0:10

Hey, everyone. Welcome to the Drive

0:12

Podcast. I'm your host, Peter

0:14

Atia. This podcast, my website, and

0:17

my weekly newsletter all focus on

0:19

the goal of translating the science of

0:21

longevity into something accessible for everyone. Our

0:24

goal is to provide the best content

0:26

and health and wellness, and we've established a

0:28

great team of analysts to make this

0:30

happen. It is extremely important to me to

0:32

provide all of this content without relying

0:34

on paid ads. To do this, our work

0:36

is made entirely possible by our members,

0:39

and in return, we offer

0:41

exclusive, member -only content and benefits

0:43

above and beyond what is available

0:45

for free. If you want

0:47

to take your knowledge of this space to the

0:49

next level, it's our goal to ensure members

0:51

get back much more than the price of a

0:53

subscription. If you want

0:55

to learn more about the

0:57

benefits of our premium membership, head

0:59

over to PeterAteaMD.com forward slash

1:01

subscribe. My

1:04

guest this week is Dr. Susan

1:06

Desmond -Hellman. Sue is

1:08

a physician who is board certified in

1:10

internal medicine and medical oncology. Her

1:13

impressive career has spanned multiple fields. She

1:15

has been a leader in the

1:17

pharmaceutical industry where she helped develop several

1:19

ground -baking drugs, worked as the

1:21

chancellor of the Health Science campus, of

1:23

a major university system, UCSF, and

1:25

served as the CEO of the Bill

1:27

and Melinda Gates Foundation. She

1:29

also served on numerous boards of

1:31

both corporations and nonprofit organizations. She

1:34

co -chaired the National Academy of

1:36

Science Committee that pioneered precision medicine

1:38

and currently sits on the board of OpenAI. I

1:40

wanted to have Sue on this

1:42

podcast to speak about her

1:44

extraordinary career spanning medicine, oncology, biotech,

1:46

and global health leadership, and

1:49

to really explore her knowledge on how

1:51

scientific innovation and leadership can drive better

1:53

health care outcomes. In this

1:55

episode, we discuss her early days in medicine

1:57

training at UCSF during the start

1:59

of the AIDS crisis before people even

2:01

knew what it was and the lessons

2:03

that she learned on handling uncertainty, balancing

2:06

public health messaging and accelerating

2:08

treatment breakthroughs, the decision that she

2:10

made to specialize in oncology and

2:12

how her time treating HIV related

2:14

cancers in Uganda reinforced the need

2:16

for integrating epidemiology patient care, and

2:18

policy to combat global health crises.

2:21

We spoke about her transition into

2:23

biotech, helping develop breakthrough cancer

2:25

drugs like taxol, Herceptin and

2:27

Avastin, and the role of precision medicine

2:29

in improving outcomes. Sue talks

2:32

about her leadership roles at UCSF

2:34

and at the Gates Foundation, driving

2:36

innovation in healthcare and global health,

2:38

and the lessons learned from leading

2:40

health research institutions and global health

2:43

initiatives, balancing financial constraints with scientific

2:45

progress and building culture. We

2:47

end this discussion with a perspective

2:49

on the future of medicine,

2:51

including AI's role in healthcare, such

2:53

as the opportunities and challenges in

2:55

leveraging AI for drug development, diagnostics,

2:57

and expanding access to high -quality

2:59

care. So without further delay,

3:01

please enjoy my conversation with Dr.

3:04

Sue Desmond -Hellman. Sue,

3:10

thank you so much for making the trip

3:12

out to Austin. Really, really was excited to

3:14

meet you last year. Just an honor to

3:16

spend part of a day with you and

3:18

then realize that I could somehow twist your

3:20

arm into coming on the podcast. I'm happy

3:22

to be here. I look forward to it.

3:24

You've had just an unbelievable career. You are

3:26

an absolute giant in many ways. I love

3:28

to always give people a sense of how

3:30

someone got to where they got. So if

3:32

I recall, you grew up in Reno. Is

3:34

that right? I did. Yes. And you went

3:37

to high school and college and even medical

3:39

school. All the way through, right? I

3:41

went to Catholic school for 12 years in

3:43

Reno. I explained that when people wonder if I

3:45

was at a casino for my child. And

3:47

then I went to University of Nevada,

3:49

both to undergrad and to medical school. Then

3:52

you ended up at UCSF for your residency? You

3:55

know this with residencies. My

3:57

dream residency was internal medicine

3:59

at UCSF. My first pick, and

4:01

I got my first pick and

4:03

went to UCSF as an internal medicine

4:05

resident. And that would have been

4:08

what year that you landed there? 1982.

4:11

Okay, so remind me where we

4:13

were in the AIDS epidemic in

4:15

San Francisco in 82. What

4:17

was known? If you read

4:19

MMWR, that's 1981, was

4:21

the first indication. In

4:23

1982, we knew that there was

4:25

something happening, especially to gay

4:27

men, but there was a sense

4:29

it was homosexuals, hemophiliacs, and

4:31

Haitians. Remember that? Yeah. Three Hs.

4:33

There was so much mystery

4:35

still involved that I was and

4:37

my colleagues were in a

4:39

study to look at drawing our

4:41

blood to see if we

4:43

had been infected as a result

4:45

of treating patients. And what

4:47

were they presenting with at the

4:49

time? It's hard for anyone

4:51

of even my generation. You've never

4:53

seen a drug naive patient.

4:55

All of my experience with HIV,

4:57

which was a lot in

4:59

Baltimore many years later, but everyone

5:01

was on something. So how

5:04

would these men present to you

5:06

as a medicine resident? Pneumocystis.

5:08

Pneumocystis carinia pneumonia was

5:10

the number one diagnosis.

5:13

So that's what you saw in the

5:15

hospital that brought patients to attention.

5:17

Tell folks why that's so unusual. Oh,

5:20

it was a disease that

5:22

immunosuppressed patients could get very rarely.

5:24

Most clinicians had never seen

5:26

it before. What was

5:28

also clear is that there

5:30

were many other infections that

5:32

were not as obvious or

5:34

life -threatening as pneumocystis was when

5:36

we saw it right away.

5:39

What was interesting from an

5:41

outpatient perspective was Capsis sarcoma.

5:43

Tell folks what that is.

5:45

Capsis sarcoma is a really

5:47

unusual purple -colored tumor, very

5:49

visible externally. It caused

5:52

nodules. In patients with

5:54

HIV infection, it also

5:56

caused internal organ involvement.

5:58

And patients would cough up blood

6:01

or they would vomit blood. But

6:03

what was really sad and difficult

6:05

is the combination of catechia and

6:07

Capsis sarcoma meant that everybody knew

6:09

you had AIDS. They sort of

6:11

wore it. And what was interesting

6:14

for me was that this old -fashioned

6:16

Capsis sarcoma was fundamentally different than

6:18

what we were seeing. We also

6:20

saw a non -Hodgkin's lymphoma. in

6:22

numbers much smaller than Capsi Sarcoma.

6:25

But Capsi Sarcoma was a very

6:27

big problem in San Francisco. It

6:29

was very common in Game N,

6:31

and it was common in the

6:33

population we saw. And was

6:35

there ever a sense of fear among the

6:37

medical staff that we don't know what this

6:40

is, we don't know how it's transmitted, and

6:42

therefore we don't know how to protect each

6:44

other or ourselves or other patients for that

6:46

matter? It's hard for me to imagine that

6:48

given how much we take for granted today. I

6:51

think it's... a reflection of

6:53

my own personality and my own

6:55

wish to be a physician, that

6:58

my memories of those days

7:00

are much more about sadness,

7:02

about my patients, and

7:04

about people my age dying, or

7:06

being pretty clear they were going to

7:08

die. I mean, a story

7:10

that brings it to life is

7:12

many patients started selling their life insurance

7:14

because they were sure they wouldn't

7:16

live long enough. and they wanted the

7:18

money now. And then when the

7:20

antiretroviral therapy came along, they wished they

7:22

hadn't, which is a good thing

7:24

to have. But I was just really

7:26

sad. There were fears about the

7:28

residents and about contagion. But in San

7:30

Francisco, there was such a wish

7:32

to help the patients and such a

7:35

good spirit about playing a role

7:37

in helping that we all persevered. But

7:39

the first patients I took care

7:41

of in the hospital, I remember very

7:43

well in 1982, We were a

7:45

gown, gloved, masked, had a cap on.

7:47

It was like we were going

7:49

into an operating room. Got it. For

7:51

all intents and purposes, you were

7:53

acting like this was Ebola without knowing.

7:55

Absolutely. Absolutely. And so you

7:57

finished your residency in internal medicine.

7:59

Did you go directly into your

8:01

fellowship? I did a chief

8:03

residency at the university hospital. And

8:05

I think that was the first that

8:08

I knew I really liked managing. I

8:10

really liked interacting with people and helping

8:12

people succeed. So I did that for

8:14

a year and then went into my

8:16

oncology fellowship after that year. And why

8:18

did you pick oncology? Oh,

8:20

to this day, I love

8:23

oncology. If you love medicine and

8:25

I do, and you love patients and

8:27

I do, it's the combination of

8:29

you get to call on your

8:31

compassion gene and your nerdy science

8:33

gene. And when I

8:35

was in Reno at the

8:37

Reno VA, I had

8:39

an attending, Stephen Hall. And he

8:41

was the oncologist who was teaching

8:43

me about medicine, third year medical

8:46

student. And I loved

8:48

everything about how he showed up.

8:50

I loved his compassion. I

8:52

loved his intellect. And

8:54

after that, I had in my

8:56

mind this bug about, I

8:58

wanted to be like him. I

9:00

can really relate to that. When I

9:02

was in my third year of medical school,

9:05

I went to the NCI for three

9:07

months with Steve Rosenberg. Oh. And it was

9:09

the exact same experience. And I remember,

9:11

learning many lessons from Steve. One of them

9:13

was that cancer diagnosis, and of course,

9:15

at the NCI, as you know, nobody's showing

9:17

up with stage one, two, or three

9:19

cancer. By definition, every patient there is showing

9:21

up with metastatic cancer, and they've progressed

9:23

through all standard treatments. So these are people

9:25

that have six months to live, and

9:27

maybe 10 % of them you actually come

9:29

up with a durable remission for. But

9:31

he said, look, cancer will do one of

9:33

two things to a family. It will

9:36

take a strong family and bring them much

9:38

closer together. it will take a fractured

9:40

family and blow them wide apart. As

9:42

a doctor, as a nurse, as an

9:44

anybody in the field of oncology, your

9:46

ability to kind of be there for

9:48

that family is as important, potentially more important

9:51

than it is in any other specialty

9:53

of medicine. That's really well said. I love

9:55

that. So, tell me a

9:57

little bit about the state of

9:59

oncology in the mid -80s when you're

10:01

embarking on your medical oncology fellowship. Help

10:03

people understand what the world of

10:05

cancer looked like roughly 40 years ago.

10:08

Let me talk about breast cancer.

10:10

That's a cancer that is such a

10:12

good example. The common

10:14

therapy that was used, cytoxin,

10:17

methotrexate, 5FU, were very

10:19

old, decades old. There

10:21

were no new chemotherapy drugs, hadn't

10:23

been in a while. The field was

10:25

stifled, I would say, in terms

10:27

of medical oncology. There wasn't a lot

10:30

going on. I was really

10:32

interested in cancer epidemiology. That was

10:34

something to me that asking the question

10:36

why did people get cancer and

10:38

couldn't we do something about it seemed

10:40

really important to me. I

10:42

wanted in the second year

10:44

of my fellowship to study the

10:47

relationship between hepatitis B and

10:49

hepatocellular carcinoma and to understand that

10:51

better and to think about

10:53

the viral link with cancer. The

10:55

mentor I was supposed to work with

10:58

ended up not coming to San Francisco.

11:00

So I decided to go to Berkeley

11:02

and get a Master's in Public Health

11:04

as a backup strategy. I really scrambled

11:06

because I didn't want to waste a

11:08

year. What was the nature of the

11:10

program? It was a three -year fellowship

11:12

with a research track on the side

11:14

because obviously UCSF is such an academic

11:17

place. It's a very academic place, but

11:19

you could do two or three years

11:21

and many people went into the lab.

11:23

I didn't want to go into the

11:25

lab. I wanted to do epidemiology. I

11:27

wanted to learn more about statistics and

11:29

epidemiology. I thought I wanted to do

11:31

it because I wanted to be a

11:34

cancer epidemiologist. And to this

11:36

day, I still think that is

11:38

one of the great opportunities to

11:40

make a big impact, but you

11:42

have to be funded. So I'm

11:44

a pragmatist. The good news

11:46

was that all that learning

11:48

at Berkeley and at UCSF

11:50

in epi and biostat, I

11:52

brought to drug development. Clinical

11:55

trials have a lot in common

11:57

with doing epidemiology. You brought up

11:59

the example of Hep B and

12:01

Hepatocellular carcinoma. Was it understood at

12:03

the time, what we now know?

