used even now. But one of the other applications

4:39:56.000,4:40:05.039
is as a way to pull funding towards some drug or 
vaccine or some product that hasn't yet been made,

4:40:05.039,4:40:10.959
so that would be another option. Can we 
develop a drug that's better than lenacapavir,

4:40:10.959,4:40:19.680
or easier to take, or so on, and fund it with this 
new model. And I think this reminds me as well,

4:40:19.680,4:40:26.160
I don't know if we mentioned it earlier, but as 
far as I know, Gilead is also trying to produce

4:40:26.160,4:40:33.279
improvements on lenacapavir that would be taken 
once per year instead of once per six months.

4:40:33.279,4:40:35.039
Imagine that, wow.

4:40:35.039,4:40:42.879
That would be very cool. I was wondering 
about why they were doing this. I mean,

4:40:42.879,4:40:50.080
if you were self-interested profit-making company, 
why not just stick with this already-amazing drug?

4:40:50.080,4:40:56.879
And it occurred to me, when you were talking 
about Merck's drug islatravir — that's this

4:40:56.879,4:41:02.639
oral pill that's once-per-month. If someone could 
choose between an oral pill once per month and

4:41:02.639,4:41:10.561
an injected drug once per six months, they might 
choose the pill. Not only the person themselves,

4:41:10.561,4:41:16.400
but the clinics might find it easier to distribute 
the drugs. It's just, you don't need a healthcare

4:41:16.400,4:41:22.798
worker, or a nurse, or someone to inject the 
drug if it's a pill. And that made me wonder,

4:41:22.799,4:41:26.959
maybe that was the incentive. That 
was the reason that they decided:

4:41:26.959,4:41:32.080
'Let's go even further, to make this 
thing that's even harder for Merck's

4:41:32.080,4:41:36.320
drug to beat.' I don't know if that's 
the case, but that's what I would guess.

4:41:36.320,4:41:41.920
And it just goes to show how much progress 
we've made with HIV. Because when you were

4:41:41.920,4:41:47.600
describing the first drugs around, they were 
not so good and there were no competitors,

4:41:47.600,4:41:54.639
and now we have great options for 
patients, and great options for people who

4:41:54.639,4:42:03.119
don't even have HIV yet who want to reduce 
their risk. So I'm glad to be alive today.

4:42:03.119,4:42:12.240
It's so much- I mean, the whole timeline is just 
incredible to think about. We talked about how,

4:42:12.240,4:42:20.160
in the early 1980s, how pessimistic or 
how scary it would've been to have HIV,

4:42:20.160,4:42:25.786
not have any treatments, have this- 
thought-of-as-this untreatable disease,

4:42:25.786,4:42:33.119
as just this behavioural problem; there's nothing 
that someone can really do medically to treat it.

4:42:34.879,4:42:41.279
Contrast that with where we are now. I think, 
personally, that process, that timeline could

4:42:41.279,4:42:49.039
have been sped up. Just reading about some of 
the details of early research in the 1980s,

4:42:49.039,4:42:54.240
but also, knowing about how long it takes to 
run a clinical trial, how long it takes to set

4:42:54.240,4:43:01.279
up the trial sites, or train the nurses and the 
healthcare workers, or to share this information

4:43:01.279,4:43:09.279
between different research groups and so on. But 
at the same time, it is just an incredible story.

4:43:09.279,4:43:18.959
Now is probably the time to step back on this 
story that we have told and conclude. I am sure

4:43:18.959,4:43:27.199
we missed out many subplots that are also ripe for 
discussion. But among what we have discussed, I'm

4:43:27.199,4:43:34.400
interested to hear: What were your main takeaways 
from this story over the last fifty years?

4:43:34.400,4:43:41.839
My takeaway... I mean, I had so many takeaways. 
One of them was just how many different ways

4:43:41.840,4:43:47.119
you can approach medical innovation. Like, 
what are the different things that you could

4:43:47.119,4:43:52.080
think are important here? I mean, partly 
it's 'Let's make a really effective drug',

4:43:52.080,4:43:59.279
but it's also 'How do we make a drug that's easier 
for people to take on a regular basis?' It's,

4:43:59.279,4:44:05.679
maybe, refining drugs that already exist — trying 
to improve on them, in terms of their safety,

4:44:05.680,4:44:12.480
efficacy, or again, how people take them.
The other was the different types of

4:44:12.480,4:44:19.759
drug development. So we talked about this 
trial-and-error process with the first drug

4:44:19.760,4:44:28.400
azidothymidine, where they just looked at some 180 
compounds: tried each of them in the lab, saw what

4:44:28.400,4:44:37.039
worked in cells in the lab, and then scaled up 
based on that. The other is, this is an example of

