Tip:
Highlight text to annotate it
X
>>Again, I want to thank everybody for coming today.
This is-- I've been looking forward to this myself for some time as these are--
we have two of the legends in our field here today and as you know,
neither one of them is a publicity seeker and getting them to agree,
even though they didn't know quite the scope of what we're going to do-- [Laughter] --
was a challenge, but I'm pleased that they were willing to do this,
and what I'm hoping that's going to happen today is that we hear a little bit
of oral history of our discipline.
This is something that all three of us experienced when each one of us went
to Harvard and-- at different times, the--
but in every instance, it was still a relatively small program there and they had a lunchroom
in the department and people would come to the lunch room.
If they were there and didn't have anything else going on so people come and go,
but it was students, fellows, faculty, visitors and it was one of my most vivid
and best memories of that whole experience, because, yes,
you get into high level discussions of epi methods and talks about a variety of things,
but you would also hear from people about what was going on, what they had gone
through other people in the discipline, you would get these stories
and it was actually a real oral history of other discipline, and so much so that
by the first time I met Richard Doll and Ernie Wynder, I knew more about them
than probably their mothers did-- [laughter] actually.
But the-- and I think what's happened in our discipline has gotten so big
and training programs have gotten so big I've--
the distinctive [inaudible] that that doesn't happen anymore and people go
and they attend their classes and they do the seminars and they find out what's going on today
and what they want to do in the future, but we've lost this appreciation of the history
and quite frankly, since most people date the real start of cancer epidemiology,
Percivall Pott with the exception here-- but mainly starting in the late '40s,
so we really only have about 65 years of cancer epidemiology and we havetwo people here
who have been here in the discipline for more than 50.
So they basically almost have-- and within their experiences early on,
and they knew all of the originators and worked with them,
so I'm really hoping what the atmosphere that established today is
to have them all pretend we're all in lunch room on Huntington Avenue-- [Laughter] --
And I've given them questions so I don't--
I'm not ambushing them, but I'm also encouraging them to take off from the questions to use it
as a platform if they want to talk about somebody or something or whatever it is,
it is just a kind of a stimulus for them to get into a discussion.
At the end, I will also ask if anybody has any questions that they'd like to ask them
from the audience, so-- with that is a long preamble and the--
from the people side of things, I think we can start with David, --
who were your most important mentors and how did they shape your career.
>> There was one mentor and his name is Abraham Lilienfeld and I was assigned to him--
I was a member of the Epidemic Intelligence Service in my second year,
he had requested someone who had a training in oncology to collaborate with him
on a multi-hospital case control study of cancer of the male breast and so Alex Langmuir,
who was the Director of the Epidemiology Program at CDC in the EIS program and DA Henderson said,
"We've got just the guy for you."
So I went up to Johns Hopkins and the chemistry between Abe Lilienfeld
and myself was just absolutely perfect.
And we did the study and he persuaded me to abandon my notion
that I would be practicing oncologist [laughter] and to pursue epidemiology in public health,
and that relationship that we developed was sustained throughout my career
and throughout his entire life so that there were periods of time when he would call on me
to join him-- one exciting time was after I returned from the Harvard School
of Public Health, he said, "Why don't you join me as a staff member
at the President's Commission on Heart Disease, Cancer and Stroke?"
That was during the time when President Kennedy had been assassinated and President Johnson
and Mary Lasker were instrumental in putting this group together.
Debakey was the chair of the commission and it was a visionary plan
for really having regional medical programs throughout the country,
biomedical research institutes, training, patient care,
facilities for these so-called killer diseases and led to a two-volume report,
a legislation that was passed by President Johnson, but no funding,
so it was in principle a visionary plan and that we see developed
through this comprehensive cancer centers as an example throughout our nation.
Abe Lilienfeld was a visionary and it's hard to imagine at that time,
there were a few iconic figures, Brian MacMahon, Abe Lilienfeld,
those were the two principle figures that I can think of
and maybe some lesser luminaries that I certainly knew.
And Lilienfeld began as a person who was going to be an obstetrician and then went
to public health service, went up to Roswell Park,
headed up there as biostatistical epidemiology unitthere and then came down to Hopkins
and created the first Department of Chronic Diseases
and then ultimately the Chair of Epidemiology.
He founded the American College of Epidemiology.
And many of the awards that we give in our organizations are in his name, you know,
so we are to him throughout, but he was my sole and principal mentor.
>> And so you can appreciate the fact that some of these names just roll of the tongues here,
DA Henderson, famous for eradication of smallpox in the-- on the-- in the world.
[Laughter] Debakey, famous cardiac surgeon who published a case control study,
but he didn't have the cleverness to call it that, of lung cancer in the late 1930s
that totally indicted smoking about 10 years before Doll and Wynder did.