12:05

It was, yeah. Palmer Beasley, one

12:07

of the fathers of that relationship,

12:09

was the guy who was supposed

12:11

to come. There were preliminary papers

12:13

and something relatively early, but it

12:15

was emerging science. Do you recall

12:17

what the incidence of Hep B

12:19

was and Hep C back then?

12:21

You know, I don't. If you

12:23

weren't in Asia, it was actually,

12:25

I think, relatively low. but I

12:27

believe increasing, which is partly why

12:29

the vaccines are so important. Tell

12:31

me about how you wound up

12:33

in Uganda. After I

12:36

got my master's in public

12:38

health, I became the oncologist at

12:40

UCSF in the University Hospital

12:42

for the AIDS Clinic. This is

12:44

Moffitt? This is Moffitt. So

12:47

San Francisco General had a very

12:49

well -known program run by oncologists

12:51

for AIDS patients who were

12:53

in the safety net hospital. But

12:55

in the university hospital, if

12:58

you were very sick and you had Capsi

13:00

Sarcoma, you saw me. And my

13:02

husband, because we had just gotten married, we

13:04

were interns together, he was

13:06

in the lab in ID doing

13:08

immunology work. So two

13:10

of the chiefs of medicine

13:13

at UCSF were approached by

13:15

the Rockefeller Foundation, who had

13:17

started to become worried about

13:19

heterosexual transmission of HIV. Remember

13:21

I talked about the Haitians

13:23

and the hemophiliacs and homosexuals?

13:25

One H wasn't heterosexual. And

13:28

so there was a lot of

13:30

disbelief about African HIV. And

13:32

in fact, some people thought it

13:34

must be gay sex, but people are

13:36

too embarrassed to admit it. There

13:38

were other theories, but people just did

13:41

not understand what was going on

13:43

in Africa. So the Rockefeller said, we'll

13:45

give you a grant at UCSF.

13:47

We'll grant you money to study heterosexual

13:49

transmission of HIV. And this was

13:51

through an epidemiologic contact tracing lens, not

13:53

necessarily going into the lab and

13:56

trying to figure this out. Not going

13:58

to the lab, but really looking

14:00

at API. And particularly, there was a

14:02

hypothesis that if it was heterosexually

14:04

transmitted, there was something to do with

14:06

sexually transmitted diseases. And that there

14:08

was something about increasing your risk if

14:11

you had untreated STDs, sexually transmitted

14:13

diseases. So we were asked

14:15

to go. UCSF had no global

14:17

health to put this into context. We

14:19

had a flat and two Honda

14:21

Civics. I still remember this. We gave

14:23

my dad power of attorney. We

14:25

sublet our flat and we sold our

14:27

Hondas. And moved to Uganda. I'm

14:30

laughing in part because I had

14:32

never been east of Chicago. I

14:34

mean, this was a pretty

14:36

dramatic thing to do. And it was

14:38

only... And I'm sorry, you, your

14:41

husband and who else? The two of

14:43

us. That's the team. That's the

14:45

dream team. That's the team. That's the

14:47

team. And Uganda was a place

14:49

where On the positive side of things,

14:51

the NCI had set up a

14:53

collaboration with Uganda Cancer Institute, where they

14:55

did some really great things in

14:57

lymphoma and Burkitt's lymphoma, if you remember

14:59

those stories. And one of the

15:01

physicians at UCSF had been associated with

15:04

that, John Ziegler. So there

15:06

was a connection to the Uganda Cancer

15:08

Institute. So on the good side of

15:10

things, there was that. And there also

15:12

was and is the Entebbe Viral Institute.

15:14

So there was some infrastructure there. Unfortunately,

15:17

most of that infrastructure had

15:19

been ruined by the idioming

15:21

regime not long before we

15:23

went to Uganda. So

15:25

when we went there, it was

15:27

pretty lawless. There were roadblocks you

15:29

had to stop at. It was

15:31

difficult to live there. It was

15:33

really difficult. And what about safety?

15:35

I would say now that I'm

15:37

used to being in more safe

15:39

situations and older and wiser, it

15:41

was probably not that smart the

15:43

way we live there, but we

15:45

weren't reckless. It seemed dangerous when

15:47

you were in the car to

15:50

have carjacking or your money go

15:52

missing or things like that. Was

15:54

Idi Amin still ruler? Idi

15:56

Amin was gone, but when we were there,

15:58

he made that attempt to come back from

16:00

Saudi Arabia and go back to Uganda, but

16:02

it was thwarted. So that was good news.

16:05

So Nick, my husband, reestablished the

16:07

sexually transmitted disease clinic and

16:09

attended in the internal medicine ward.

16:11

And I like to say

16:13

I doubled the population of oncologists

16:16

in Uganda when I was

16:18

there. So my colleague, Edward

16:20

Ambiti, who's Ugandan, put all his

16:22

focus on the pediatric unit. And

16:24

I put all my focus on

16:26

the adult unit, which was so

16:28

many cases of Capsis sarcoma. Give

16:31

me a sense of what this

16:33

meant. So we're talking late 80s

16:35

now. This is 89, 90, and

16:37

91. Is AZT out yet? Not

16:39

yet. Okay. Just on the brink.

16:41

Okay. So we have nothing. And

16:43

what is the approximate conversion? So

16:46

for a patient who develops

16:48

AIDS, what fraction of those

16:50

will go on to develop KS?

16:52

If you were in Uganda

16:54

at the time, especially

16:56

amongst males, but also males

16:58

and females, it's so

17:00

hard to give those numbers. But I

17:03

would say about a third of

17:05

patients who sought medical attention probably had

17:07

KS, some KS. What was the

17:09

prevalence of HIV AIDS in the population

17:11

in Uganda? Depending on

17:13

the population you treated, it was

17:15

double digits in the country as a

17:17

whole. If you were 16 years

17:19

old, if you were a 16 -year -old

17:21

girl and you went to the STD clinic,

17:24

you had a 50 % chance of being

17:26

HIV positive. 16 and most of

17:28

those girls was their first and only

17:30

sexual partner. It was Russian roulette to have

17:33

sex. in Uganda then. I mean, worst

17:35

Russian rule, that's one in six if you've

17:37

only got one bullet in the chamber.

17:39

You got the bullet in one of the

17:41

two chambers. Yeah. And

17:43

the best business in

17:45

town, coffin maker. We

17:48

would go, we would drive back

17:50

to where we stayed and you would

17:52

see if you've ever been in

17:54

an African village like they'll prop up

17:56

the coffins made of wood and

17:58

you just see them because that was

18:00

the feeling of being scared and

18:02

sad in San Francisco in 1982. multiply

18:05

that by 1 ,000

18:07

in 1989, it

18:10

was terrifying. If we

18:12

hadn't gotten ARVs, this was

18:14

killing people. But you know, the

18:16

same time, the first time

18:18

we went back to San Francisco

18:20

from Uganda was six months after we

18:22

had left. I went

18:24

back to the Capsis Sarcoma Clinic that

18:27

I had led and said to the

18:29

nurse, oh yeah, ask about your patients.

18:31

I had so many great guys who

18:33

I cared for. All my patients were

18:35

dead, all of them, six months. The

18:38

sense of how bad

18:40

HIV was before antiretrovirals,

18:42

it's impossible to overstate

18:44

it, just impossible. And

18:47

when we were in Uganda, it

18:50

was really clear that you could

18:52

see someone's immune status with a good

18:54

physical exam if they had Kaposi

18:56

sarcoma. I wrote a paper that I

18:58

think is a good paper if

19:01

you do global health and you have

19:03

limited resources. It was a

19:05

paper that had one observation. If

19:07

you had Kaposi sarcoma on your soft

19:09

palate, on the roof of your

19:11

mouth, you had HIV, 100 %

19:13

predictive. Kaposi sarcoma, there's

19:15

a Mediterranean form and an

19:17

African form. It happens on your

19:19

skin. It can cause elephantiasis,

19:21

but it doesn't go in. The

19:23

mouth is just a surrogate

19:25

for your GI tract. Doesn't happen

19:27

unless you're immunosuppressed with HIV. These

19:30

patients weren't... dying

19:32

from the KS directly.

19:35

That's a proxy for how weak their immune system

19:37

was. I assume they were ultimately dying from

19:39

pneumonia. Many would die

19:42

from pneumonia. There was severe catechia,

19:44

and then they were prone

19:46

to pneumonia and other problems. But

19:48

Kaposi's sarcoma in the lungs

19:50

or the stomach can also cause

19:52

bleeding, and you can die

19:54

from that. What did you know

19:56

at this point in time about HIV? Because

19:59

the virus had been identified by this point.

20:01

What was known and what was unknown? We

20:03

knew most of the clinical

20:05

syndromes associated with HIV. Gallow, was

20:07

it Gallow? Yeah, Bob Gallow

20:09

was one of the... Luke Monnier

20:11

was... They had a fight

20:14

over who deserved the credit. But

20:16

yeah, we knew about HIV then

20:18

and we knew the biology and we

20:20

knew as soon as we got

20:22

to Uganda and examined patients that this

20:24

was heterosexual transmission of HIV. And

20:27

we knew that untreated STDs

20:29

were a big reason. And that

20:31

was a very important thing.

20:33

Going back to these 16 -year -old

20:35

girls, is the reason

20:37

that the heterosexual transmission was so

20:39

high because the viral loads were

20:41

through the roof? Because today, if

20:43

a male with HIV had unprotected sex with

20:46

a female, it would not be that high, it?

20:48

It wouldn't be that high. No. So

20:50

one of the really

20:52

important aspects of STDs

20:54

is high frequency of

20:56

herpes and chancroid. really

20:58

open lesions that are

21:00

very, very, if not

21:02

treated. one -two punch. Super

21:05

high viral load. High viral load

21:07

and transmissible. opening. Yes. Yes. So

21:09

we knew all of

21:11

that. Now, we also knew

21:13

that some of these

21:16

were treatable, that both medication,

21:18

also Museveni, the still leader of

21:20

Uganda, had this very funny campaign

21:23

called Zero Grazing. So they raise

21:25

a lot of cows, and this

21:27

is very important in Uganda is

21:29

having a herd of cows. It

21:31

means you're an important man, you

21:33

know. Moose Avenue wears this hat

21:36

like he's raising cows. So zero

21:38

grazing, the farmers and many people knew what

21:40

that meant. One wife, one

21:42

partner, no grazing. And

21:44

so there was a pretty good public

21:46

campaign. We did a lot of condom

21:48

distribution. And so the

21:50

government was receptive to this. Yeah. They

21:52

understood the science. They understood

21:55

the epidemiology. And they were

21:57

completely on board with the campaign. They

21:59

were very on board. They also knew that

22:01

this was going to be a geopolitical

22:03

problem for them if people were dying in

22:05

the prime of their lives at the

22:07

rates they were. They got that. This was

22:09

really clear to them. What

22:11

other countries in Africa were afflicted to this

22:13

extent? In East Africa, there was quite

22:15

a bit. There was a lot of HIV

22:17

in Kenya, and there were programs like

22:20

the one that we had in Kenya. Tanzania,

22:22

there were others where it was more

22:24

unknown, I think not talked about. I

22:27

mean, the program I know

22:29

about most today is the program

22:31

my husband's been working with

22:33

for 15 years, which is Elizabeth

22:35

Glazer Pediatric AIDS Foundation. They

22:38

work now, I think, in 12 countries

22:40

in sub -Saharan Africa, and many of

22:42

the southernmost countries are heavily affected by

22:44

HIV still. Can you estimate in a

22:46

year how many people died from AIDS

22:48

in Uganda when you were there? Oh,

22:50

no, I can't estimate it. I guess

22:52

the point is it's a staggering number.

22:54

Yeah. And yet there were so few

22:56

of you that were on the front

22:58

lines. If there's 16 million people,

23:00

it wouldn't have surprised me if there

23:02

were a million people who died. I mean,

23:05

it's that kind of numbers. I'm probably

23:07

exaggerating, but not by much. And

23:09

I think the sense of feeling

23:11

overwhelmed is just really important. What

23:13

I realized I was doing I

23:15

don't know if you've interacted with

23:17

people in the military much, but

23:19

if they were on the battlefield,

23:21

they triage. I triaged. I triaged

23:23

in San Francisco. If you didn't

23:25

need chemotherapy, but you had Capsis

23:27

sarcoma, I didn't see you. What was the chemo?

23:30

The simple one was vincristine. Vincristine's

23:32

actually reasonably good against KS. I

23:34

used it in Uganda a lot.

23:36

It does cause some neuropathy, but

23:38

if you're careful about how much,

23:41

and then bleomycin. Again, you have

23:43

to be careful because of the

23:45

pulmonary toxicity. Good old -fashioned vincustine

23:47

and blio, and then Texel. Texel

23:49

was approved for Capsule Sarcoma after

23:51

I left Uganda. It wasn't a

23:53

drug before then. I would

23:55

see the patient and I would literally ask

23:57

them and their family, can you walk? If

23:59

you can walk. If yes, you're too healthy

24:02

for me. too healthy. We'll delay. There was

24:04

triage because I only had on the shelf

24:06

a certain amount of chemotherapy. How

24:08

did you manage the personal toll

24:10

of the grief and the death of

24:12

seeing this. I mean, look, I think

24:14

every doctor, to some extent, goes through

24:16

this where you try to sort of

24:18

compartmentalize what you're seeing. But the truth

24:21

of the matter is virtually no doctor

24:23

can really comprehend what you're describing there.

