The Future of Medicine
Welcome to The Future of Medicine, a podcast from Stanford's Department of Medicine.
We bring you into conversation with the thought leaders who are reshaping how we understand disease, deliver care, and imagine what's possible in human health. This show is built around the extraordinary speakers who join us for Medicine Grand Rounds – one of the longest-running and most respected forums in academic medicine.
Our guests include world-renowned physicians, scientists, innovators, and policy leaders from across the globe, as well as the remarkable faculty at Stanford. Together, they represent the full spectrum of modern biomedical discovery: from breakthrough therapeutics and cutting-edge genomics, to health equity, digital health, global health, neuroscience, AI, and the re-design of care systems.
This is The Future of Medicine.
The Future of Medicine
Valter Longo on Fasting-Mimicking Diets and Longevity
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
What can fasting teach us about aging, and can the body receive some of its benefits without giving up food completely? Valter Longo, PhD, professor of gerontology and biological sciences, director of the Longevity Institute at the University of Southern California, and author of The Longevity Diet, joins The Future of Medicine for a wide-ranging conversation about nutrition, fasting-mimicking diets, protein, chronic disease, and the biology of longevity.
In this episode, Longo traces an unconventional path from aspiring rock guitarist to longevity scientist. He describes how the early death of his grandfather and the striking differences he observed between relatives living in Italy and Chicago helped shape his interest in why people age and develop disease.
Longo explains how experiments in yeast led his laboratory to study the effects of sugar, amino acids, and fasting on lifespan. He also discusses the development of the fasting-mimicking diet, a five-day, low-sugar, and low-protein eating plan designed to trigger some of the biological responses associated with fasting while still providing food.
The conversation also examines caloric restriction, time-restricted eating, breakfast skipping, and today’s enthusiasm for high-protein diets. Longo distinguishes between short-term changes in weight or metabolic markers and the much harder question of whether a dietary practice improves long-term health.
Together, Longo and host Euan Ashley explore what scientists are learning about nutrition and healthy aging, why extreme approaches can carry tradeoffs, and how carefully designed dietary interventions might one day become part of preventive medicine.
Thank you for listening!
Call to action: If you enjoy The Future of Medicine, subscribe for more conversations with leading scientists shaping the next era of healthcare. Please rate and review the podcast to help others discover these important discussions. Share with friends and colleagues who are curious about how science becomes medicine.
0:00:00.000,0:00:05.440
Ketogenic diet short-term is good, long-term it
kills you early. The 16-hour fasting short-term is
0:00:05.440,0:00:11.280
good. Long-term it kills you early. Very low carb
diets. Short-term is good. Long-term it kills you.
0:00:11.280,0:00:16.320
Dr. Valter Longo is a professor of gerontology
and biological science and the director of the
0:00:16.320,0:00:21.200
Longevity Institute at the University of
Southern California. His research focuses
0:00:21.200,0:00:27.040
on the fundamental mechanisms of aging, including
the relationship between nutrition and longevity.
0:00:27.040,0:00:31.520
Five days a month completely reversed
everything — the cholesterol,
0:00:31.520,0:00:36.160
the heart problem — and we didn't think
it was going to be that much better.
0:00:36.160,0:00:40.320
In this conversation, we talk about his early
years touring in a rock band, the family
0:00:40.320,0:00:45.840
history that informed his interest in aging — as
a five-year-old, just seeing somebody dying in the
0:00:45.840,0:00:51.840
room, his grandfather, thinking this has got to be
the most important thing that I can do: figure out
0:00:51.840,0:00:57.600
why people die — and how fasting mimicking diets
can help people live longer, healthier lives.
0:00:57.600,0:01:03.040
Welcome to Stanford Department of Medicine's
inside look at the future of medicine.
0:01:03.040,0:01:05.200
Well, Valter, welcome to Stanford.
0:01:05.200,0:01:07.280
Oh, thanks for inviting me.
0:01:07.280,0:01:11.840
Yeah, it's really great to have you here, and
we've been really excited about your visit for
0:01:11.840,0:01:18.000
some time. As I was mentioning a few minutes ago
in the warm-up here, we really love to dig into
0:01:18.000,0:01:23.280
people's background, and you have a particularly
interesting background, and the way it took
0:01:23.280,0:01:28.240
you to where you are today, and the amazing work
you're going to share with us at Grand Rounds. So
0:01:28.240,0:01:36.560
let's go back to the past first of all — tell me
about your life story, where you grew up, and in
0:01:36.560,0:01:41.280
particular I'm kind of interested in your musical
career, so make sure you tell us about that.
0:01:41.280,0:01:50.320
Yeah, so I grew up in Genova, Italy, but between
Genova, which is in the northwest, and then
0:01:50.320,0:01:57.840
Calabria, which is in the south, the southern
tip of Italy. And then I moved when I was 16 to
0:01:57.840,0:02:10.960
Chicago with some uncles and aunts, and I think
watching my family in Chicago — having diabetes
0:02:10.960,0:02:17.760
and cardiovascular disease, and they were all
Italian, 100% Italian — and yet I didn't remember
0:02:17.760,0:02:25.600
anybody in Italy in my family getting the same
diseases. So that was an early motivation for me.
0:02:26.400,0:02:31.360
I wasn't thinking too much about it because I was
a musician, I was a guitar player, but I started
0:02:31.360,0:02:38.640
thinking that's strange — multiple of them having
advanced stage diabetes and lots of problems.
0:02:38.640,0:02:46.640
So 16 years in Italy, then you
moved to Chicago — tell us more
0:02:46.640,0:02:49.920
about growing up in Italy there
and what your aspirations were
0:02:49.920,0:02:54.000
for the future. You mentioned you were a
musician — was that your plan for life?