4:44:37.039,4:44:44.879
screening existing compounds, or just compounds 
in nature, that helps repurpose this previous

4:44:44.879,4:44:52.959
cancer therapeutic that didn't work, for HIV.
Then we had some examples where this understanding

4:44:52.959,4:44:59.039
of the specifics of HIV, or how the enzymes 
work, what they look like, what will fit

4:44:59.039,4:45:08.719
into these little gaps between them — that was 
another option for developing a new drug. But,

4:45:08.719,4:45:14.080
at the same time, there was so much iteration 
and adjustment — that was tinkering — that was

4:45:14.080,4:45:21.039
important there. The move from this drug that was 
potentially promising, to one that actually met

4:45:21.039,4:45:27.680
several criteria that you would have, with the 
efficacy, the safety, how long lasting it was.

4:45:27.680,4:45:35.119
And then I guess, there are these other- 
improving on the drugs that already exist,

4:45:35.119,4:45:43.680
that is not just trying to use existing 
information. If there has been, already,

4:45:43.680,4:45:50.320
a protease inhibitor invented, can you now develop 
a new one based on that knowledge? Can you develop

4:45:50.320,4:45:59.039
a different type of nucleoside inhibitor, like 
AZT? Will people develop more capsid inhibitors,

4:45:59.039,4:46:05.039
based on the knowledge that they have from this? I 
mean, all of this, I think, is super interesting.

4:46:05.039,4:46:14.959
I totally agree on how much the tinkering and 
iteration stands out, as important. Basically,

4:46:14.959,4:46:19.680
at every level, there's.. so much of 
that's happening at the screening stage,

4:46:19.680,4:46:26.719
so much of that is happening, as you just said, 
at designing a capsid inhibitor that makes sense

4:46:26.719,4:46:32.879
for patients in a particular context.
Another thing that stood out to me is

4:46:32.879,4:46:42.480
how hard science is to predict — in the sense 
of, thank goodness people in the last forty,

4:46:42.480,4:46:50.638
fifty years, scientists did not only work 
on vaccines. We don't have an HIV vaccine,

4:46:50.639,4:46:56.639
and we do have a HIV preventive 
drug that you can get injected with,

4:46:56.639,4:47:02.320
and kind of feels like a vaccine. I'm 
so grateful that people were exploring

4:47:02.320,4:47:09.759
different parts of the technology tree there. 
We don't have a cure either. We don't have

4:47:09.760,4:47:15.199
a cure for HIV and we don't have a vaccine, 
but guess what? We have game-changing tools,

4:47:15.199,4:47:22.560
and that came from a part of exploration you 
may not have been able to predict back in 1981.

4:47:22.561,4:47:30.799
Also, the fact that the treatments could be used 
as a prevention as preventive drugs was probably

4:47:30.799,4:47:37.199
not that obvious to scientists at the time. 
That itself was quite unpredictable, I think.

4:47:37.199,4:47:38.400
Absolutely.

4:47:38.400,4:47:43.839
The other thing that reminded me of 
was just how many different aspects,

4:47:43.840,4:47:49.359
or how many different types of science- 
or what is involved in developing a drug,

4:47:49.359,4:47:58.240
is not just one person tinkering with it in the 
lab. It's this whole network of clinical trials

4:47:58.240,4:48:05.840
that are running; there's the basic research, 
there's stuff like developing microscopes

4:48:05.840,4:48:10.719
with a high-enough resolution that you can 
really see what is happening inside the cell,

4:48:10.719,4:48:17.840
what this virus looks like, what the proteins look 
like. There's the DNA sequencing technologies,

4:48:17.840,4:48:25.359
there's the protein development. Like, all 
of this stuff comes together to develop these

4:48:25.359,4:48:30.320
drugs. And then there's the medicinal chemistry, 
and the pharmacology — which I think, probably,

4:48:30.320,4:48:37.439
I had kind of underrated before as just, okay, 
this seems like this last-minute thing; after

4:48:37.439,4:48:42.878
you've developed a drug, you now want to make 
sure that it's safe and effective. I previously

4:48:42.879,4:48:49.600
had this assumption that that was what medicinal 
chemistry was about, and now I'm thinking this

4:48:50.480,4:48:57.519
can make a huge difference on whether a drug is 
useful or effective at all in the real world.