24:25

How did you process that? I

24:27

have this philosophy, which I

24:29

don't recommend it for others. It's

24:31

just my philosophy. I love people.

24:34

I love interacting with people. I

24:36

love getting to know the patients

24:38

who I care for. And

24:40

it makes me happy to think

24:42

I'm helping. Helping might be

24:44

helping them get better. Helping

24:46

might be helping with their pain or they

24:49

can talk about dying with me because

24:51

it doesn't make me scared. So I get

24:53

a lot of joy in trying to

24:55

contribute. Even if I feel overwhelmed and if

24:57

I step back and think, how can

24:59

we cope with this? My coping is... Is

25:01

leaning in. Yeah. Does your husband share

25:03

that? Was there a yin and a yang

25:05

to the relationship where you supported each

25:08

other in a way that was helpful in

25:10

that? I do understand what you're saying,

25:12

and I appreciate that there is a joy

25:14

that comes from helping people. But I

25:16

can also at least personally say that there

25:18

are moments when it breaks down and

25:20

you feel so overwhelmed by sadness. Well,

25:23

first of all, my husband is more

25:25

introverted and probably gets more sad. But

25:27

we are also a good team because

25:29

we're there for each other. And I

25:31

think it's a special thing done in

25:33

small amounts, not too much, to be

25:35

able to come home and say, boy,

25:38

that was tough. Here's what I dealt

25:40

with today, or I need to tell

25:42

this story, or I want to talk

25:44

about this. The other thing we did,

25:46

which is I think so important, is

25:48

I do drive a lot of joy

25:50

in trying to help, but I'm not

25:52

a martyr. I don't believe in it.

25:54

Okay, you worked hard, I worked harder.

25:57

You suffered, I suffered more. I hate

25:59

that, you know? So we went to

26:01

Greece. We still laugh about going to

26:03

Greece and eating our way through Greece

26:05

for a week. When we were in

26:07

Uganda, we had a couple

26:09

of other good trips. We went

26:11

on a hilarious safari to a place

26:13

that was Moya Lodge that had

26:15

been closed to all tourists, had just

26:17

reopened. And it was so

26:20

great. We saw hippos and elephants. And

26:22

we realized we were the only, what

26:24

you call in Uganda, a mazungu, which

26:26

is a white person there. So it

26:28

was a grand adventure. So we had

26:30

some grand adventures. played tennis,

26:32

enjoyed friends. We did as much

26:34

to keep our spirits up as one

26:36

can. And so you came back

26:38

to the US after about three

26:41

years. And did you go back

26:43

to UCSF? Well, we wanted to go

26:45

back to UCSF, but we had

26:47

not kept our academic careers going

26:49

as much as we should have.

26:51

We didn't publish enough and they didn't

26:53

have a global health program or

26:55

money for us. Taking care of

26:57

a million people with HIV wasn't

26:59

enough to justify coming back to UCSF.

27:01

It actually wasn't. So we said,

27:03

well, gee, when the chief of

27:06

medicine outlined for us the plan

27:08

for us to stay, a large part

27:10

of it was taking care of

27:12

patients to pay our way. So

27:14

we said, boy, taking care of

27:16

patients, we know what that looks like.

27:18

So we went into private practice.

27:20

In San Francisco? No, we moved

27:22

back to Kentucky where Nick is

27:24

from. Okay. So we moved back to

27:27

Kentucky and I was in a

27:29

two -person oncology practice with a former

27:31

classmate of Nick's in Lexington. And

27:33

you were doing at this point

27:35

oncology unrelated to, not necessarily focusing

27:37

on HIV and AIDS related cancer, breast

27:39

cancer. I was doing good old

27:41

fashioned American oncology. I didn't take

27:43

my oncology boards when we went

27:45

to Uganda because I was in Uganda.

27:48

So, I still sort of

27:50

laugh about taking the DeVita

27:52

oncology book with a yellow

27:54

sharpie. I reread the

27:57

big oncology book twice. Is

27:59

this the DeVita Hellman Rosenberg book? Yes. Yeah,

28:01

yeah, yeah. Of course. Yeah, it's brown these days,

28:03

I think. I reread it twice, took

28:05

my boards and did fine. So, I was

28:07

ready. This is

28:09

unbelievable. So, you're sitting in Kentucky

28:12

practicing garden variety oncology. Talk to

28:14

me about what that's like. completely

28:17

orthogonal to what you've been doing for the

28:19

past couple of years. was so, so, so,

28:21

so different. And Nick was in

28:23

a practice where he was more like a hospitalist.

28:25

Somebody'd get a fever in the ICU and they'd call

28:27

that ID group. And my practice was a two -person

28:29

practice. It was very classy. I saw a lot

28:31

of lung cancer. It was Kentucky. So there's

28:33

a lot of smoking, a lot of

28:36

people from Appalachian. I actually like

28:38

taking care of patients. So that part

28:40

I liked, but I really missed.

28:42

intellectual research, collegial stuff that I was

28:44

used to at UCSF because we

28:46

had been there nine years by that

28:48

time because we were still UCSF

28:51

faculty when we were in Uganda. Nick

28:53

was called about Bristol

28:56

-Meyer Squibb search for

28:58

an expert on HIV

29:00

because they were trying

29:02

to follow AZT with

29:04

the next antiretroviral. I

29:06

think it was DDI and D4T

29:08

were both in development then. And so

29:10

they recruited Nick to come and

29:12

work at Bristol Myers Squibb out of

29:15

private practice. And Nick

29:17

said, I won't come unless you have a

29:19

job for my wife. And they said, no,

29:21

we have a nepotism clause. We don't allow

29:23

couples to work at Bristol Myers Squibb. So

29:25

he said, fine, I won't come. He's a

29:27

good husband. This is

29:29

one of our favorite stories because it's a true

29:31

story. So they called him back and they

29:33

said, we really, really want you to come because

29:35

we want this program to do well. And

29:37

could your wife be a consultant? Would

29:40

she agree to be a consultant and not

29:42

a full -time employee? And he said, yeah,

29:44

that'll work. So we moved to Connecticut. He

29:46

had a job and I was the trailing spouse. And

29:49

I can just see what this looked like.

29:51

I'm making this up now. Oh,

29:54

God, we've got an LMD. You know

29:56

what an LMD is. We got this

29:58

lady from Lexington, Kentucky. She's in private

30:00

practice, oncology, and we're stuck with her.

30:03

Let's have her do drug safety on

30:05

tax law. We have this new drug

30:07

and it's really busy. It looks like

30:09

it might work. And so we'll put

30:11

her on drug safety. She can't hurt

30:13

anything doing that. But

30:15

did they not understand what you had spent

30:17

the last couple of years doing prior

30:19

to being in Kentucky? Did they not know

30:22

what you had done in Uganda? I

30:24

don't think that registered because

30:26

there were so many people

30:29

there. who were very traditionally

30:31

trained at NCI or at

30:33

Yale or wherever they were.

30:36

And they were traditionally trained in

30:38

oncology. My experience in Kampala

30:40

in Uganda didn't make an impact.

30:42

But here's what was funny. Nick

30:45

and I didn't have a statistician.

30:47

As I told you, we just, the

30:49

two of us went. So we

30:51

brought this little compact computer and all

30:53

the SAS manuals. We didn't have

30:55

a TV. We didn't have newspapers. We

30:57

didn't have anything. So we taught

30:59

ourselves how to do SAS programming. When

31:02

I got to Bristol Myers Squibb,

31:04

one of the really interesting things

31:06

about taxol is it causes severe

31:08

neutropenia, but it's short. It's like

31:10

this short severe neutropenia. And

31:12

so I wanted to study that

31:14

because I thought it was really important

31:16

in why people weren't really getting

31:18

infections. I always have to remind myself,

31:20

tell people what taxol is, how

31:22

it works. Just give them a quick,

31:24

what is neutropenia? Why would we

31:26

care? When I talked before about how

31:28

few new chemo drugs there were,

31:30

Taxol was one of the first new

31:32

chemotherapy drugs. So, taxol is a

31:34

product of the utri and it's a

31:36

microtubule poison. It is, if

31:39

you think about it coming from the

31:41

utri and you think about sap, think

31:43

about trying to dissolve sap in water

31:45

and give that to a patient. That's

31:47

plenty hard. The dissolving fluid

31:49

that's given with taxol. And the reason

31:51

we would give somebody with cancer

31:53

a microtubule inhibitor is because that prevents

31:55

cells that are dividing. You can't

31:58

divide. They need these microtubules when they

32:00

create new cells and we want

32:02

to block that. And we want to

32:04

block that. So this was not

32:06

just a good way to block cellular

32:08

division, which is so important in

32:10

cancer therapy. It's really important because it's

32:13

very different than some of the

32:15

old chemotherapy drugs. And if you're resistant

32:17

to those old drugs, here you

32:19

have a brand new mechanism of action.

32:21

So that's a terrific thing. But

32:23

it's not easy to dissolve it. So

32:26

the dissolving agents are like soap.

32:28

They dissolve the taxol. And when the

32:30

National Cancer Institute tried to use

32:32

it, some patients got severe allergic reactions

32:34

from that. And they got scared

32:36

and put it on the shelf. So

32:39

Bristol Myers Squibb went to the

32:41

National Cancer Institute and said, you know,

32:43

that drug might really be active.

32:45

We're willing to carefully go back in

32:47

the clinic and test it and

32:49

give people agents to counterbalance. the allergic

32:51

reactions and see if we can

32:53

get away with it. So they

32:56

did that and they got an

32:58

approval in ovarian cancer, a brand

33:00

new agent. Now, Taxol was

33:02

really exciting because first ovarian and

33:04

then breast cancer were these indications where

33:06

we had not had new drugs

33:08

or really any drugs that were active

33:10

in the case of ovarian for

33:12

a long time. And because I was

33:14

a safety person, I was really

33:16

trying to understand and put into context

33:18

all these safety issues so it

33:20

was possible to safely treat patients with

33:22

these drugs. It had already

33:25

been approved. It had been approved for ovarian

33:27

cancer when I showed up. So now

33:29

you're doing post -marketing surveillance on safety. We're

33:31

doing post -marketing surveillance on ovarian and putting

33:33

together a U .S. submission and a European

33:35

submission for cancer. For breast. So I started

33:37

talking to the statisticians there about how

33:39

I wanted them to program to get the

33:41

data we needed for the safety label.

33:44

And I'll never forget the guy looking at

33:46

me and saying, Do you know how

33:48

to do this?" I said, well, I had

33:50

to learn in Uganda because I didn't

33:52

have somebody like you, you know. So

33:54

it was sort of funny that

33:56

I was very happy to prove

33:58

myself. It didn't bug me. It

34:00

made me more feisty. Like, I'll

34:03

show you, I'm not underdosed in

34:05

the kinds of things you need

34:07

to do in this place. And

34:09

by the way, I loved every

34:11

minute of being at Bristol -Marsquibb.

34:13

They were pros at cancer drug

34:15

development. They were pros. at monitoring

34:17

safety. And I thought

34:19

it was so much fun because

34:21

you got to make drugs. What

34:24

was the pharma landscape like in the

34:26

early 90s? So you had Bristol -Meyer, you

34:28

had Pfizer, you had Merck. You had Merck,

34:30

you had Novartis. I think Novartis was

34:32

a combo of a couple. It

34:34

was smaller, much smaller and

34:36

cancer was Bristol -Meyer script. I

34:38

mean, they had made cisplatinum, carboplatinum.

34:40

They had a lot of those drugs

34:42

and people who had made those

34:45

drugs were still there. And I was

34:47

really happy to learn from them.

34:49

I felt very lucky to get to

34:51

be around these folks who knew

34:53

about Taxol. So we got Taxol approved

34:55

in the U .S. and in Europe

34:57

for breast cancer, it became Pristamar

34:59

Squibb's number one drug. I became the

35:01

project team leader for Taxol. How

35:03

long did it take them to thank

35:05

your husband for forcing them to

35:07

bring you along? Too

35:10

long. He

35:12

really enjoys that story because he, like

35:14

all good family stories, he gets embellished

35:17

over the years and he tells the

35:19

story like, you should actually want her.

35:21

You don't know this, but he's a

35:23

good husband. When you pause at where

35:25

we are in this story to think

35:27

of everything that would come from this

35:29

moment forward and to realize there's a

35:31

scenario under which nobody knows everything that's

35:34

about to happen and you're an oncologist

35:36

in Kentucky right now. That's right. That's

35:38

right. I'd be better at tennis. Yeah.

35:40

I had more, I had

35:42

more free time. But

35:44

yeah, no, I think that's the

35:46

thing that I love to mentor. I

35:49

think it's really underrated to listen

35:51

to students and hear what's on their

35:53

minds. And I remind students about

35:55

the role of serendipity. And I think

35:57

I'm a poster child for the

35:59

role of serendipity. So you

36:01

left Bristol Myers Squibb in 95.