0:02:54.000,0:02:59.760
Yeah, I wanted to be a rock star, right?
That's for sure. I told my parents when
0:02:59.760,0:03:04.320
I was 12 I wanted to go to London. Of
course they were laughing. But then a
0:03:04.320,0:03:09.520
few years later they sent me to Chicago
thinking I would come back in a month,
0:03:09.520,0:03:16.080
and then of course I never went back. But Chicago
was great, because I started taking lots of
0:03:16.080,0:03:21.440
music lessons from lots of people, and there was
somebody called Stuart Pierce who was doing bebop.
0:03:21.440,0:03:25.280
I heard you were a jazz guitarist as well, right?
0:03:25.280,0:03:27.280
Yeah. Well, then I went to
University of North Texas,
0:03:27.280,0:03:33.040
which at the time was the leading jazz
program in the US — believe it or not,
0:03:33.040,0:03:38.960
in northern Texas. So I studied there,
started with music, jazz performance.
0:03:38.960,0:03:45.120
And who were your major
influences in jazz, or on guitar?
0:03:45.120,0:03:50.800
Well, I was listening to people
like John Scofield and Pat Metheny,
0:03:51.840,0:03:56.560
but I was really training to be a
rock guitarist — I wanted to get the
0:03:58.160,0:04:05.680
jazz background to be a better rock guitar
player. That was the idea, to come to LA,
0:04:05.680,0:04:13.920
and then being in a rock band eventually, which
I did. But then, the second year of college,
0:04:13.920,0:04:19.760
as a jazz performance major in Texas, they say,
"You have to direct the marching band." And I say,
0:04:19.760,0:04:28.960
"There is no way I'm directing the marching
band." So that's when I switched to biochemistry.
0:04:32.800,0:04:39.200
So this music college had a marching band? I
didn't think music colleges had marching bands.
0:04:39.200,0:04:45.840
The marching band — you were required, as a
music major, that was part of the curriculum.
0:04:49.440,0:04:51.280
And then, just to jump back before you got there,
0:04:51.280,0:04:54.560
in Chicago you were already
playing publicly, right?
0:04:54.560,0:05:00.560
Yeah, I was already going to Rush Street, and
I would go at night and just plug in with the
0:05:01.120,0:05:09.600
blues musicians. As a teenager, it was great
— I would take the train in the middle of the
0:05:09.600,0:05:17.120
night, the "L," they call it in Chicago, which was
not a good idea. But I would just take my guitar,
0:05:17.120,0:05:24.240
and I never got robbed, believe it
or not. But it was an experience.
0:05:24.240,0:05:29.120
I'm a big jazz fan, and I actually
— if I didn't play the saxophone,
0:05:29.120,0:05:37.920
I would play the guitar. Jim Hall, Joe Pass,
John Scofield, Mike Stern — and Bill Frisell,
0:05:37.920,0:05:42.560
I love his sense of melody.
We have to jam at some point.
0:05:42.560,0:05:45.360
That would be great, though I'm pretty rusty.
0:05:45.360,0:05:48.720
Well, you went to music school — I
had considered going to music school,
0:05:48.720,0:05:53.840
actually. But you actually did, so I
think that makes you a proper musician.
0:05:53.840,0:05:58.720
Yeah, and I have to say, when I
switched from music to biochemistry,
0:05:58.720,0:06:01.840
even though the biochemistry professor
was like, "I don't think you're going to
0:06:01.840,0:06:09.680
make it," I was like, this is so much
easier. That school is really tough.
0:06:10.800,0:06:17.600
So they almost wanted you out of the music school?
0:06:18.240,0:06:20.800
They were trying hard — it's like *Whiplash*.
0:06:23.600,0:06:27.520
Yeah, that was a great
movie, by the way, well done.
0:06:27.520,0:06:33.120
So then you switched your major to biochemistry,
but you were still doing a minor in music?
0:06:33.840,0:06:39.120
Yeah, I was doing my music, and
then I had a band all the way to my
0:06:39.760,0:06:45.280
last year in the PhD at UCLA.
We would leave Thursday night,
0:06:46.000,0:06:54.080
go to San Francisco, Portland, Seattle,
Spokane, then come back Monday morning.
0:06:54.080,0:06:57.440
That's wild — was that a rock
band, blues band, or jazz?
0:06:57.440,0:07:03.200
No, that was a rock band. This is
the grunge era. So it was great,
0:07:03.920,0:07:09.280
until professors at UCLA started
coming to me and saying, "That's
0:07:09.280,0:07:18.080
really nice that you're doing that, but..."
So that was the end of it, during my PhD.
0:07:18.080,0:07:18.320
They weren't so impressed
with your music distraction?
0:07:18.320,0:07:23.040
I mean, I was thinking that was cool,
but they were not so impressed with my
0:07:23.840,0:07:33.520
music distraction, because it was taking up
so much time — it wasn't like a minor hobby,
0:07:33.520,0:07:43.040
I was touring with a band. We'd be playing all
the major rock clubs, like the Whisky, the Roxy.
0:07:59.840,0:08:05.840
The Whisky a Go Go — I'd heard of it before,
because there's a gene that causes long QT
0:08:05.840,0:08:15.120
syndrome, an inherited arrhythmia, called *human
ether-a-go-go related gene*, and it's because when
0:08:15.120,0:08:19.520
these flies were exposed to ether they would
shake, apparently, like they'd dance in this
0:08:19.520,0:08:29.920
Whisky a Go Go bar. Geneticists have a sense of
humor, you know. Anyway, it's a potassium channel
0:08:29.920,0:08:35.680
gene. So I'd happened to have heard of this place
before — it was a pretty famous music venue.