So Joe, who are your most important mentors and how did they shape you?
>> Well, I guess in addition to David of course, the person that really encouraged me to go
into epidemiology was a-- the Chairman of the Department of Medicine
at Memorial Sloan-Kettering Cancer Center whose name was Rulon Rawson.
And Rulon was a thyroidologist and in my experience, thyroid cancer specialist have--
seem to have a special interest in epidemiology, and I'm not entirely sure why that is,
but they seem fascinated by epidemiology, and Rulon was one of those.
And my great interest at that time was in cancer--
clinical aspects of cancer particularly looking at the patterns
of disease associated with cancer.
And we would go on rounds and Rulon would see the patients and I was responsible
for guiding him to see the most interesting cases that we have on the wards.
And Rulon one day said to me, "Joe, you're thinking like an epidemiologist,"
and I didn't know what to make of that because epidemiology wasn't on my radar screen really
at the time and so he encouraged me we have to go
to the library and read up on things and I did.
I still couldn't quite figure out what he was talking about so-- [Laughter] --
but he says it's the patterns that you're always focusing on and that's epidemiology,
and at the time where the Vietnam War was heating up,
and fortunately David was a senior fellow in Medical Oncology at Memorial and he told me
about his experiences at the epidemic intelligence service at CDC and--
but, what interested me was NIH and so Rulon at the last minute,
my tenure as a chief resident was running out and he said,
"You can't be chief resident all your life.
You have to do something else and why don't you go to the Cancer Institute."
And he pulled strings and somehow managed to get me in,
although usually there was a waiting period of 2 to 3 years, so when I arrived at NCI,
Bob Miller was the Chief of the Epidemiology Branch and he recruited me and he was different
from any other epidemiologists that I've ever met.
He was extremely clinically oriented and he sort of paved the way for someone like myself,
who was very clinically oriented and to discover what the contributions I could make
to epidemiology.
At the time, there were many statisticians at NCI, lots of them,
and they were all outstanding, but they were very few people who were medically trained.
So Bob was really a very key figure in my development.
>> Thank you.
So start with Joe this time for the next one.
Excluding each other, your mentors and others in this room, [laughter] who were the 3
or 4 cancer epidemiologists you have known-- or 2 or 3, however many you want,
who have made the most important contributions to our field and what were those contributions.
>> Well, I think Richard Doll really stood out.
He is a towering figure in cancer epidemiology and in epidemiology in general.
He was, along with Ernie Wynder, the first to really pin down the relationship between smoking
and lung cancer and-- but his contributions extended well beyond that particular finding
which is probably the landmark finding in cancer epidemiology.
He went-- he looked into occupational cancers.
He was really the first-- I believe the first to carry out a cohort study of asbestos workers
and he had other occupational carcinogens that he discovered.
Radiation was an interest of his and he worked with Court-Brown to study the patients
with ankylosing spondylitis and quantify the risks of cancer associated with that exposure.
He had tremendous influence over the field, all the cancer research generally and this--
just as an illustration, it was an experience that I have with Richard Doll.
He invited me to co-edit a book that he wanted to write on cancer trends and it's hard
to say no to any request from Richard, so I agreed to do it on one condition and that is--
he said that he would take care of all the European authors,
I would take care of the North American ones and I said, "I will do it if you write the letter
of invitation," because working with David on our volumes, I knew how difficult it was
to get-- sometimes to get authors to contribute and also to send
in their manuscripts by a certain deadline.
[Laughter] So I said, "I will do it if you write the letters," and he said, "I'll do it."
He wrote the letters, everybody accepted on the spot without any hesitation,
and the manuscripts were on time.
[Laughter] Not a day late, it was just incredible.
So that was a-- and I got to know him quite well during that time.
And one other person that I would like to mention is Bill Haenszel who was Chief
of the Biometry Branch when I arrived.
His work on migrant studies I think was so key to understanding of cancer etiology.
I mean he really was so masterful in the studies that he carried out
and I think he deserves a tremendous amount of credit.
>> David?
>> Yes, well I think it's not surprising that we would agree--
[Laughter] I want to mention another name.
We only corresponded at the time near his death in 1991 and that was Austin Bradford Hill.
I don't know if you're familiar with his name, but he was an extraordinary man
and one who really-- not having worked closely with Richard Doll in the initial studies
on smoking and lung cancer in Britain and the British medical physicians, but his introduction
of Principles of Medical Statistics into the literature
that impacted not only the medical community, but also as me a young student.
I used to read his textbook, as well as a series of articles--
he wrote about 17 articles in The Lancet in the 1930s and then probably his first edition
of the Principles of Medical Statistics.