36:03

95. And you went to Genentech.

36:05

Tell me where Genentech was in

36:07

its life cycle then. So Genentech

36:10

had been around for a while. I

36:12

mean, when did Genentech get founded in

36:14

the mid -70s? 1976. Okay. Give folks a

36:17

little bit of a history of Genentech. Genentech's

36:19

a storied company, but also a different

36:21

company in that it was founded on a

36:23

new technology. Genentech's a

36:25

really interesting company because it claims

36:27

to be the first biotech

36:29

company. There's some cedas back and

36:31

forth about that, but it

36:33

was based on genetic technology. That's

36:35

where the Genentech name came

36:37

from. And what Herb

36:40

and Bob, the co -founders of Genentech,

36:42

wanted to do is kind of

36:44

do a proof of concept that

36:46

you could use genetic technology and

36:48

make medicines, make big

36:50

medicines, proteins, antibodies, medicines

36:52

that would almost certainly have to be

36:54

injected rather than swallowed because they're

36:56

large and they're protein, so you break

36:58

them down if you swallow them. But

37:01

their initial goals were focused

37:03

on insulin. which they out

37:05

licensed to Lily and Pfizer,

37:08

and growth hormone, human growth hormone. Before

37:11

Genentech, when you were a

37:13

parent and your child was

37:15

short, you needed to

37:17

give that child growth hormone that

37:19

came from cadavers. That had

37:21

a risk of this slow virus

37:23

disease, and that was not a

37:26

good trade -off for parents. The

37:28

concept of having recombinant, of

37:31

having human -like growth hormone

37:33

was a really wonderful thing.

37:35

So Genentech's first drug was

37:37

human growth hormone. And it

37:39

was a tour de force. It

37:41

was really amazing that in the

37:43

late 70s, they were able to

37:45

do this fermentation and purification because

37:47

they had to prove to FDA

37:50

it was pure human growth hormone

37:52

with no contaminants. And it became

37:54

famous for that. And people were

37:56

excited and thought this was cool.

37:58

Tell people briefly how this worked.

38:00

What was recombinant DNA technology? What

38:02

were they putting the gene into? How

38:05

did they get the gene to make

38:07

the protein? We take this all for

38:09

granted today because we have... It's tricky.

38:12

Yeah. Yeah, but it's so incredible. So

38:14

what you see if you go to

38:16

Genentech or a company like Genentech is

38:18

you see these tanks and the tanks

38:21

are like a cell ICU, like an

38:23

ICU for a cell. So

38:25

the cell, what you're doing is

38:27

you're teaching the cell to make

38:29

it very high amounts growth hormone

38:31

way higher than your cells or

38:33

my cells would. And then you're

38:35

teaching the cell through this genetic

38:37

engineering to secrete it into a

38:39

medium, into this soup that is

38:41

really a lot of growth hormone.

38:43

And then you're taking away the

38:45

cells after they secrete it. You're

38:48

purifying that growth hormone and you

38:50

put in little vials. And that's

38:52

the process of biotechnology. And you

38:54

can trick a cell into making

38:56

almost anything you want, not completely,

38:58

but almost anything you want and

39:00

make it very much like human,

39:02

which is neat. So you don't

39:04

have to go to a human

39:06

to donate you with hormone because

39:08

it would be too small. Or

39:10

in the case of insulin, I

39:13

mean, they were using insulin from

39:15

pigs and - Poursine insulin that made

39:17

allergies and expensive. So human

39:19

insulin really changed how you thought

39:21

about treating people who have diabetes.

39:24

So Genentech made growth hormone, and Genentech

39:26

sold growth hormone and set up

39:28

something I actually think is a really

39:30

neat thing that Genentech did, which

39:32

people said, how do you know that

39:35

by giving kids extra growth hormone,

39:37

it won't cause leukemia or a fourth

39:39

arm to sprout out or, you

39:41

know, weird things to happen? And

39:43

Genentech said, well, we'll follow every child.

39:46

So they set up a patient

39:48

registry, one of the first patient

39:50

registries ever, and followed every child

39:52

until they reached their final adult

39:54

height. And the physicians and their

39:56

staff enter this into a computer. And

39:59

so this is an amazing amount of

40:01

information. So if the FDA ever asked

40:03

us, do you have this with growth

40:05

hormone? Did you have that with growth

40:07

hormone? We had not an example. You

40:09

wouldn't even do statistics on it. Every

40:11

child ever treated with genetic growth hormone.

40:13

Do you have a sense of how

40:15

many kids that was? Oh, hundreds of

40:17

kids, thousands of kids by now. Yeah,

40:19

yeah. So Genentech got really good at

40:21

that. And in fact, when I went

40:23

to Genentech in 1995, the

40:25

chief medical officer of Genentech was

40:28

a pediatric endocrinologist, an expert on short

40:30

stature and growth hormone. But

40:32

it's a pretty small market. This is

40:34

uncommon. By this point, it was

40:36

being used rampantly in sports. Yeah.

40:38

And the FDA was not happy about that

40:40

and pushed really hard on Genentech to

40:42

control that use. Was it being used by

40:45

this point also pretty heavily in HIV,

40:47

right? People were using it in HIV, they

40:49

were using it in sports, anything where

40:51

you wanted to have more muscle mass. That's

40:54

exactly right. Genetic had done some studies

40:56

to look at whether that was a

40:58

good idea and none of the studies

41:00

came out successful. Meaning there

41:02

was no benefit to an HIV

41:04

patient being on growth hormone? The

41:06

benefits did not outweigh risks of

41:08

having increased blood sugar and some

41:10

other things that would happen. So

41:12

one of the aspects of Genentech

41:14

that happened in the early years

41:16

before I was there is they

41:19

learned how to make enzymes. Same

41:21

genetic technology. Childing the cell make

41:23

these enzymes. And some of

41:25

the enzymes actually got out

41:27

licensed to make commercial enzymes like

41:29

that you use when you

41:31

wash clothes and things like that.

41:33

So that wasn't core to

41:35

Genentech. But they had an enzyme

41:37

activase, a TPA, tissue pledge

41:39

menogen activator that could break down blood

41:41

clots. Did they go after that

41:43

knowing what they were doing or was

41:45

this a bit of a fishing expedition

41:47

where they realized in the process

41:49

of trying to do many things that,

41:51

oh my god, we can actually make

41:53

TPA, which you're going to explain

41:55

in a minute why that changed the

41:58

game of cardiovascular medicine. It was intentional

42:00

that a really great... There's a

42:02

clinician researcher, Dave Stump, who

42:04

is a clotting expert. He's

42:06

he -monk on the heme side. He was

42:08

there and really pushed them to do

42:10

this. And the concept was that

42:12

if you could break down the blood

42:14

clot, you could cure the heart attack.

42:16

You could save lives. And

42:18

the interesting thing, if you are interested

42:20

in doing trials, is they started

42:22

the concept of a large, simple trial.

42:25

This was early on, and people

42:27

in cardiovascular gene Brownwald and his followers

42:29

had started these large, simple trials.

42:31

So, genetic kind of bet the farm

42:33

on this TPA. And the farm

42:35

was that they could change the outcomes

42:38

in 30 days. There'd be more

42:40

people alive than dead if they were

42:42

treated with Activase. And two of

42:44

the people involved in the studies, you

42:46

probably know their names, were Rob

42:48

Califf and Eric Topol. Was Eric at

42:50

Scripps at that time or where

42:53

was he? I think he was at

42:55

the Cleveland Clinic. And

42:57

they ran a group called the Timmy Group

42:59

and they did all these studies named Timmy. And

43:01

so they did this big trial and it

43:03

worked. If you treated with

43:05

Activase, you could break down the

43:07

blood clots. So Genentech

43:09

started this franchise in Cardiovascular, and

43:12

again, did this really interesting patient

43:14

registry to look at 30 -day

43:16

outcomes for post -marketing, but stents came

43:18

along. And so the franchise

43:20

of Genentech and people who were

43:22

treated with TPA really went down. And

43:24

the stroke indication was tricky. You

43:26

had to make sure it wasn't hemorrhagic

43:29

or you could make things so

43:31

much worse. Yeah. Stroke

43:33

indication on paper was really

43:35

cool, but pragmatically was really tough

43:37

for hospitals to execute. Genotech

43:39

also made another enzyme

43:42

-like molecule, DNAs, palmosyme,

43:44

for cystic fibrosis. And

43:46

that was approved very

43:48

tiny. It decreases how

43:51

thick your secretions are, but

43:53

with vertexes, CF drugs, it's

43:55

also been scooped. So when

43:57

I came in 95, Genotech

43:59

was really struggling. They had those

44:01

three drugs. They had growth hormone,

44:04

TPA, and Palmazyme. Why did

44:06

they out -license insulin? I think they needed the

44:08

money. Got it. I don't think they have

44:10

the scope to even it. And why did you

44:12

decide to leave? Bristol Meyer Squibb

44:14

is just crushing it. You finally earned

44:16

the respect you deserve. You've

44:18

got this struggling company,

44:21

Genentech. Was it the opportunity?

44:24

Oh, yeah. Yeah. For sure it was the opportunity. I

44:26

will tell you, if you were

44:28

me in 1995, sitting down

44:30

with Art Levinson, And they had

44:32

a research then. And he was talking

44:34

about the future in oncology, what the

44:36

plans were, you'd have gone to.

44:38

You'd have gone to. For sure it

44:41

was an opportunity. We were

44:43

doing well. I will say that further

44:45

down on the list of pros

44:47

and cons was West Coast's home. I

44:49

mean, Connecticut was snowy and cold. I

44:52

didn't come over on the Mayflower, it turns out.

44:55

I mean, I loved people at Bristol -Miresquibb. I

44:57

loved the job. But heading back to San

44:59

Francisco. in San Francisco. Yeah, that was a

45:01

big deal. But I believed Art when he

45:03

said we're going to be a cancer company.

45:05

So what was the first thing you worked on?

45:08

Ah, thrombopoetin. They

45:10

hired me to work on thrombopoetin. It

45:13

was going to be the third leg

45:15

of the stool, EPO. So make

45:17

your red cells go up, nupogen

45:19

for your white cells and TPO

45:21

for your platelets. And it was

45:23

a big race. Amgen

45:25

was in the race. Who developed EPO?

45:27

Amgen. Amgen, okay. EPO and nupogen.

45:29

Got it. Amgen and Genentech had always

45:32

been kind of rivals. And

45:34

when they cloned Trump a poet in

45:36

that Genentech, I read the paper and then

45:38

they called me, did I want to

45:40

come work on it? It was that kind

45:42

of thing. When you clone it, if

45:45

you publish it, that doesn't give you the

45:47

right, it's a race for everybody. It's

45:49

a race. So let me ask a silly

45:51

question. Why do you publish the results

45:53

of the cloning before you've gone and made

45:55

the recombinant protein yourself? So you patent

45:57

it, then you publish. then you make the

46:00

recombinant. Okay. So once it's published after

46:02

the patent, you get to make it. Genentech,

46:05

one of their great assets

46:07

started by her boyer is

46:09

they publish. They don't stop

46:11

the scientists from publishing. They get the lawyers and

46:13

they're quick and they make sure that they protect

46:15

the IP of the company, but they want people

46:17

to publish. It's very academic. So thrombopoietons,

46:19

Epo and Newpagen are as

46:21

if you design them. to make

46:23

recombinant forms and give them

46:25

for cancer patients or other patients

46:27

who need them. Thrombopoietin,

46:29

not so much. To make

46:32

it simple, if you said, okay,

46:34

your platelets are going way down

46:36

from your chemotherapy, and I'm

46:38

going to give you thrombopoietin to make them go

46:41

up, they come back

46:43

up really late and they

46:45

go too high. So I'm

46:47

making you at risk for a blood

46:49

clot by giving you a million platelets.

46:51

but later than you need to when

46:53

you're recovering on your own already. Was

46:55

that known only once you started developing

46:57

and you understood the kinetics of it?

46:59

Once we looked at how it worked

47:01

in patients, we knew better than we

47:03

had before. The kinetics of

47:05

really recovering from not all chemotherapy, as

47:07

you know, causes your platelets to go

47:09

low. Yeah, so you can't give it

47:11

prophylactically because you don't know who's going

47:13

to get thrombocytopenia. Right, right. And this

47:15

tricky thing about going too high if

47:17

you're wrong, it's a problem. What do

47:19

you do? You plasma -ferrice the patients if

47:21

you've overshot or not plasma -ferrice, platelet -ferrice.

47:23

You could platelet -ferrice them. I mean,

47:25

there are remedies, but you don't want

47:27

to do that. Yeah, that's it. Anyway,

47:29

so thrombopoietin proved to be very, very

47:31

difficult drug. And I learned

47:33

a lot about the cancer

47:36

equivalent or the product development

47:38

equivalent of tulip mania. When

47:41

everybody's so excited, you get

47:43

excited too. And it's like,

47:45

oh, I did learn a

47:47

lot. I've often reflected on

47:49

what might happen that I'm

47:51

not thinking about now. But

47:53

what also happened is the

47:55

labs at Genentech had been

47:57

working on Herceptin on Trestuzumab

47:59

for a while. Art

48:01

became the CEO. Art Levinson

48:03

became the CEO in 95. And

48:05

he wanted to push on having

48:07

Trestuzumab Herceptin get into the clinic.