4:48:57.520,4:49:03.039
Even right at the discovery stage of lenacapavir. 
So important, how stable it is and how it doesn't

4:49:03.039,4:49:09.600
break down very quickly. I totally agree with 
what you just said about interlocking parts of

4:49:09.600,4:49:20.240
the medical innovation system. It's amazing that 
we just about had recombinant DNA at the time when

4:49:20.240,4:49:28.879
HIV started becoming a crisis. So we could use 
recombinant DNA in the lab as a research tool

4:49:28.879,4:49:33.600
and that we did not have all of those other things 
we just mentioned. We didn't have PCR, we didn't

4:49:33.600,4:49:35.039
have electron microscopes- well, I dunno about 
that- we didn't have cryo-electron microscopes,

4:49:35.039,4:49:41.920
certainly. We, only in the last few years, have 
learned the capsid, which we are now inhibiting

4:49:41.920,4:49:48.639
with lenacapavir, in fact stays intact into the 
nucleus. There's so much more that we're still

4:49:48.639,4:49:55.520
going to learn over the next coming years that 
might open up new frontiers. And that's true not

4:49:55.520,4:50:02.240
just for HIV, I bet you that things that HIV 
researchers have learned will be useful for

4:50:02.240,4:50:09.519
hepatitis B, for other cancers, for this, that, 
and the other. And the final interlocking surprise

4:50:09.520,4:50:19.199
of science that I learned from you many hours 
ago was that it was only two years before 1981

4:50:19.199,4:50:26.240
that the first human retrovirus was discovered. 
Thank goodness for that. And it makes you wonder

4:50:26.240,4:50:33.359
what we don't yet understand, that will make 
solving and curing diseases in the future easier.

4:50:33.359,4:50:41.279
Yes! And, and, above all of that, the other 
interesting thing for me was learning about

4:50:41.279,4:50:47.119
how drug pricing works or how these patents 
work. How does manufacturing actually work

4:50:47.119,4:50:55.119
at large scale? What do the funding models- what 
they have to do with whether drugs are developed,

4:50:55.119,4:51:03.599
how fast they're rolled out, who's ready to 
pay for certain drugs? And it's every aspect

4:51:03.600,4:51:09.920
of this whole process: not just the lab, not 
just the clinical trials, but the funders,

4:51:09.920,4:51:15.760
the people who decide, who show support 
for foreign aid spending, for example.

4:51:15.760,4:51:22.799
Everything comes together when we're talking 
about any disease, but particularly for HIV, it's

4:51:22.799,4:51:30.561
so salient because of these huge programs that 
have transformed the lives of millions of people.

4:51:30.561,4:51:36.320
And going into the future. There's no point for 
all of this wonderful science unless we remain

4:51:36.320,4:51:43.439
ambitious. And unless we make sure that people who 
need these drugs can access them. It's possible,

4:51:43.439,4:51:55.839
we've done it before and into the future 
we go, with uncertainty and with resolve.

4:51:55.840,4:52:01.039
This episode was only possible from a lot of 
work done by a lot of people publishing in the

4:52:01.039,4:52:05.600
open — whose papers we read and whose reviews 
we read. I won't thank them all here, but we'll

4:52:05.600,4:52:10.400
leave, in the show notes, some of the research 
that we base this episode on. I, in particular,

4:52:10.400,4:52:17.359
would like to thank Anne de Bruyn Kops, who wrote 
a great review of long-lasting injectables for

4:52:17.359,4:52:23.279
many different diseases for Open Philanthropy, 
that I learned a lot from. I want to thank Sanela

4:52:23.279,4:52:31.679
Rankovic, who was the HIV researcher who knows 
all about PF-74. And then, of course, I'm sure

4:52:31.680,4:52:38.160
we both want to thank Douglas Chukwu, who joined 
us for our first ever phone-a-friend section.

4:52:38.959,4:52:45.359
Yes. And the team at our Works in 
Progress, Aria Babu, who helped

4:52:45.359,4:52:50.480
us really, actually, get this 
podcast to run Adrian Bradley,

4:52:50.480,4:52:57.279
who's here with us now producing and keeping 
us on track with this episode. Then, the team

4:52:57.279,4:53:04.080
at Works in Progress and Open Philanthropy, 
who were sponsoring this podcast. And then,

4:53:04.080,4:53:10.320
I would say, also, all of the scientists who 
were involved in developing all of these drugs,

4:53:10.320,4:53:14.799
all the people who were participating in all 
the clinical trials, all the healthcare workers

4:53:14.799,4:53:22.320
who worked in them, everyone involved in this 
massive program, PEPFAR, everything. It's just-

4:53:23.279,4:53:24.080
It's so cool.

4:53:24.080,4:53:25.600
It's very inspiring.

4:53:25.600,4:53:30.400
And with that, I will ask you as 
listeners, if you enjoyed this episode,

4:53:30.400,4:53:36.240
feel free to subscribe. We will be talking 
about other Hard Drugs in the future,

4:53:36.240,4:53:39.920
and check out the show notes 
for more details on this one.

4:53:39.920,4:53:43.359
Bye. Bye.