It's a beautifully written book, and it went through 12 editions,
and so I thought that he had an enormous impact on the medical profession which tended not
to think numerically or quantitatively, and he kind of laid
out what I would call the introductory principles of biostatistics.
He was one of-- was a strong advocate of randomized clinical trials
and how they should be used in making decisions about efficacy of treatment,
conducted the first trial of streptomycin on tuberculosis.
He also, as we know, put forward these principles,
guideposts about how we make judgments about causal inference.
This followed after the '64 Surgeon General's Report that we often think
and invoke Bradford Hill's principles bywhich we evaluate data and the plausibility and coherence
of the data with respect to inferential reasoning.
I also want to cheat a little bit here--
>> Yeah.
>> And list a bunch of names because I think we need to really bow to the individuals
who introduce methodology into our field and we know about Nathan Mantel and Bill Haenszel,
Jerry Cornfield, there-- and of course Norman Breslow and Nick Day.
I think also Ken Rothman and Sander Greenland.
I mean these individuals have made enormous contributions to our discipline.
We build upon them, we use their body of literature,
and it has advanced our field into a rigorous science.
>> Thank you.
And maybe we'll go to David again here that--
other than the contributions noted for the luminaries you just discussed,
what findings over the last 50 years do you think are particularly notable?
>> Findings, we are down to here [laughter].
All right.
>> It just links to the other questions.
>> Oh sure.
Well clearly, the literature on tobacco and carcinogenesis,
the literature on biologic agents and cancer that was driven by the new technology
that we had to make compelling arguments about associations with different biologic agents.
I think the human genome project and the relationship
with molecular epidemiology is enormously advancing field that will I'm sure be part
of the evolution of DCEG and as well as the discipline across our country.
If I could-- we also mentioned some other things that are important contributions to our field.
The IARC Monographs, a body of literature that have been enormously informative and helpful
to us, indeed the IARC volume 100, that will be-- it has a series of these committee reports,
A through E should be reviewed by everybody.
It really brings us up to date about all of the lifestyle and environmental risk factors.
And I think the EPIC studies, E-P-I-C,
that are being conducted among the European countries, the cohort studies.
These are-- a body of literature, and I'll also mention the health professional cohorts
at Harvard have informed us enormously about associations, I can go on and on,
but I think you wanted to be just the-- [Simultaneous Talking]
>> That's fine.
That's great.
>> Yes.
>> Joe?
>> What's the question?
[Laughter]
>> I Other than the contributions noted for the luminaries that you've discussed,
what findings over the last 50 years do you think are particularly notable?
>> Oh there's so many, Bob-- when I got here , which was 1962,
exactly 50 years ago, cancer seemed like a black box.
There was-- except for smoking, and the hazards of smoking in particularly lung cancer,
very little was known about the causes of cancer.
And then there was this succession
of epidemiological discoveries that seemed to appear.
Part of it was due to the waves of fashion, ideological fashions,
that we went through in the 1960s, the emphasis was on viruses.
I recall-- the first week I was here, the only other person in the branch was the branch chief,
Bob Miller and he was always traveling and so I would get calls.
And one call that I got that really took me off guard was a call from the head
of our virus oncology program, who said, "Joe, you have to go to Green Bay, Wisconsin."
And I said, "Well, I couldn't think what that could be.
I knew about the Green Bay Packers [laughter], and but why go to Green Bay?"
He said, "There's a cluster of cancer, particularly leukemia in children and you have
to go and get some specimens for us to look at".
So I tracked down Bob Miller and said what, what's this about?
Then I realized that everybody in the Institute practically was thinking viruses are the cause
of cancer.
And if we could only study these clusters that were occurring or thought
to be occurring, we would get the answer.
So Bob gave me a wise advice.
He said, "Refer them to CDC," [laughter].
And all I can say is, "Thank God, [laughter] for CDC, they have done so many things for us."
And they were very good at studying clusters.
I'm not sure much came of these particular types of clusters.
And then, well this was in 1960s, in the 1970s, fashion changed
and it was environmental and occupational hazards.
In all this time, genetics was really underground.
There were very few people studying genetics.
There was Henry Lynch, studying families one after the other,
but hardly anybody was interested in genetics.
1980s, lifestyle took over, particularly diet, and there were debates as to what was
in the diet that was causing cancer.
Then in the 1990s, the genomic revolution occurred
and finally our basic scientists were coming up with the tools
that we could use in epidemiology.
And this was a-- there was an explosion of work in family-based studies
with high-penetrant genes, and then GWAS, exciting findings from GWAS
on the low-penetrant common genetic variants, but then there were also individual discoveries
that all added up, the HPV and cervical cancer by zur Hausen and this was followed
up by epidemiologic studies carried out by Nubia Munoz
and Mark Schiffman, and others in our group.