48:09

Tell folks how that drug worked,

48:12

what it was for. Trestuzumab,

48:14

or I'll call it Herceptin because it's

48:16

less of a mouthful, is a antibody.

48:18

Like you and I have antibodies that

48:20

fight disease in our bodies. And

48:23

it's an antibody that

48:25

targets this protein called

48:27

HER2. And HER2 matters

48:30

because about one in four women

48:32

with breast cancer have too much of

48:34

it. And when you have too

48:36

much of it, if you've got too

48:38

much HER2 from the time you're

48:40

diagnosed, Your median

48:42

average survival is three

48:44

years. If you don't have

48:47

too much of it, it's seven years. So

48:49

you know what matters. So the concept with

48:51

this antibody is turn that off. Whatever

48:53

bad thing that drives it down to

48:55

three years, turn that off and go

48:57

back to seven years. Pretty simple concept.

48:59

Why do you mechanistically do you think

49:01

that the over expression of HER2 was

49:03

impeding immune clearance? What was the thesis

49:06

at the time for why overexpression of

49:08

her two was cutting life expectancy down?

49:10

The thesis was that it was telling

49:12

the cell to grow. That

49:14

it was getting a growth

49:16

signal to the nucleus to say

49:18

grow more. And if you could

49:20

shut that off, you'd grow less. Now,

49:22

later, we armed her

49:24

septum. We put a payload on it.

49:26

So then you could say, both change

49:29

the grow more signal and you've got

49:31

a little bomb on there. Yeah, kill

49:33

signal. Yeah. So you'd get a twofer.

49:35

In fact, another company has

49:37

one AstraZeneca that's so powerful with

49:39

a bystander effect, you don't

49:41

even have to have overexpression. So

49:44

that antibody, I'm talking

49:46

about it now because anybody in breast cancer

49:48

knows about her too. Of course, yeah.

49:50

You see it on TV and direct to

49:52

consumer ads. Yeah, but I think what

49:54

I really enjoy about this type of discussion

49:57

though, Sue, is One, it's the story

49:59

of your career, but it's also the story

50:01

of oncology. It is, yeah. It's the

50:03

story of modern oncology. So you're one of

50:05

the few people whose careers takes us

50:07

through the walk of modern oncology. I mentioned

50:09

that because it seems impossible. There

50:11

was a lot of people

50:14

at Genentech who were negative

50:16

about her septum, did not

50:18

think we should invest. why.

50:20

The dogma was that antibodies

50:22

were all hype. They'd been

50:24

overpromised. as smart bombs, smart

50:26

missiles, Time Magazine, all

50:28

of this, but that they had

50:30

flopped. What was the biggest failure

50:32

at that point commercially? I

50:34

don't think the things had even been commercialized.

50:36

I don't think they had gotten out of

50:39

the clinic, that people just weren't seeing benefit. I

50:41

have a very good friend who's an

50:43

oncologist, and he said, you

50:45

just can't treat a solid tumor,

50:48

a solid tumor versus leukemia or lymphoma

50:50

with an antibody. you need something

50:52

more powerful. And remember what was happening

50:54

at the same time is, if

50:56

you want to talk about the history

50:58

of oncology, the amazing

51:00

thing is being at

51:02

ASCO, the American Society of

51:04

Clinical Oncology, and two

51:07

different rooms. One room, we

51:09

hear that her septum is going to

51:11

change forever how we think about antibodies and

51:13

breast cancer. It way better

51:15

than we thought, improved survival. The

51:17

other room doing bone

51:19

marrow transplant for breast

51:21

cancer and having the South African

51:23

group who published a paper saying it

51:25

worked, retract the paper and go

51:27

through and talk about how much of

51:29

it was fraudulent, fraud, fraud, fraud. So

51:32

at the same time, this

51:35

nearly toxic, nearly lethal bone

51:37

marrow transplant for breast cancer was

51:39

debunked at the same time

51:41

as we said what we call

51:43

now a naked antibody, no

51:45

payload, no chemo. You give her

51:47

septum, you're going to help

51:49

that patient with breast cancer. Just

51:51

an antibody. Welcome to modern

51:53

oncology. It could not have been

51:55

more clear. I almost

51:58

forgot the BMT stuff.

52:00

It's so archaic. And

52:02

it was a distraction because

52:04

people felt like you just

52:06

needed to hit the cancer

52:08

hard. You just needed to

52:10

hit the cancer smart. The

52:12

tarred wasn't the point. Yeah. I mean, we've

52:14

got to be getting close to a Vastin now

52:17

too, right? Yes, but don't

52:19

forget Ritux. Yes. So when

52:21

I said people didn't believe in

52:23

you could treat a solid

52:25

tumor, they thought you could treat

52:27

lymphoma because we did. Antibodies

52:30

were so disliked. People

52:32

did not believe in

52:34

antibodies that in 95,

52:36

96, IDEC was

52:38

going to run out of money.

52:40

So iDEC had made an

52:42

antibody to CD20, a very important

52:44

marker on all lymphomas. And

52:47

they were going to run out of money.

52:50

So some of our business development

52:52

folks talked to them about

52:54

Genentech doing a deal with them

52:56

on Rotexan. It is impossible

52:59

to overstate how important Rotexan is

53:01

in lymphoma. I often think

53:03

when I'm in product development of

53:05

patients I've cared for, I

53:07

had this great 83 -year -old pharmacist

53:10

when I was in practice

53:12

and he had a lymphoma that

53:14

was low grade, a little

53:16

tired, he was fine. And

53:19

so we did watch and wait.

53:21

My not favorite strategy of oncology.

53:23

Let's watch and wait as you

53:25

dwindle. He'd be a

53:27

perfect candidate for Rotexan.

53:29

Four doses, you can repeat

53:31

it. In fact, it works so

53:33

well. Here's when I changed

53:35

my mind on antibodies. Somebody runs

53:37

in my office and said, oh, we

53:40

have a case of tumor lysis syndrome. So

53:42

tumor lysis syndrome being somebody got

53:44

rituxand. They had a lot of lymphoma

53:47

and the cells are breaking down

53:49

so fast their kidneys can't keep up

53:51

and they have to be dialyzed.

53:53

Oh, that's only an antibody. No

53:55

chemo, no payload, nothing. That's

53:58

when you know you've got a good

54:00

drug. How many cells in the body, how

54:02

many types of cells in the body

54:04

express CD20? It's mainly a B cell. But

54:06

it's not as specific as CD19, is

54:08

it? I think 19 and 20 are both

54:10

B cells. I'd have to look at

54:12

it. I know CD19 is on the B

54:14

cell, but I didn't know the... Okay.

54:16

Yeah. We were talking about this earlier. This

54:18

is chimeric. It's chimeric, yeah. Yeah, so

54:20

tell folks what that implies, because that's another

54:22

wrinkle in the story. Well, that's the

54:24

other thing that I think is... There's so

54:26

many dogmas that we believe until data

54:28

proves otherwise. So one of the warts of

54:31

Rottuxen was thought that it was a

54:33

chimera. It had too many mouse parts to

54:35

human parts and that we would cause

54:37

human anti -chimeric antibody, HACA. And FDA was

54:39

very concerned about this. So we measured and

54:41

measured and measured. And it

54:43

turned out probably because the patients had

54:45

lymphoma that they didn't get HACA. Very

54:47

tiny numbers and they not clinically relevant.

54:49

And you can treat them a lot.

54:51

You can treat patients over and over

54:53

again. Perceptin's 93 % human,

54:55

so not a chimera, but

54:57

not fully humanized. And none of

54:59

the patients treated with antibodies

55:02

that I've seen, not with genetic

55:04

or idic antibodies, have really

55:06

had problems. They've had

55:08

other problems based on target related

55:10

problems, not based on the antibody.

55:12

And the CD20 antibody was just

55:14

also a straight naked antibody? Straight

55:16

naked antibody. And targeted for an

55:18

immune destruction? It's targeted to destroy

55:20

the CD20 positive cells, but it

55:22

turns out that if you have

55:24

lymphoma, you've got a reserve in

55:27

your marrow of other CD20 cells

55:29

of more immature that grow up

55:31

and replace. So it's not like

55:33

you're really at a huge risk

55:35

for untoward reactions from your CD20.

55:37

And why do those patients with

55:39

marrow that's still producing CD20 positive

55:41

cells not go on in a

55:43

constant state of lymphoma requiring.

55:45

In other words, why is it that you can

55:48

treat this and create a durable remission? I

55:50

can guess. I don't know if anyone's

55:52

done a formal study. I do think

55:54

at some point in oncology treatment, if

55:56

you have a tiny amount of disease

55:58

left, especially something like lymphoma, you may

56:01

take care of it yourself. So

56:03

it's just getting rid of enough of

56:05

the diseased B cells until you get

56:07

the load down low enough that the

56:09

immune system can wipe out the clone.

56:11

And honestly, if it comes back, That's

56:13

the other thing that I find really

56:15

interesting about antibodies. The

56:17

dogma with chemotherapy, if

56:19

you are on taxol and your tumor comes

56:21

back, I wouldn't give you taxol again. If

56:23

you're on Herceptin or Rituxin and your tumor

56:25

comes back, in a heartbeat, I'd give it

56:28

to you again. Yeah, it's a very different

56:30

thing than chemotherapy. What was

56:32

the price of these drugs at the

56:34

time they came out? Were these the

56:36

first chemotherapeutic agents or you kind of

56:38

want to distinguish them from traditional chemo,

56:40

but were these the first oncology drugs

56:42

that came with big price tags? Probably

56:44

they were. I think Taxol kind

56:46

of went to that next level

56:48

and then they went to the further

56:50

level compared to today's prices low. But

56:53

Rituxin, I think more than

56:55

Herceptin, because people started using it

56:57

more like you'd use four

56:59

times or eight times and recurring,

57:02

Rituxin sales went very high, very

57:04

fast. And about this time,

57:06

we get the whole anti -vegev story,

57:08

right? Yeah. Yeah, that was... So, Judah

57:10

Folkman over at Boston Children's. Yeah, I

57:12

never had the chance to meet him.

57:14

Oh, you didn't him? I've never met

57:16

Judah. He wrote a fantastic book that

57:19

I read in medical school, poured over

57:21

the book. I'm blanking on the name

57:23

of it. Do you remember the book?

57:25

It was his story. Yeah, I don't

57:27

remember. Again, a beautiful story. I'm smiling

57:29

because I had a word for Judah

57:31

Folkman talks, which I heard many of.

57:33

He was a phenomenologist. He

57:35

would say, this patient had this, so it

57:37

must mean that. He just connected dots all

57:39

the time. I mean, some of it made

57:41

no sense to me, but some of it

57:43

was like, wow, I wish I thought of

57:45

that. He was just a really fun person

57:47

to listen to. I used to

57:50

tell this story so many times. His

57:52

thing was the cancer can't grow larger

57:54

than a BB if it doesn't have

57:56

new blood vessels. That was

57:58

his thing. It stuck at

58:00

a certain size. And so VEGF

58:02

is the primary way, vascular

58:04

endothelial growth factor, is the primary

58:06

way that you grow new

58:09

blood vessels if you're a tumor

58:11

cell. People went crazy about

58:13

this hypothesis. It was more

58:15

than the TPO that I was describing.

58:18

Because it was due to fulcrum and

58:20

he's very compelling and very charismatic. And

58:22

just because the hypothesis It's

58:24

logical. It's logical. It resonates. It

58:26

sounds like a good thing. You

58:29

know, the saying that I love

58:31

is, it's the description of science

58:33

as a beautiful, compelling hypothesis slayed

58:35

by an ugly fact. That's

58:39

perfect. Yeah. Napa

58:41

Ferrara, also a great

58:43

character. Italian OBGYN, who

58:46

came to Genentech and worked

58:48

in one of the labs,

58:50

made an antibody, actually same

58:52

backbone as Herceptin, to VEGF.

58:54

Mostly human. mostly human again,

58:56

I think about 93, 94 %

58:58

human. And so

59:00

we decided we should go

59:02

after an antibody for VEGF

59:05

as our next big oncology

59:07

program. I still remember, by

59:09

the way, one of the things

59:11

that Gwen Fife, who's a great

59:13

clinical oncology trials expert, was in

59:15

charge of the program. And

59:17

Gwen and I talked the day before

59:19

the first patient was going to get treated

59:21

with anti -VEGF. And Glenn said, my nightmare

59:23

is that all the blood vessels fall

59:25

apart. We've just put that into

59:27

someone's body. And I said, well, you know, we

59:29

did all the talk studies. We're like, I

59:31

don't think it's going to be that bad, but

59:33

we have no idea what. I mean, it

59:35

just felt. And I'm sorry, this was before the

59:37

first phase one patient. This was the first

59:39

phase This was the first phase one. So you're

59:41

going very low dose. We're going to dose.