0:08:36.560,0:08:42.000
Very famous — the Doors, and lots of all the top
0:08:42.960,0:08:49.840
names in music history have been through
there. But we were very small, of course.
0:08:49.840,0:08:54.000
So what led you to go back —
what led you to pursue a PhD,
0:08:54.000,0:09:02.320
given that you now had your biochemistry degree,
and science and music were both going — was the
0:09:02.320,0:09:10.560
idea that science would continue in the PhD
and music would continue through the band?
0:09:10.560,0:09:16.800
Actually, when I switched to biochemistry I wasn't
interested in biochemistry — I was interested in
0:09:16.800,0:09:23.600
aging. I don't know, for whatever reason, and
I always speculate that when my grandfather
0:09:23.600,0:09:31.920
died — I was five — and I was in the room when he
died. As a five-year-old, I thought no problem,
0:09:31.920,0:09:41.440
but just seeing somebody dying in the room, my
grandfather, I think that stuck in my head for 13,
0:09:41.440,0:09:47.200
however many years, and thinking this has
got to be the most important thing that I
0:09:47.200,0:09:53.920
can do — figure out why people die,
right? So when I left music school,
0:09:54.960,0:09:59.360
I was sure I had to study aging. It wasn't
like, oh, let me look around for what else
0:09:59.360,0:10:06.320
I can do. I thought, well, what better than
chemistry and biology together to study aging.
0:10:06.320,0:10:11.920
So even at that early moment, your choice
of biochemistry was already dictated by a
0:10:11.920,0:10:19.280
vision to study longevity. When was this in time?
0:10:19.280,0:10:31.040
18, 19 years of age. This is 1987.
0:10:31.760,0:10:38.960
Late '80s, right? So the longevity movement
that you've helped fuel, essentially,
0:10:38.960,0:10:45.520
probably didn't gain huge momentum or real
front-page news for another 10 or 20 years.
0:10:45.520,0:10:51.520
Maybe 20, right, because I always say that when I
was at UCLA and they would ask us in biochemistry,
0:10:51.520,0:10:56.800
"What are you working on?" We wouldn't say
aging, right — being embarrassed. We'd say, "Oh,
0:10:57.360,0:11:04.640
free radical biochemistry." People
were making fun — I literally
0:11:05.920,0:11:11.040
remember, I think it was at Washington
University, and this must have been like
0:11:11.040,0:11:14.080
2002 or 2003, and they were still making fun of
the aging field — somebody said, "Oh, there's
0:11:14.080,0:11:24.880
this journal called *Aging*," and I'm like,
"Hey, I'm a co-editor." And, "Oh, oops, sorry."
0:11:32.800,0:11:37.600
Not anymore, no — it's really in a good spot now,
which I think is a testament to the foundation
0:11:37.600,0:11:44.320
that you and others built, and just a realization
that we know so little about specifically cellular
0:11:44.320,0:11:50.880
senescence and how important that could be. But
since aging is something all of us inescapably
0:11:51.440,0:11:55.520
experience, it seems like something
we should all be interested in. So
0:11:57.040,0:12:00.800
let's go back to your professor who said
you got to do a bit less music and a bit
0:12:00.800,0:12:07.120
more science — how did that work out? How did
you balance that going forward in your PhD?
0:12:07.120,0:12:18.800
Well, I balanced that in the sense that I did both
until I got to my postdoc — then I stopped, right,
0:12:18.800,0:12:30.000
basically stopped. And I think they had a point,
right — you just can't do both, at least at that
0:12:30.000,0:12:37.360
stage. So that was a good idea. And then, with
the end of the PhD also came the end of my music
0:12:37.360,0:12:48.480
career. We had gotten a small contract from
Interscope Records, but I think I would have
0:12:48.480,0:12:56.000
needed a new band if I wanted to go anywhere. So
it was a good time, but it was time to move on.
0:12:56.960,0:13:02.400
Well, science is a team sport, and that's
sometimes a bit like playing in a band — almost
0:13:02.400,0:13:05.840
all science is collaborative in some way,
and people are playing their individual
0:13:05.840,0:13:11.360
roles. Tell us then, as you moved through
your postdoc, how you started to really
0:13:11.360,0:13:19.040
focus in on this area that would be your
obsession for the next couple of decades.
0:13:19.040,0:13:24.160
Yeah, so the postdoc, of course, I was
a student at UCLA of Roy Walford first,
0:13:24.160,0:13:34.800
and Walford at the time was a superstar in
aging — he was one of the pioneers of caloric
0:13:34.800,0:13:42.000
restriction, a very simple intervention that had
been established to be probably the most powerful
0:13:42.000,0:13:47.600
anti-aging intervention ever. As medical students,
that was all anyone taught us about aging,
0:13:47.600,0:13:52.880
right — that's all we knew. So Walford had a lot
to do with it. And Walford was the type of guy
0:13:52.880,0:14:03.840
who would be on *Larry King Live* — Larry King had
a big show where they interviewed famous people,
0:14:03.840,0:14:11.520
and Walford would be one of the ones interviewed.
So there was a very good time that I had with
0:14:11.520,0:14:18.640
Walford, and then back to biochemistry. By
the time I got to my postdoc, of course, I'd
0:14:19.440,0:14:26.240
had a lot of exposure to some of the best people
in the world working in aging — not just UCLA,
0:14:26.240,0:14:34.480
but the Leonard Guarente lab at MIT was doing
really, really good work, and Cynthia Kenyon up
0:14:34.480,0:14:44.880
here in the Bay. There was a lot of exposure to
conferences and the work by other, maybe six or
0:14:44.880,0:14:53.600
seven, labs that were like a really relatively
small community at that point. Everybody knew
0:14:53.600,0:14:59.760
each other, and we'd go to Gordon conferences
and it'd be like the same 10 people presenting.