59:43

Very low. is the first time it's going into

59:45

a human. the first human dosing of anti -VegF,

59:47

and we knew how important VegF was. So

59:50

we were scared. And this is also mid -90s.

59:52

Yeah. Yeah. This is all happening when you arrive.

59:54

Yeah, they just got there. I

59:56

mean, what a time to be a Genentech.

59:58

Yeah, it was wild. It was wild. Wow.

1:00:00

So we're into the clinic and we

1:00:02

make progress and it's really good news. and

1:00:04

lots of studies and we're ambitious. We want

1:00:06

to do a lot of different, we wanted

1:00:08

to do lung cancer and we ought

1:00:10

to do breast cancer and... How are you

1:00:12

picking the cancer to study something like this?

1:00:14

Herceptin's obvious because you're targeting a receptor. Herceptin

1:00:16

and retoxin were easy. You know what you're

1:00:19

doing but here you could be targeting

1:00:21

anything. There is one tumor

1:00:23

where VEGF plays a seminal role

1:00:25

and that is renal cell carcinoma. And

1:00:28

yet renal cells are really tough

1:00:30

to study. It's just,

1:00:32

it's not set up clinical trials

1:00:34

wise. Why is that? You've

1:00:36

got the IL2 stuff going on where

1:00:39

you've got 10 % of people will respond

1:00:41

to it, but 90 % won't. It should

1:00:43

be easier. It may be the sites and

1:00:45

where the clinicians are who care for

1:00:47

it. It may just be that pragmatic. We

1:00:50

kept struggling to figure out how

1:00:52

to do a good renal cell

1:00:54

study. But we thought we could

1:00:56

do a breast cancer study because

1:00:58

we had a lot of networks

1:01:00

of breast cancer patients and particularly

1:01:02

patients who weren't eligible for herceptin

1:01:04

because many of them weren't. So

1:01:06

we wanted to do a late

1:01:08

-stage breast cancer study because if

1:01:10

we could help these patients, we

1:01:12

would find out right away. These

1:01:14

were patients with metastatic breast cancer?

1:01:17

Metastatic breast cancer who had already

1:01:19

progressed through everything. Everything. So really

1:01:21

tough high bar. And the standard

1:01:23

you're going to hold yourself to

1:01:25

in the phase two is 50 %

1:01:27

shrinkage. We wanted to have

1:01:29

at least 50 % shrinkage. We wanted to

1:01:31

change time progression. This is a great

1:01:33

time to actually hit pause. I wanted

1:01:35

to do this later, but I think

1:01:37

this is the right drug to go

1:01:39

through two things. One, even though

1:01:41

I've done this probably half a dozen

1:01:43

times on the podcast, you should never assume

1:01:45

somebody remembers it. I want people to

1:01:47

understand what the difference is between a phase

1:01:49

one, a phase two, a phase three

1:01:51

study, also understand what's preclinical. It's not

1:01:54

intuitive to people why it costs a billion dollars

1:01:56

to get a drug to market and why it can

1:01:58

take a decade. And then

1:02:00

within that, if you could just

1:02:02

embed enough of the details

1:02:04

about decisions that you can make

1:02:06

that will make or break

1:02:08

you, how many times has a

1:02:10

drug failed because the experimental

1:02:12

design, the wrong patient

1:02:14

selection, the wrong disease selection? You

1:02:16

have got to line up four

1:02:18

pieces of Swiss cheese just right

1:02:20

to get the pen through to

1:02:22

hit it. Sorry to interrupt, but

1:02:24

let's go back good.

1:02:27

You got Judah Folkman talking about Vegev,

1:02:29

Vegev, Vegev, that then turns into,

1:02:31

well, if we made an antibody to

1:02:33

Vegev, okay, so there's your idea. Now

1:02:36

start the clock and start

1:02:38

the dollars. So if you start

1:02:40

with a target, often in oncology

1:02:42

today, we'll start with a target. There's

1:02:45

two things you have to

1:02:47

start with. One is, what's

1:02:49

the best way to turn down

1:02:51

or turn off that target? Is

1:02:53

it a small molecule? Is it an

1:02:56

antibody? Tell folks the difference. How

1:02:58

do you think of small molecule versus antibody? Where

1:03:00

do we draw the line? So here's a really

1:03:02

simple way I think of that helps me. Small

1:03:04

molecule is chemistry. Small molecule,

1:03:06

it can be, not always,

1:03:08

a pill. A small molecule

1:03:10

you're impacting on often pathways

1:03:13

or enzymes or things that

1:03:15

happen in the cell. A

1:03:17

large molecule, whether it's a

1:03:19

protein or especially an antibody, an

1:03:22

antibody's biology. An antibody you're

1:03:24

trying to do something that may

1:03:26

be immune in nature or

1:03:29

you use the antibody as a

1:03:31

delivery device. You're getting something

1:03:33

to the cell. A company

1:03:35

like Genentech and many modern

1:03:37

companies really like antibodies. I like

1:03:39

antibodies because when something happens,

1:03:42

it's on target. It doesn't tend

1:03:44

to be off target. Small

1:03:46

molecules at chemistry tends to have

1:03:48

surprises in negative ways off

1:03:50

target, like liver toxicity or kidney

1:03:52

toxicity. I do this through

1:03:55

cardiovascular medicine to explain to people

1:03:57

the difference between a statin

1:03:59

and a PCSK9 inhibitor. You have

1:04:01

these two very common drugs

1:04:03

that are used to lower cholesterol,

1:04:06

but a statin is a small molecule.

1:04:08

I don't say this in an insulting

1:04:11

way, but we use the terminology it's

1:04:13

dirty. It does block an enzyme, but

1:04:15

it's got all these off -target things and

1:04:17

your liver function gets whacked, you get

1:04:19

insulin resistance, some people get horrible muscle

1:04:21

soreness. So five to 10

1:04:23

% of people taking this drug are

1:04:25

going to have a side effect

1:04:27

that prevents them from taking the drug.

1:04:29

I've never seen a person yet

1:04:32

who couldn't tolerate a PCSK9 inhibitor where

1:04:34

you inject an antibody into them

1:04:36

that binds to a protein and shuts

1:04:38

it off. That's really a good

1:04:40

example. And the choice of molecule is

1:04:42

driven by that. When I

1:04:44

was first in product development, there was this

1:04:46

thing of, oh, you need a pill,

1:04:48

especially for chronic indications. You need a pill

1:04:50

for compliance. Right. Who would take an

1:04:53

injection for cholesterol? Look at obesity drugs. Turns

1:04:55

out a lot of people would take

1:04:57

an injection if they want to. But once

1:04:59

you have your selection, you

1:05:01

need to make sure you can make

1:05:03

it. And one of the critical

1:05:05

things for a biotech company, if it's

1:05:07

a protein or an antibody, is

1:05:09

the small scale production of it. In

1:05:12

small, they call it a mini -firm.

1:05:14

The mini -firm has to resemble what is actually

1:05:16

going to be used, because the next thing

1:05:18

you start doing is a bunch of models.

1:05:21

Judah Folkman, giving a great talk, doesn't

1:05:23

mean you believe that blocking VEGF

1:05:25

will help cancer. So we

1:05:27

do models in mice. We may

1:05:29

do larger. animals. We do fewer

1:05:31

animal models than we used to

1:05:34

because they're really limited. I would

1:05:36

rather have a great target with

1:05:38

good biology than an animal model,

1:05:40

but it's still helpful. It's still

1:05:42

helpful. And then the critical

1:05:44

thing is the preclinical work that

1:05:46

you do, what FDA is going

1:05:48

to want to ask you when they should,

1:05:50

this is not them being bad, this is

1:05:52

them being good, they're going to want to

1:05:54

ask you about toxicology. What's your safety plan?

1:05:57

Based on biology of VEGF, what are you

1:05:59

most worried about? I'm most worried about bleeding.

1:06:01

It's an antibody. I'm most worried about an

1:06:03

allergy to the antibody. Did any of the

1:06:05

tox studies show allergy to the antibody? What

1:06:08

are you going to look for and how

1:06:10

are you going to look? How often are you

1:06:12

going to measure the patient? So the preclinical

1:06:14

safety plan is really important and based on what

1:06:16

you find in toxicology. The

1:06:18

other thing that's essential is, and

1:06:20

especially modern oncology, if you

1:06:22

have a targeted therapy, you must

1:06:25

have a diagnostic. And that

1:06:27

is wicked hard because you've got

1:06:29

the therapeutic and the diagnostic

1:06:31

at the same time. Now,

1:06:33

things like CD20, things

1:06:35

like VEGF are very ubiquitous, so

1:06:37

it's not really targeted in the sense

1:06:39

of HER2, where we needed a

1:06:41

diagnostic. But if you need to have

1:06:43

that, we had what we call

1:06:45

the clinical trials assay for Herceptin that

1:06:47

wasn't to be marketed. Did you

1:06:50

guys have to have somebody in parallel

1:06:52

developing a CLIA certified assay? that

1:06:54

a pathologist was going to use, or

1:06:56

did you do that in -house? So

1:06:58

we had an in -house clinical trials

1:07:00

assay that we used all the

1:07:02

way through phase three. And you could

1:07:04

quality control the hell out of

1:07:06

it? It was fine. It was nothing

1:07:08

wrong with it, except it wasn't

1:07:10

approved. So not fine. So

1:07:12

when we went to FDA, they said we're

1:07:14

not approving her sept until you have an

1:07:17

approved diagnostic. So whose responsibility is like, how

1:07:19

do you encourage the world to make that

1:07:21

happen? So we went to DACO. We went

1:07:23

to several diagnostic companies and DACO said, we'll

1:07:25

make you an immunohistochemistry test for her too.

1:07:27

And her sub -test is made by DACO.

1:07:29

But we had to go back and correlate

1:07:31

that with the clinical trials to make sure

1:07:33

that it was the same as the clinical

1:07:35

trials acid. Now, why didn't you guys do

1:07:37

that in parallel? Was the cost too great?

1:07:39

And did you want to de -risk the drug

1:07:41

before you sunk the cost into that? We

1:07:44

were... an experience to realize we should have.

1:07:46

I see. Okay, so nowadays we're doing that

1:07:48

in parallel. Oh, for sure on parallel. Yeah,

1:07:50

for sure. It was a mistake. So how

1:07:52

long, just to again go back to helping

1:07:54

people think through the arc of time, from

1:07:56

the moment you guys hit a go decision on,

1:07:59

we want to do this, we want to pursue

1:08:01

this path, how long until you file the IND?

1:08:03

Oh gosh. It could be

1:08:05

years. It could be two, three years

1:08:07

because you're doing animal models. Maybe

1:08:10

tell folks what the IND is so

1:08:12

they understand why that's an important

1:08:14

milestone. So the investigational new drug is

1:08:16

asking the Food and Drug Administration

1:08:18

permission to ship an unapproved drug across

1:08:20

state lines. If you and I

1:08:22

wanted to do something in Austin, we

1:08:24

could actually do it, which is sort of

1:08:26

weird when you think about it, but most people

1:08:29

don't really want to do that. Peter Sue

1:08:31

drug. It's going to be amazing. We're going to

1:08:33

set the lab up right over there. But

1:08:35

the moment we want to run a clinical trial.

1:08:37

And ship it. And get it out of

1:08:39

the state. You got to have the IND. So

1:08:41

the investigational new drug is the request. And

1:08:43

what happens is that you take all this information

1:08:46

I've been talking about, that you know you

1:08:48

have a molecule, you trust

1:08:50

the way you're producing the molecule,

1:08:52

you understand the biology enough,

1:08:54

you have a safety plan, and

1:08:56

you have a phase one

1:08:59

protocol. So phase one has one

1:09:01

purpose. We're all greedy. I've

1:09:03

been there. It is only

1:09:05

for safety. Phase one is is it

1:09:07

safe to give humans this molecule? Is it

1:09:09

safe to give it once? Is

1:09:11

it safe to give it multiple times?

1:09:13

And there's an art to knowing where to

1:09:15

start the dose because it's an escalating

1:09:18

dose trial. That's right. But you're extrapolating. from

1:09:20

what you learned about toxicity in a

1:09:22

totally different organism that never translates one to

1:09:24

one to the organism of choice. It's

1:09:26

absolutely true. And it's not uncommon. And you

1:09:28

see people all the time backing up

1:09:30

on the dose thinking, oh, that was more

1:09:32

than we needed or more than we

1:09:34

wanted. But phase one, with a

1:09:37

good preclinical package, a good IND,

1:09:39

phase one should be uneventful. And because

1:09:41

we are greedy in oncology, we

1:09:43

always look to see if anybody responds

1:09:45

just because... what we do. But

1:09:47

phase one often, to be fair, has

1:09:49

some really tough patients who are

1:09:51

trying something and have tried a lot

1:09:53

of other things. So the patient

1:09:55

population can be tough to find any

1:09:57

efficacy in. So phase one, I

1:09:59

always think of phase one as it

1:10:01

might be a year that you're

1:10:03

in phase one if you're doing really

1:10:05

good job. And typical cost given

1:10:07

the relatively low numbers of patients. Oh

1:10:10

gosh, in the tens

1:10:12

of millions? Tens of millions, yep. Yeah, tens of

1:10:14

millions. And then you get into the 20s

1:10:16

and 30s and 40s of millions with

1:10:18

the phase two, depending on how big

1:10:20

your phase two is. And phase two,

1:10:23

I think, that's where people can use

1:10:25

their intellect, I think in many ways.