0:15:02.480,0:15:04.480
Was there a sense then, among that group,
0:15:04.480,0:15:08.000
that you knew something others didn't
— that it was just a matter of time
0:15:08.000,0:15:16.000
before this broke into the big time — or was
there a different sense at those conferences?
0:15:16.000,0:15:22.480
No, I think the sense was more that everybody was
very passionate about it — and people were making
0:15:22.480,0:15:31.840
fun of the field, so it wasn't exactly like, "Oh,
now it's going to explode." We were wondering why
0:15:31.840,0:15:39.760
nobody was interested. And by the way, I picked
the worst field in that sense, and then the worst
0:15:39.760,0:15:47.360
organism, and even the worst within that — I was
working on starving yeast. People were making fun
0:15:47.360,0:15:52.400
of yeast, and then yeast people were making fun
of the people working on starving yeast. So I was
0:15:52.400,0:16:01.280
in the absolute worst nightmare selection
for a career, and everybody kept asking,
0:16:01.280,0:16:05.840
"When are you going to stop working on this yeast?
Why don't you start working on a real organism?"
0:16:07.840,0:16:11.520
For context, because I'm sure some
people listening don't have a good
0:16:11.520,0:16:18.320
sense of this — today it seems obvious
that you'd work on longevity and aging,
0:16:18.320,0:16:22.880
among the most important things you
could work on — give people a sense
0:16:22.880,0:16:26.720
of why they were making fun of you,
or thinking this isn't real science.
0:16:31.040,0:16:35.760
I'm told — and I don't know the details
— that the same was true for neurobiology
0:16:37.680,0:16:42.800
60, 70 years ago: people said it's too
complicated, why waste time learning about the
0:16:42.800,0:16:51.280
brain. Or even deep learning and AI models more
recently. I think if something is novel enough, it
0:16:51.280,0:16:58.480
seems like a crazy field, a bunch of crazy people.
Walford was a UCLA professor, a medical doctor,
0:16:58.480,0:17:09.440
but people in biochemistry would say, "We don't
know what they do over there," about pathology.
0:17:11.040,0:17:18.080
I think the overarching paradigm was: if we want
to help humanity, we should focus on disease. If
0:17:18.080,0:17:23.840
someone shows up with a disease, that's something
very concrete, something we can research,
0:17:23.840,0:17:31.920
write a grant on — heart disease, atherosclerosis,
cancer. Aging just seemed like a more amorphous
0:17:31.920,0:17:40.320
concept — too complicated and too diffuse. What
is it that you're going to get out of this?
0:17:41.120,0:17:48.720
And then, with the yeast — I made the mistake
of saying I didn't want to do something called
0:17:48.720,0:17:53.760
replicative lifespan, which is a unicellular
eukaryote measure — everybody was measuring aging
0:17:53.760,0:18:02.320
by how many times a mother cell generates daughter
cells, and I said I don't want to do this,
0:18:02.320,0:18:05.360
I want to study starving yeast instead —
what I call chronological aging. And so
0:18:05.360,0:18:09.520
it became an even bigger nightmare, because
now nobody else was doing that — I was the
0:18:09.520,0:18:17.920
only person on the planet studying it
that way. It seemed like I was dead,
0:18:17.920,0:18:22.960
right — I even started looking for
something alternative to the PhD,
0:18:24.480,0:18:35.280
maybe getting a job working in some biotech
lab, because it just didn't look good at all.
0:18:36.720,0:18:41.520
So then, track forward for us, because
obviously this all changed over time.
0:18:42.320,0:18:56.560
Yeah, things changed within yeast — we started
getting big papers because of the genetics. Once
0:18:56.560,0:19:03.520
we started studying yeast chronologically, like
every other organism, the techniques for yeast
0:19:03.520,0:19:13.920
were remarkable. At UCLA, the microbiology and
biochemistry departments were incredible — we
0:19:13.920,0:19:22.480
had hundreds of mutants ready to go, so we had
so much more power than any other organism.
0:19:22.480,0:19:26.400
*C. elegans* had nothing, by comparison
— we could do transposon mutagenesis,
0:19:26.400,0:19:34.000
look for any gene we wanted, and bring the
power of genomics to bear early with yeast.
0:19:34.800,0:19:39.040
You could do a triple mutant in weeks, where
with mice it would take years or decades.
0:19:39.040,0:19:42.080
Exactly right — if you wanted to do a
triple mutant with mice, it would take
0:19:42.080,0:19:52.400
years or decades. But with yeast, I could just
go upstairs, grab a double mutant, then make my
0:19:52.400,0:20:01.520
third mutation in a couple weeks, and I'd have a
triple mutant. And there's been this sense — these
0:20:01.520,0:20:07.920
competing ideas — that sure, humans are very
far away from yeast, but on the other hand,
0:20:07.920,0:20:17.200
so much of biology is shared that there's a lot of
fundamental biology you can learn from yeast. We
0:20:17.760,0:20:25.360
forget that we've been co-evolving for three and
a half billion years. Things split at some point,
0:20:25.360,0:20:34.320
but we're still co-evolving, right — so there
are fundamentals that are very central and
0:20:34.320,0:20:41.120
controlling of everything, and that's still
now the most important part of what we do.
0:20:42.880,0:20:45.680
Would you rather biohack your
way? We're in Silicon Valley,
0:20:45.680,0:20:48.240
right, and here it's all about biohacking.
0:20:48.240,0:20:55.600
But biohacking — I always say, you cut
yourself, and within a couple weeks it's gone,
0:20:55.600,0:21:00.560
perfect repair, right? That's three and a
half billion years of evolution. So let's
0:21:00.560,0:21:07.280
say you want to biohack your way to that — how
many decades will it take you, if you stop all
0:21:07.280,0:21:14.240
the natural processes, to biohack your way to
almost-perfect repair like that? So that's why
0:21:14.240,0:21:23.120
it was so important to identify already-evolved
alternative aging modalities. In yeast,
0:21:23.120,0:21:29.440
for example, you have three modalities: you
can be in what's called post-diauxic phase,
0:21:29.440,0:21:34.320
and you live about three, four days; you can
be in stationary phase, you live three weeks;
0:21:34.320,0:21:39.760
and then you can be in the spore state, you live
two years — it's a hundredfold difference in
0:21:39.760,0:21:50.720
lifespan between normal yeast and a spore, which
is genetically the same yeast. I always thought,
0:21:50.720,0:21:59.920
that's a trick that's already there. So I could
biohack my way to the spore, or learn — can I
0:21:59.920,0:22:16.640
get a spore that's high metabolism? So even with
AI and all the AI in the world, three and a half
0:22:16.640,0:22:21.760
billion years of evolution — you're not going
to do it very quickly. You're going to do it,
0:22:21.760,0:22:27.920
but how long is it going to take to get
that sophistication? Very, very difficult.
0:22:27.920,0:22:36.640
And I think much of the proof of the importance of
what you say — that biology really is preserved,
0:22:36.640,0:22:42.720
at least the important biology is preserved — is
that your work has had significant impact also
0:22:42.720,0:22:49.040
on humans, and some of the work you're best
known for. So draw us a line between the work
0:22:49.040,0:22:56.720
you're doing in yeast and where many people first
perhaps think about the foundation that was built
0:22:56.720,0:23:04.400
by caloric restriction, and specifically fasting,
since obviously a lot of your work has been about
0:23:05.920,0:23:11.760
the mimicking of fasting and the importance of
fasting — something humans definitely understand
0:23:11.760,0:23:18.000
because they do it every night. But also in
today's world, where there's a lot of people
0:23:18.000,0:23:26.240
skipping breakfast to fast for longer, or adopting
the "5 plus 2" phenomenon you've spoken about,
0:23:26.240,0:23:32.560
where they'll eat more normally for five days and
fast for two — draw the line between your work in
0:23:32.560,0:23:41.120
yeast and some of this work that many humans
are thinking about or living day-to-day now.
0:23:41.120,0:23:52.640
Yeah, so in yeast we had two mutations that were
very effective: one in the sugar pathway, RAS,
0:23:53.200,0:23:57.760
and one in the amino acid pathway — protein
kinase A signaling, to keep it simple. This
0:23:57.760,0:24:06.000
was very clear from the mid-'90s. If you
do a mutation in the amino acid pathway,
0:24:12.960,0:24:16.880
the yeast live three times as long,
and if you do it in the sugar pathway,
0:24:16.880,0:24:21.520
they live twice as long. And if you do it in both
pathways, they live five times as long. And if you
0:24:21.520,0:24:27.840
do it in both pathways and you starve them, it's
ten times. So that was very clear in my head from
0:24:27.840,0:24:34.080
the very beginning — fasting can do things,
but so can genes, and so can the regulation
0:24:34.080,0:24:49.440
of genes. That's everything we do, based on
that experiment. And I was lucky, because I
0:24:49.440,0:24:56.400
had Walford, and Walford, when I joined the lab
in 1992, was in Biosphere 2 — in the middle of
0:24:56.400,0:25:02.240
the Arizona desert near Tucson — doing a human
experiment on himself and seven other people.
0:25:18.320,0:25:20.640
Give us a one-minute sense of
Biosphere 2 for people listening.
0:25:20.640,0:25:27.200
Biosphere 2 was eight people, and it was
supposed to be like a moon station or Mars
0:25:27.200,0:25:35.680
station — so there was an electrician and
other different jobs, and Walford was the
0:25:35.680,0:25:43.120
medical doctor inside Biosphere 2. They were
locked inside for two years, that was the idea,
0:25:43.120,0:25:46.560
and they were supposed to grow their own food. But
within a couple months of entering they realized
0:25:46.560,0:25:59.600
they weren't going to make it — some bacteria were
degrading the cement, and oxygen levels dropped,
0:25:59.600,0:26:06.880
and there were issues growing their own
food. So Walford said, "Let's go on caloric
0:26:06.880,0:26:11.760
restriction" — conveniently, the world-leading
figure on caloric restriction saying, this is
0:26:11.760,0:26:18.720
the way to save food, eat less. So they started
the first human study on caloric restriction, and
0:26:20.080,0:26:32.240
the results were dramatic — cholesterol level,
blood pressure, glycemia, extraordinary. But
0:26:32.240,0:26:37.440
then you looked at them and thought there's
something wrong with this group of people,
0:26:37.440,0:26:42.000
even though the markers are great — so
thin, they looked thin but stressed out.
0:26:43.040,0:26:49.760
So that exposure was very important for me,
because I then said, I've got to combine the
0:26:49.760,0:26:54.480
yeast and *C. elegans* and everything we're
learning from all these organisms and mice,
0:26:54.480,0:27:05.280
and make it feasible. And the caloric restriction
data was starting to show good and bad — it would
0:27:05.280,0:27:10.800
take decades until the real bad stuff came out,
just like with the 16 hours of fasting now.