1:10:27

Phase two, you start to look at

1:10:29

what's the right dose and schedule. Very,

1:10:31

very important to get the right dose

1:10:33

and schedule and... What's the right outcome?

1:10:35

What's the right patient? Who do you

1:10:37

want to treat? And really

1:10:39

what phase two is supposed to do,

1:10:42

with one exception, phase two is supposed to

1:10:44

get you ready for phase three. You've

1:10:46

got a dose, you've got a schedule, you've

1:10:48

got a patient selection criteria, and

1:10:50

you've got a hypothesis of where this is going

1:10:52

to be a drug. The exception in

1:10:54

oncology is sometimes you want to get

1:10:56

an approval off phase two. When we tested

1:10:58

anti -veg F in breast cancer in phase

1:11:00

two, we wanted to use that as

1:11:02

an approval study because we would go in

1:11:05

and say, look, it can be a

1:11:07

contingent approval, but these patients have nothing else

1:11:09

to do. And so I think that

1:11:11

can happen, especially targeted therapy. where you've got

1:11:13

the perfect target and FDA is feeling

1:11:15

good about it too. That can be a

1:11:17

phase two study. But most the time

1:11:20

you're getting ready for phase three. So

1:11:22

where were you guys with anti -vegetarian

1:11:24

phase two? You're at breast cancer

1:11:26

and did you do colon? We did

1:11:28

colon but not the kind of

1:11:30

study that I just mentioned for approval.

1:11:32

We did a traditional phase two

1:11:34

and colon. So what happened is the

1:11:36

breast cancer study failed. And I

1:11:38

was so disappointed. I was so hoping

1:11:40

that that would work. I still

1:11:42

remember that day. For me, it was

1:11:44

like, oh, we need more better

1:11:46

drugs for breast cancer because I often

1:11:48

heard from people when they're the

1:11:50

three out of four patients to her

1:11:52

septum wasn't for them. If you

1:11:54

looked at your stock that day, it

1:11:56

also looked really bad because all

1:11:58

the hype about Avastin was there. But

1:12:01

in colon cancer, we had a phase

1:12:03

two, got ready on the dose and schedule,

1:12:05

and then we went to a phase

1:12:07

three study in colon cancer, much more traditional.

1:12:09

just plus minus a vast. 5FU and

1:12:11

the usual suspects. 5FU and the usual suspects

1:12:13

plus minus a vast. And that succeeded.

1:12:15

That was a stage four? Only stage four.

1:12:17

This is a median survival study. You're

1:12:19

not doing overall survival, correct? That was a

1:12:22

median survival study. Actually, I don't remember

1:12:24

all the details of that one. I feel

1:12:26

like it was eight more months of

1:12:28

median survival. Does that sound about right? Probably

1:12:30

right. It was the first new thing

1:12:32

in colon cancer for a while too. So

1:12:34

people were pretty excited. Now, at this

1:12:36

point, I'm in medical school just down the

1:12:38

street. I'm at Stanford. And I remember

1:12:40

we had a big discussion about this. I'm

1:12:42

in my first year of medical school,

1:12:44

and the discussion we had in class was,

1:12:47

I think at the time we've asked them was

1:12:49

$100 ,000 for the treatment. Extends median survival

1:12:51

by whatever, but I think it was eight months.

1:12:54

The UK said no. The NHS said

1:12:56

we are not paying for this because

1:12:58

at the time the NHS had this

1:13:00

$100 ,000 quality adjusted life year hurdle. which

1:13:02

is understandable, right? That's how the Uthrottle

1:13:04

supply site economics. They said, look, we

1:13:06

can't pay for a drug. We can't

1:13:08

pay more than $100 ,000 for an

1:13:10

incremental year of quality adjusted life year.

1:13:12

This is only giving you eight months.

1:13:14

That's why I know it was less

1:13:16

than 12. And so the NHS

1:13:18

flatly said, we're not paying for this. And I

1:13:20

do believe people in the UK could pay out

1:13:22

of pocket for it. You can get it out

1:13:24

of pocket, but not through the National Health Service.

1:13:27

That's right. People in Canada could not because you

1:13:29

couldn't have private insurance in Canada, though you could

1:13:31

come to the US for treatment. So of course,

1:13:33

this just became a great topic of discussion for

1:13:35

med school freshman, what was your

1:13:37

thought at the time of, have

1:13:39

we moved the needle enough? How do we

1:13:41

think about the economics of this? So

1:13:43

I had a lot of different

1:13:45

reactions. First of all, with Avastin, it

1:13:47

was the first time I remember

1:13:49

reading, and I think it was one

1:13:51

of those curtain raiser things in

1:13:53

the Wall Street Journal for the breast

1:13:55

cancer study. The headline was, Avastin

1:13:58

might help breast cancer patients, but

1:14:00

can they afford to take it? I

1:14:02

was shocked that was the first

1:14:04

time I had read and as long

1:14:06

as I had been at Genentech

1:14:08

that somebody couldn't afford one of our

1:14:10

drugs. That instead of saying, oh,

1:14:12

isn't that cool? Britaxin, Herceptin, Avastin, it

1:14:14

was too much money. And

1:14:16

that felt really important to me and

1:14:18

really not good. We had as a company

1:14:21

a philosophy that no patient should go

1:14:23

without any of our drugs because of an

1:14:25

inability to pay. So we had a

1:14:27

bunch of patient, what do you call those?

1:14:29

Patient support programs, whatever. So we had

1:14:31

a bunch of different things in place. So

1:14:33

I knew we had those programs, but

1:14:35

that doesn't help the patients in the UK.

1:14:38

And it doesn't help the overall cost

1:14:40

because we're actually supplementing, but the cost is

1:14:42

still really high. And we started to

1:14:44

have a lot more discussion at the executive

1:14:46

committee about the price and how we

1:14:48

would think about it and how we would

1:14:50

price the drugs because that was, like

1:14:53

I say, that was not good. Did you

1:14:55

go straight from Genentech to being a

1:14:57

chancellor at UCSF? Just because I

1:14:59

want to stay with the story, I want to continue

1:15:01

the arc of your career. We're at the halfway point,

1:15:03

right? We're one -third of the way

1:15:05

into... Early days. We're in

1:15:07

the 90s. How did you, given

1:15:09

you are truly on the

1:15:11

cusp? of what is happening in

1:15:13

oncology and biotechnology. And

1:15:15

now the same institution that said, you

1:15:17

can't come back here to have

1:15:19

a clinical appointment after saving the people

1:15:21

of Uganda is now offering you

1:15:23

the highest post essentially outside of a

1:15:25

provost, I'm guessing, right? I don't

1:15:28

even know where the is the highest

1:15:30

post. It is the highest post.

1:15:32

Okay, so that's kind of remarkable. And

1:15:34

does that just speak to what

1:15:36

they saw as the vision of that

1:15:38

institution, which was... people have learned

1:15:40

what Sue has learned in the last

1:15:42

10 years, and we want that

1:15:44

type of leadership here. So first of

1:15:46

all, on the genetic side of

1:15:48

things, Roche bought us. That's right. That

1:15:50

was 99. That

1:15:52

was 2009. Oh,

1:15:55

2009. Oh my God. Okay. I'm off

1:15:57

by a whole decade. I thought that was

1:15:59

sooner. 2009, which was not what we

1:16:01

wanted. It was a... It's a hostile takeover.

1:16:03

A hostile takeover. Yeah. They call it in the

1:16:05

business world, a squeeze out. They squeezed us out. So

1:16:07

I knew I was going to leave. I knew

1:16:09

I was going to do something different. And

1:16:12

UCSF, I had been really

1:16:14

close with the chief of medicine

1:16:16

who was my chief of medicine

1:16:18

when I was an intern, Holly

1:16:20

Smith. Do you know Holly

1:16:23

Smith? No, I don't. So Holly

1:16:25

Smith was a South Carolinian,

1:16:27

Harvard trained, amazing person. And between

1:16:29

him and Bill Rutter, who

1:16:31

founded Chiron, who's a biochemist, they

1:16:33

decided a long time ago

1:16:36

that UCSF should not be a

1:16:38

backwater medical school and should

1:16:40

be a serious medical school. So

1:16:42

Holly on the clinical front and

1:16:44

Bill on the scientific front just decided

1:16:46

they would start recruiting people to

1:16:49

have a great institution, like a pretty

1:16:51

amazing thing. Matt was starting

1:16:53

in the 80s. It was

1:16:55

probably 80s, 70s, 80s. I was

1:16:57

still friends with Holly. I

1:16:59

just think there's so many wonderful

1:17:01

things about Holly that I

1:17:03

admired. that I was still

1:17:05

friends with him. So Holly called

1:17:08

me and said, Mike Bishop is

1:17:10

stepping down of Bishop Envarmus, that

1:17:12

Mike Bishop. Of Nobel fame. And

1:17:14

you should be Chancellor. Of

1:17:17

course, I said to Holly, I'm not a professor.

1:17:19

And I said, well, if they're interested

1:17:21

in me, I'd be open to talking

1:17:23

to them. They were, and I went

1:17:25

through the interview process. I'm sure that

1:17:27

when you're going through that, sorry to interrupt.

1:17:30

Because you're now interviewing with the Board of

1:17:32

Trustees. They have a search committee. Yeah. Yeah.

1:17:34

And so they must be asking you to

1:17:36

present a vision. They're not interviewing you to

1:17:38

make sure you know how to use PowerPoint.

1:17:40

Yeah. Do you remember what the vision is

1:17:42

that you presented to them? What

1:17:44

I do remember, and I remember

1:17:46

if you remember 2009, was

1:17:48

just... Post -apocalypse. Horrible recession

1:17:50

and California being particularly

1:17:53

in a bad place. So

1:17:55

I just talked to

1:17:57

them about how I think

1:17:59

about managing people, how

1:18:01

I think about making sure that you

1:18:03

use whatever assets you have maximally. I

1:18:05

admitted that I had never done fundraising,

1:18:07

but I had done a lot of

1:18:09

work with Wall Street and I could

1:18:11

talk. You were the president. What was

1:18:13

your title at Genentech before left? President.

1:18:15

Yeah. Yeah. And I did a massive

1:18:17

amount of investor relations because art didn't

1:18:19

like traveling or talking. So...

1:18:22

My kind of guy. Mine

1:18:25

too. No, we were a good

1:18:27

partnership. But I also said, look, if

1:18:29

you think about running the faculty

1:18:31

meeting, I'm probably not your guy, but

1:18:33

that's what the provost does. I

1:18:35

think they really thought they wanted somebody

1:18:37

who could work on the business

1:18:39

aspects of the campus. We needed to

1:18:41

finish Bay. You're really the CEO

1:18:43

of the system. doing. Yeah. Yeah. And

1:18:45

you manage the hospital CEO and

1:18:47

it's a big hospital system. There's no

1:18:49

undergrads at UCSF. So that was

1:18:51

what they were looking for. And I

1:18:53

thought, well, why not? Were

1:18:55

you nervous? I mean, when they called you

1:18:57

and said you've got the job, was there

1:18:59

a moment where you thought, have I bit

1:19:01

off more than I can chew? This is

1:19:03

a huge responsibility. I was really concerned. I

1:19:06

was really concerned about it. And I also

1:19:08

realized, because some of the faculty were pretty

1:19:10

negative when I first started. Because you were

1:19:12

an outsider, you were an alum, but you

1:19:14

weren't coming up through the ranks as the

1:19:16

CEO of the hospital or something. I actually

1:19:18

think they were just as nervous as I

1:19:20

was about money. They weren't convinced that I

1:19:23

knew how to get them money. Because, you

1:19:25

know, if you're running a program, you need

1:19:27

money. It's the mother's milk of doing science.

1:19:30

So tell me about the budget

1:19:32

of UCSF, because it's a state

1:19:34

school, presumably California provides what fraction

1:19:36

of it? Almost nothing. So what's

1:19:38

the benefit of being a UC? The

1:19:40

brand. There's a curriculum thing. There's some

1:19:42

things that. So in that sense, you

1:19:44

and Stanford aren't that far apart? Not

1:19:46

that different. I mean, the most important

1:19:48

things, you've got clinical income. You've got

1:19:50

NIH -driven income, right? You've got other

1:19:52

grant income and you have philanthropy. Show

1:19:55

me the P &L on those things there.

1:19:57

So NIH is bringing in how much?

1:19:59

Oh gosh. Percentage -wise, roughly. Of

1:20:01

the money that you use every day,

1:20:03

there's this discussion of overheads now. yeah.

1:20:05

It probably gets up to a third.

1:20:07

A third of the revenue for general

1:20:09

operations is coming out of the NIH

1:20:11

overhead. Probably, yeah. And then clinical revenue.