0:27:12.080,0:27:16.400
We were starting to get an idea this wasn't
just going to be all good — it was going to
0:27:16.400,0:27:20.480
be good and bad. And then the monkey study that
Richard Weindruch eventually did — he was in the
0:27:20.480,0:27:25.280
same lab as I was, then went to University
of Wisconsin and did a 30-year monkey study
0:27:25.280,0:27:28.560
on caloric restriction — and it showed
exactly that: you get a lot of benefits,
0:27:28.560,0:27:33.440
and you get a lot of problems at the end. You
don't live that much longer if you're chronically
0:27:33.440,0:27:39.280
calorie restricted. I think the combination of
that was really important then, to develop: okay,
0:27:39.280,0:27:45.600
what if we intervene once in a while, for five
days, and we don't fast people — we give them
0:27:45.600,0:27:51.360
food, but food that mimics fasting? And I thought,
everybody can do that — well, not everybody,
0:27:51.360,0:28:05.040
but maybe 50% of people, say every three months,
can do a five-day fasting-mimicking diet. And
0:28:05.040,0:28:11.600
would we have the same side effects? No, because
we had data from mice showing we didn't see lean
0:28:11.600,0:28:17.280
mass loss, we didn't see the problems — we
just saw the benefits of caloric restriction.
0:28:17.280,0:28:21.040
So lay that out in a little more detail
— what were those people experiencing,
0:28:21.040,0:28:28.320
what would they eat daily, what was
their diet like over a seven-day period?
0:28:28.320,0:28:36.080
Just five days, right — the FMD is five days,
and it was designed based on the yeast work.
0:28:36.080,0:28:46.480
We knew each ingredient that's blocking the
fasting response — amino acids, sugars — that
0:28:46.480,0:28:52.800
block the fasting response. So we eliminate
everything, and you end up with a low-sugar,
0:28:52.800,0:29:06.240
low-protein, high-fat, very healthy diet. And
I had a good idea, back 20 years ago when I
0:29:06.240,0:29:14.640
started working on it: let's make it like the
longevity blue-zone-based food — Okinawa, Japan,
0:29:14.640,0:29:21.760
Italy — let's take the common denominator, the
healthiest food in the world, and use that to
0:29:21.760,0:29:32.000
make the fasting-mimicking diet. I could have
made it with lard instead of olive oil and nuts,
0:29:34.320,0:29:43.200
and it still would have been fasting-mimicking,
but I think we would have had a lower benefit.
0:29:43.200,0:29:46.080
But also, thinking back to
your own personal history,
0:29:46.080,0:29:51.280
and the story you told at the beginning about
moving to Chicago and seeing your Mediterranean
0:29:51.280,0:29:57.040
relatives have disease — there's a whole
sense, at least in current human history,
0:29:58.240,0:30:04.160
of data and a feeling that a Mediterranean
diet has hidden properties of benefit.
0:30:04.160,0:30:10.720
Yeah, although — one of the things I show
is a study in mice from a few years ago,
0:30:10.720,0:30:18.160
where we take mice and give them the worst
nightmare diet, the Western high-fat, high-sugar,
0:30:18.160,0:30:24.240
high-calorie diet, and they become huge. And
then we take another set of mice and give them
0:30:24.240,0:30:31.040
the same bad diet for 25 days a month or so,
and then we give them five days of the FMD,
0:30:31.040,0:30:34.320
identical to the ones that
are always on the good diet.
0:30:34.320,0:30:35.200
Oh, really?
0:30:35.200,0:30:39.840
Yeah, so at least for mice, there was no
difference whether they had a good diet or
0:30:39.840,0:30:46.640
bad diet — and I'm not advertising my book, *The
Longevity Diet*, which talks about a very healthy,
0:30:46.640,0:30:55.280
even better than Mediterranean, everyday diet
— but at least that study showed, what if it
0:30:55.280,0:31:05.173
doesn't really matter what you eat, right? I mean,
it's remarkable, because we did the whole lifespan
0:31:05.173,0:31:12.000
study — so you have the control diet and they live
very long, and then you put them on the bad diet
0:31:12.000,0:31:20.960
and it almost cuts the lifespan in half. It's
remarkable, even in a mouse, how bad diet is at
0:31:20.960,0:31:26.640
killing you early, giving you high cholesterol,
high glycemia, all the problems people get.
0:31:26.640,0:31:32.720
But then five days a month completely reversed
everything — the cholesterol, the heart problem
0:31:32.720,0:31:36.800
— and we thought it was going to be a little
bit better, but it wasn't just a little bit.
0:31:36.800,0:31:39.760
And you've shown this in mice and in humans?
0:31:39.760,0:31:46.720
Well, in humans, we finished the trial — we have
a 500-person trial, finished in southern Italy.
0:31:46.720,0:31:54.480
I can't talk about the results yet, but the
trial has three arms: one is nothing, obese
0:31:54.480,0:32:01.040
and overweight people with one risk factor — high
blood pressure, high cholesterol — everybody's got
0:32:01.040,0:32:08.600
a problem, not a big problem. Then one group
does an FMD every three months — a lot of the
0:32:08.600,0:32:14.080
earlier trials we did were every month, and now
we think most people can do it every three months,
0:32:14.080,0:32:23.920
five days each time — they just get a box, like
medicine, a very standardized box. And the third
0:32:23.920,0:32:33.440
arm is the "longevity diet" — we throw everything
at them: 12-hour time-restricted eating, mostly
0:32:33.440,0:32:40.480
vegan, pescatarian, everything you can think
of, plus the FMD. So we'll see — that's coming.
0:32:41.840,0:32:50.160
You heard it here first. One of the things that's
obviously less common, especially at the moment
0:32:50.160,0:32:58.800
with the FMD, is that it's low protein, and a
lot of health influencers and even doctors are
0:32:58.800,0:33:05.760
recommending higher-protein diets. What do
you make of this trend toward more protein?
0:33:08.000,0:33:14.800
We like to look at multiple pillars, right — a
lot of people look at epidemiology and that's it,
0:33:15.520,0:33:19.920
and we think epidemiology is extremely
important, probably the number one pillar
0:33:19.920,0:33:25.600
in our view. But then, what about centenarians,
what about clinical trials, what about mice and
0:33:25.600,0:33:33.520
rats? If you put it all together, you come up with
low-but-sufficient protein — low but sufficient,
0:33:33.520,0:33:37.760
you cannot be malnourished, like we said about
caloric restriction. At a certain point the level
0:33:37.760,0:33:45.360
becomes too low. And soon enough we have several
papers that are now accepted, or close to it,
0:33:46.000,0:33:51.920
making the case that "protein" as a number is
irrelevant — it should be amino acids, it should
0:33:51.920,0:34:01.760
be amino acid profile, because from a legume
to fish, the essential amino acid content could
0:34:01.760,0:34:06.720
be a three- or four-fold difference. So
if you're talking about a gram of protein,
0:34:06.720,0:34:15.120
it could be one gram or four grams, depending on
how it matches the essential amino acid profile.
0:34:15.120,0:34:24.480
So your strategy is to provide the
essential amino acids and not much more?
0:34:26.800,0:34:32.960
I think it's looking that way
— it's like medicine, right,
0:34:32.960,0:34:39.680
how powerful a few amino acids are. A few
of the essential amino acids can completely
0:34:39.680,0:34:49.360
change the frailty or the lifespan of a mouse,
at least, and of rats. Now, if you go to humans,
0:34:49.360,0:34:54.800
is all that lost, like it might be in yeast and
worms and flies and mice and rats and probably
0:34:54.800,0:35:06.720
monkeys? Probably not. So my argument would
be that we need to know your frailty level,
0:35:06.720,0:35:17.280
and then match the amino acids so that it's
sufficient for what you're trying to achieve.
0:35:17.280,0:35:22.720
Because clearly there are a lot of people trying
to build muscle, either for cosmetic reasons,
0:35:22.720,0:35:29.360
strength reasons, or anti-frailty reasons, and
there's some reasonable evidence — though none
0:35:29.360,0:35:36.480
of it as strong as maybe the health influencers
suggest — that increasing protein in your diet,
0:35:36.480,0:35:43.840
up to a certain level, improves muscle bulk
gains in response to resistance training.
0:35:43.840,0:35:48.960
Right, but these are short-term effects, right?
0:35:51.600,0:35:56.240
There's a lot of things you can do short-term
— ketogenic diet is one. Short-term is good,
0:35:56.240,0:36:02.640
long-term it kills you early. The 16-hour fasting
— short-term is good, long-term it kills you
0:36:02.640,0:36:06.960
early. Very low-carb diets — short-term
is good, long-term it kills you early.
0:36:06.960,0:36:10.320
But do we actually have evidence that
the 16-hour fast itself kills you early,
0:36:10.320,0:36:14.160
or is that from meta-analysis
on breakfast skipping?
0:36:14.160,0:36:19.760
We have a lot of evidence for that — 16
hours, with breakfast skipping. The new
0:36:19.760,0:36:26.320
data is indicating it doesn't even matter
if it's breakfast skipping or not, but long
0:36:26.320,0:36:34.480
fasting — which is what most people doing 16 hours
are actually doing — is associated with increased
0:36:34.480,0:36:40.720
overall mortality, increased cardiovascular
mortality, and not a small increase — a
0:36:41.600,0:36:47.600
doubling of cardiovascular mortality in people
who skip breakfast. This is meta-analysis,
0:36:47.600,0:36:54.560
not just one study, so it's very clear. So
breakfast skippers consistently do very poorly,
0:36:54.560,0:37:00.320
and my colleagues argue it's because
people who skip breakfast have bad
0:37:00.320,0:37:06.800
lifestyles overall. But my argument is: if
it was so good for you — and it is good,
0:37:06.800,0:37:13.760
you get a lot of benefits — why wouldn't you
at least see them come back to normal? It could
0:37:13.760,0:37:25.760
be weight loss, insulin resistance, maybe better
sleep in some cases — so if they have bad habits,
0:37:25.760,0:37:29.440
and now they're doing something very healthy,
why don't we see them at least going back
0:37:29.440,0:37:35.440
to normal — no good, no bad? Instead
we see negative, over and over. This
0:37:38.080,0:37:48.160
tells you it's not a good place to start. Now,
if you go to 12 hours of time-restricted eating,
0:37:48.160,0:37:54.880
you see no problems — and of course 12 hours,
which most people don't really do anymore, since
0:37:54.880,0:38:00.160
you could say 12 hours is just eating normally.
Well, it used to be, but it's not anymore. So I
0:38:00.160,0:38:09.360
think 12 hours gets you there more slowly, but
there's no association with increased mortality.
0:38:10.400,0:38:17.360
So the difference between 12 hours and 16 hours
is a doubling in cardiovascular mortality?
0:38:17.360,0:38:18.800
Nobody's really tested 12 hours
specifically, because it's
0:38:18.800,0:38:19.369
never been directly compared — most of the data is
on breakfast skipping, and that's very negative.