1:20:13

Clinical is a lot, but a lot

1:20:15

of it goes back to the hospital.

1:20:17

You know, it's a not -for -profit. They

1:20:19

spend it on the hospital. Okay. And

1:20:21

then philanthropy is some direct, some into

1:20:23

the endowment where you're living off the

1:20:25

interest. Yeah. And that's basically what your

1:20:27

revenue streams are, those four things. And

1:20:29

then the tuition is tiny. Well, especially

1:20:31

because you don't have undergrads, right? Yeah.

1:20:33

Right. Because the number of students is

1:20:35

really low. And the really good news

1:20:37

at UCSF. I didn't even apply to

1:20:39

UCSF, by the way, because I mean,

1:20:42

I was not in California when I

1:20:44

was applying to medical school. I

1:20:46

was told, well, such a great medical school

1:20:48

that I was like, there's no way I'm

1:20:50

going to get in. as a non -Californian. So

1:20:52

I didn't even apply. So the funny thing

1:20:54

for me is... But you did and you

1:20:56

got in. Well, there's a story there. We

1:20:58

talked about me being at University of Nevada.

1:21:00

What we didn't talk about is my first

1:21:02

year at University of Nevada, my

1:21:04

youngest sister was born. So

1:21:06

I lived at home and helped

1:21:08

my mom in Reno. Because there's like

1:21:10

seven of you, right? There's seven

1:21:13

of us. Yeah. And number seven was

1:21:15

born when I was a freshman.

1:21:17

It was kind of crazy. But when

1:21:19

I went to medical school, I

1:21:21

used to have like this sign on

1:21:23

the stairs, be quiet. I'm studying

1:21:25

down here. It's like probably a giant

1:21:27

pain. But I wanted to

1:21:29

go to San Francisco. My dad was

1:21:31

born and raised in San Francisco. My

1:21:33

grandma lived in San Francisco. So even

1:21:35

though I was still pretty young in

1:21:38

terms of ever living outside the home,

1:21:40

I knew San Francisco and wanted to

1:21:42

go. But UCSF had never taken. a

1:21:44

university in Nevada student because we were

1:21:46

a pretty new medical school. I was

1:21:48

only in the second four -year class. Oh,

1:21:50

wow. Yeah. And I was excited like

1:21:53

my head would blow off. Like I

1:21:55

wanted to go to UCSF so big

1:21:57

time. So I'm going

1:21:59

back to UCSF and the

1:22:01

really good news, what saved

1:22:03

me as chancellor is philanthropy. It

1:22:06

turns out that we needed people

1:22:08

to care about the mission and the

1:22:11

projects at UCSF. right at the

1:22:13

same time as the Mark Zuckerbergs of

1:22:15

the world and the venture capitalists

1:22:17

of the world. And a lot of

1:22:19

people had come into a lot

1:22:21

of money. Even though the overall economy

1:22:23

was bad, it was coming back. And

1:22:26

we had some spectacular

1:22:28

successes. And my successor, Sam

1:22:30

Hoggard, continues to have that kind of

1:22:33

success. And people are just really

1:22:35

generous. Americans are hands down

1:22:37

the most generous people in the

1:22:39

world. I think that's a demonstrable

1:22:41

fact. Yeah, I'm big on New

1:22:43

Year's resolutions. One of

1:22:45

my New Year's resolutions. I

1:22:47

have a mini list. One

1:22:49

thing that's always on the list is be more

1:22:51

generous. I'm no Mark Zuckerberg, but

1:22:53

I can be generous in other ways. I'm

1:22:55

sure that you must deserve some of

1:22:57

the credit for that. I don't think it's

1:23:00

just that a bunch of people in

1:23:02

the Bay Area came into money at that

1:23:04

period of time. What was the approach

1:23:06

you took towards philanthropy and how did you

1:23:08

reach donors that maybe previously hadn't been

1:23:10

involved in UCSF because again one of the

1:23:12

things that's working against you is you

1:23:14

don't have an undergraduate. So Stanford

1:23:16

has a big advantage over you in

1:23:18

which you've got a lot of people

1:23:20

that are coming through doing engineering degrees,

1:23:23

doing CS degrees who are going

1:23:25

on to create enormous enterprises. Anyone

1:23:27

who's an alum of UCSF went

1:23:29

to graduate school there. There's no

1:23:31

business school. There's no law school.

1:23:33

So you're missing out on a

1:23:35

lot of Don't forget the hospital.

1:23:37

Yes. The most important donor base

1:23:39

is Grateful Patients. Yes. Or people

1:23:41

who love science, technology. I

1:23:43

hired John Ford, who had

1:23:45

retired as the head of

1:23:47

Stanford's fundraising. Moved up

1:23:49

to the Northeast and he was my head

1:23:51

of development. And I talked to John

1:23:53

and I said, how do you do this?

1:23:56

Teach me how to be a good

1:23:58

fundraiser. And he talked about tell people your

1:24:00

hopes and dreams, tell people what you're

1:24:02

excited about and ask them what they're excited

1:24:04

about and see if there's a match.

1:24:06

And I think that was really important. And

1:24:08

then I also think that because I

1:24:11

had been at Genentech and I was sort

1:24:13

of gregarious and knew a lot of

1:24:15

people and people knew that I had a

1:24:17

decent business savvy. I wasn't going to

1:24:19

waste their money. We were very committed to

1:24:21

use the money wisely, especially in the

1:24:23

hard times and do special things at UCSF.

1:24:26

I was surprised. I

1:24:28

sort of worried that I would be sad

1:24:30

if people said no. That

1:24:32

it would be weird, especially if I knew

1:24:34

them well. So I would get myself psyched up

1:24:36

for the beginning of it. And then by

1:24:38

the end, I'd be like, oh, that's fine. Next

1:24:40

time if you're in town, let's talk again

1:24:42

and maybe it'll change or whatever. But it was

1:24:44

actually fun. I got to talk to and

1:24:47

meet a lot of great people. What

1:24:49

percentage of your time was

1:24:51

spent externally versus internally? Probably

1:24:54

40 % externally. a lot external. And

1:24:56

what was the internal focus then? Working

1:24:58

with the team. So who were your

1:25:00

direct reports? The provost? Provost, CEO of

1:25:02

the hospital, lawyer, all

1:25:05

the deans, that was really

1:25:07

important. And then part of

1:25:09

it was monthly we met with all

1:25:11

the chancellors with the president of the university.

1:25:13

We was marked off at the time

1:25:16

and then he stepped down after a while.

1:25:18

But the chancellors meetings were funny because

1:25:20

they all had undergrads. I always felt like

1:25:22

I was squirming like, are we done

1:25:24

yet? I thought

1:25:26

it was just really important. My favorite thing,

1:25:29

every Friday, lunch, Mission

1:25:31

Bay, they had a science talk. And

1:25:34

they'd have some pizza, Chinese food,

1:25:36

something, and you'd look around

1:25:38

the room and it'd be like Bruce

1:25:40

Alberts, Liz Blackburn, there'd be four Nobel

1:25:42

Prize winners in this little cramped room

1:25:44

listening to science. This would be something

1:25:46

you did as a scientist, not necessarily

1:25:48

as the chancellor. Not as a chancellor.

1:25:50

The chancellors weren't doing this all the

1:25:53

time. I just go over on the

1:25:55

shuttle bus, eat a slice of pizza

1:25:57

and enjoy. Yeah, amazing. It was really

1:25:59

good. So then let's get to

1:26:01

the next chapter. What all of a

1:26:03

sudden in 2013, 2014 leads to the next

1:26:05

transition to being the CEO of the

1:26:07

Bill and Melinda Gates Foundation? Well,

1:26:09

to my surprise, I think I

1:26:11

got an email or I think

1:26:13

it was an email from Melinda.

1:26:16

Did I have time to talk?

1:26:18

And UCSF throws a big event

1:26:20

every year that's kind of friend

1:26:22

-making, fundraising, everything, and we give

1:26:24

out awards, recognition to people whose

1:26:26

work we respect a lot. So

1:26:28

I had invited Melinda the year

1:26:30

before and thought, she'll never come. You

1:26:33

invite people, the throwaway invite. And she

1:26:35

accepted and came, actually came with her

1:26:37

mom and dad even better. And I

1:26:39

thought that was nice. So

1:26:41

she sent me an email and

1:26:43

she said that she and Bill wanted

1:26:46

me to look at being the

1:26:48

next CEO of the Gates Foundation.

1:26:51

And I was surprised. I had

1:26:53

not expected that. And I started

1:26:55

having discussions with them. It was

1:26:57

actually funny. I went up to

1:26:59

Seattle and they were having all

1:27:01

this hush -hush. You know, this was

1:27:03

very cloak and dagger. So

1:27:05

I went to their house because I had

1:27:07

had a meeting with Melinda and I needed

1:27:09

to meet with Bill. And it was Halloween.

1:27:13

The kids are coming and going. It was

1:27:15

kind of crazy. as I had talked to

1:27:17

him. And I talked to my husband. You know,

1:27:19

my husband worked at the Gates Foundation. He

1:27:21

led the HIV and TB programs

1:27:23

about five years before. Was

1:27:26

he still there? No, no, he had gone.

1:27:28

He was commuting to Seattle, which was dreadful. But

1:27:30

he knew the Gates Foundation. So he and

1:27:32

I were talking about this and I was like,

1:27:34

oh, God, you know, I've only been at

1:27:36

UCSF five years. I just found where the bathrooms

1:27:38

are, you know, that kind of thing. And

1:27:40

it was going well. I was happy with that.

1:27:42

They asked for a teleconference and

1:27:44

they got on the phone and

1:27:46

Especially Bill made this big pitch

1:27:48

that a lot of people could

1:27:50

do the UCSF job with all

1:27:52

due respect as Bill went for

1:27:54

it. But I was the only

1:27:57

person who they both wanted and

1:27:59

who could do this job. I

1:28:01

was perfect for the job and

1:28:03

it's really important for the world

1:28:05

and I needed to do it.

1:28:07

I assume that the rationale for

1:28:09

that was obviously their focus is

1:28:11

on global health and You

1:28:13

have the background in the clinical side,

1:28:15

the research side, the epidemiologic side,

1:28:17

the management side. So there's kind of

1:28:19

those are four legs of a

1:28:21

chair. Were there other things that I'm

1:28:23

missing that they felt were kind

1:28:25

of essential pillars? I think it

1:28:27

was less obvious then, but I think

1:28:29

now they had started to kind of have

1:28:32

disparate views of how the foundation should

1:28:34

operate. Melinda has been

1:28:36

really all over women's issues, all

1:28:38

over. And Bill would do another

1:28:41

run at polio. It's like the

1:28:43

goal broadly that all lives have

1:28:45

equal value, which by the way,

1:28:47

I think is a wonderful thing.

1:28:49

They share, but they come at

1:28:51

it from different ways. And

1:28:53

so I think that the thing

1:28:55

that resonated for me is that I

1:28:58

could see both those points of

1:29:00

view. But those points of view don't

1:29:02

strike me as mutually exclusive for

1:29:04

an organization with enormous resources. Yes and

1:29:06

no. It's one thing to have

1:29:08

enormous resources. It's another to know where

1:29:10

one of the most important assets

1:29:12

they have is the time and energy

1:29:14

of Bill and Melinda. They actually

1:29:16

show up. Things happen. So

1:29:18

how did you weigh this

1:29:20

decision? I thought that I

1:29:22

could add value. I thought I would learn

1:29:24

a lot. And I did think that UCSF

1:29:27

would be fine without me. I felt like

1:29:29

we were back on our feet financially. I

1:29:31

thought that Sam Hoggood, who was the Dean

1:29:33

of the School of Medicine, I had a

1:29:35

ton of respect for him and thought he

1:29:37

was the obvious person to take my place

1:29:39

and that it would be okay. Were they

1:29:41

surprised? Did they try to talk you out

1:29:44

of this? I don't think they did. I

1:29:46

actually think they had a lot of respect

1:29:48

for the Gates Foundation and thought, oh, well,

1:29:50

that's a cool job. At least the

1:29:52

way they showed up with me. Maybe when I wasn't

1:29:54

there, they did. Okay. So you head up to

1:29:56

Seattle now. Yes. When you show up to the foundation,

1:29:58

how many employees are there? What does it look

1:30:01

like? It's not for profit, but does it run like

1:30:03

Microsoft? I mean, how does it operate? It's

1:30:05

a couple thousand. It's a big

1:30:07

foundation, big building, big foundation with

1:30:09

people all over the world now.

1:30:12

There was a lot I wasn't

1:30:14

surprised by, like the global

1:30:16

health stuff I knew what they

1:30:18

were doing, and I thought

1:30:20

it was interesting and great. And

1:30:22

the challenge for me was

1:30:25

Bill's endless appetite for things like

1:30:27

learning things. driving things, funding

1:30:29

things, and me feeling like

1:30:31

I could get my hands around a

1:30:33

strategic plan. It was a little bit

1:30:35

like, okay, this staff would be like,

1:30:37

Bill's gonna love this. Let's present that,

1:30:39

you know, it was that kind of