0:38:19.369,0:38:21.520
There's newer data saying it's fasting of 14, 15,
16 hours that's bad, but I've never seen anything
0:38:21.520,0:38:31.440
negative at 12 hours — there's a Chinese group I
think using enhanced data that looked at over 14
0:38:31.440,0:38:53.520
hours of fasting versus less than 12 hours. All
the data is showing that breakfast skipping is
0:38:57.520,0:39:07.280
negative, and now nobody has carefully compared
the two, but certainly if you do 12 hours,
0:39:07.280,0:39:14.800
a lot of Sachin Panda's data indicates you
get benefits — maybe not as big or as quick
0:39:14.800,0:39:20.960
as at 16 hours. So that's why we're saying
it's much easier, much safer, probably less
0:39:20.960,0:39:27.920
likely to cause side effects like muscle loss
— so that's probably a much better way to go.
0:39:27.920,0:39:30.400
Do you think these studies are
overall large enough — I mean,
0:39:30.400,0:39:35.440
you described a randomized trial you're doing,
which is obviously our gold standard — do you
0:39:35.440,0:39:42.400
think most of the data in the field is large
enough to draw these conclusions confidently?
0:39:44.080,0:39:50.640
I don't think you can draw conclusions
confidently, but I think you can say:
0:39:50.640,0:39:57.120
there's 30 years of research, epidemiological
study after epidemiological study coming up
0:39:57.120,0:40:06.160
with big numbers of increased risk —
there's consistently increased risk
0:40:06.160,0:40:13.280
in breakfast skippers. That doesn't give you
confidence to say "I know for sure," but it
0:40:13.280,0:40:18.800
gives me confidence to say I'm not starting with
that — let's start with something that doesn't
0:40:18.800,0:40:29.200
have that association. Maybe it is fine, and some
people I talk to — a physician, last night — say,
0:40:29.200,0:40:34.560
"I do 16 hours, and maybe you're going
to live forever doing your 16 hours,
0:40:34.560,0:40:42.240
but maybe not — take a chance." That's really
a statistical issue, and that's all it is.
0:40:43.360,0:40:48.480
But then, going back to what you said
a few minutes ago — you have data that
0:40:48.480,0:40:54.000
would support that you could have a very
unhealthy diet for 25 days, and as long as
0:40:54.000,0:40:59.760
you have five days of the fasting-mimicking
diet, you can reverse your blood markers?
0:40:59.760,0:41:06.320
Yeah, but I wouldn't — this is why
we say to everybody, do it all,
0:41:06.320,0:41:12.880
right — have the healthy diet, and maybe
20 years from now we'll know for sure,
0:41:12.880,0:41:16.560
once 20 trials have been done and
we keep getting the same result.
0:41:17.120,0:41:20.560
And those markers — are they things
like cholesterol and blood sugar,
0:41:20.560,0:41:24.640
or much more extensive,
inflammatory markers as well?
0:41:24.640,0:41:29.640
We have inflammatory markers, yeah, and
we have lifespan — in mice we look at
0:41:29.640,0:41:31.520
cholesterol, fasting glucose, lots of different
things, and in humans we have like a hundred
0:41:31.520,0:41:50.560
different markers, epigenetic clocks, telomere
length, name it. But the most important of all
0:41:50.560,0:42:01.200
is lifespan — the five days are bringing
the lifespan back to statistically normal,
0:42:02.240,0:42:09.520
but a little bit shorter than the good diet.
So bad diet plus FMD is statistically the same,
0:42:09.520,0:42:16.320
but if you look at the curves, they're dying a
little bit earlier compared to the good diet.
0:42:16.320,0:42:22.800
And if you take the same markers at week
two, in the middle of the poor-diet period,
0:42:22.800,0:42:29.760
I assume those markers are off, in the negative?
0:42:32.640,0:42:45.440
We take them in the middle of the bad diet — a lot
of markers are not off, actually. Ketone bodies,
0:42:45.440,0:42:53.440
for example, during the bad diet, are still
elevated, which means the fat cells seem to
0:42:53.440,0:43:01.280
continue in a "good diet" mode even though
they're on the bad diet. We did RNA-seq,
0:43:01.280,0:43:10.880
and the fat cells are very active — a
lot of mitochondrial biogenesis markers.
0:43:12.000,0:43:16.400
So there's even a forward effect
from the five days into the next
0:43:16.400,0:43:21.120
period of bad diet that seems
to be protective in the moment?
0:43:21.120,0:43:26.800
I think it's a modality — they switch to
something else and stay in it, which makes
0:43:26.800,0:43:34.080
sense if you think about evolution: periods where
you're exposed to lots of food, and periods where
0:43:34.080,0:43:39.680
you're exposed to no food at all, like grizzly
bears and emperor penguins of the South Pole.
0:43:47.520,0:43:55.920
Emperor penguins do two months with no food, so
beforehand they have to accumulate fat, and once
0:43:55.920,0:44:05.200
they're sitting on the egg for two months, they
can't eat — so they switch to burning the fat,
0:44:05.200,0:44:10.800
no matter what, and even if they eat something,
that modality probably doesn't change. So it looks
0:44:10.800,0:44:18.160
like the FMD is that switch — once it switches,
the body starts using the fat that's been stored,
0:44:18.160,0:44:23.920
and it keeps going. This is why the ketone
bodies, even days later, are still high — a
0:44:23.920,0:44:32.080
major increase — because everybody else
is putting away fat, and you're using fat.
0:44:32.640,0:44:36.800
Interesting, really interesting. Well,
Valter, thank you so much for coming to
0:44:36.800,0:44:41.360
Stanford — we're really excited to see your
data and have you present at Grand Rounds,
0:44:41.360,0:44:44.320
and thank you so much for joining
us here on the future of medicine.
0:44:44.320,0:44:44.820
Thanks a lot.
0:44:46.880,0:44:51.120
The preceding program is copyrighted
by the Board of Trustees of the Leland
0:44:51.120,0:44:57.360
Stanford Jr. University. Please
visit us at med.stanford.edu.