Hepatitis B by Palmer Beasley and relationship to liver cancer,
H. pylori came as a big surprise and-- but that was a major discovery,
in relationship to stomach cancer.
And then asbestos was a big deal and estrogens, thanks to Brian MacMahon, and Bob Hoover,
we learned a lot about the role of estrogens in breast cancer.
And radiation, I can't forget radiation, the work of Gil Beebe
at the Atomic Bomb Casualty Commission,
now called the Radiation Effects Research Foundation,
the classic study of the 'N' bomb survivors which still goes on to this day.
And so much, passive smoking was I think a major finding because it really, in my mind,
underscored the value of epidemiology, the power of epidemiology to identify low-level risks.
At one time, we thought of for just less than 7-fold, didn't mean much, then it was 3-fold
and then suddenly we were down to 1 and 2-fold.
And then GWAS also has shown us that with the large enough studies in replication,
we could identify very low levels of risk.
So-- and the Doll-Peto monograph of course was a huge undertaking, a tour de force,
which sort of put things in perspective, and a Surgeon General's Report
in the 1964 was a classic and it was interesting, but even after it came out,
there were still people at NIH who did not believe
that cigarette smoking was a real health hazard.
So, and people in power [laughter] at NIH.
So, anyways, that's it Bob.
>> That's great.
Okay, can we lighten up the conversation a little bit.
[laughter] It's onto-- in the personality area over your career,
who were the most notable characters in our field and what made them so interesting.
Okay? [Laughter].
>> Well, you almost know what I'm going to say.
[Laughter] Ernst Wynder.
[Laughter] Ernst Wynder, when he was a medical student, at Washington University,
was approached by Evarts Graham, who was a thoracic surgeon,
about doing a study on tobacco and lung cancer.
And Ernst took it over, did a case control study and published, while in medical school in 1950,
a paper that underscored the association.
If we would take that paper and put it before our students today, they could critique it,
that's good, but no less the association was so compelling that is was hard
to imagine any serious source of bias or confounding or whatever.
He then came to Memorial Sloan-Kettering Cancer Center and was enormously productive.
Ernst, who was Germanic in his manner, was flamboyant, charismatic
and at times infuriating [laughter].
But he had endless ideas, the only issue was, you pursue them [laughter].
And-- but he was enormously productive, hundreds and hundreds of publications,
and he would do these case control studies, have these team of interviewers within hospitals,
particularly Memorial Sloan-Kettering.
He then had this laboratory that did tobacco carcinogenesis studies with Dietrich Hoffmann
and his group and began to identify the chemical constituents of tobacco smoke
that were carcinogens, promote as co-carcinogens and a body of literature came out of there.
He was one of the early proponents of the concept
of what he called metabolic epidemiology, which we have different terms for now,
but where he wanted to bring the laboratory in the study of metabolites into analytic studies
and hypothesis testing studies so that there is more mechanistic implications
of associations that we were reporting.
And he created this American Health Foundation, which was a basis for the laboratory studies,
as well as these health educational initiatives that he was advancing.
He had enormous ability to get funding.
He had access to offices within NCI and a most remarkable way
to extract funds from [laughter] decision makers.
And he also was not beyond bringing a Hollywood starlet
into our professional staff dining room [laughter] and saying, "Hi boys," [laughter].
And this beautiful model would come into the room
and you'd keep shaking your head about Ernst Wynder.
So he was a notable character.
[ Laughter ]
>> Yes he was [laughter].
Joseph?
>> Well, you know, David took the words out of my mouth.
[Laughter].
Nobody that I know could somehow manage such an active, both professional life and social life.
He was a bachelor and he thrived on, as David says, squiring very attractive ladies.
And I'll never forget the time he walked into the cafeteria at Memorial Hospital arm
in arm with Kim Novak [laughter].
This-- I mean, he just brought down the house.
[ Laughter ]
>> Okay, well that lightened it up a little, didn't it?
Okay, in your view, over your entire career,
which non-epidemiologists have been the most helpful to and supportive of our discipline,
and the second part of that question, if you choose to answer it,
is similarly who have been the most non-helpful?
[ Laughter ]
[ Inaudible Remarks ]
>> Well, I would say, we've been very fortunate at NCI to have support
of leadership at the highest levels.
And, the person who had the wisdom to make our program a division was Rick Klausner
who was the director at that time and I think we owe him a debt of gratitude.
He was a basic scientist, but somehow he had a breadth of vision which was truly remarkable.
I think at the international level, John Higginson, also a character,
and a pathologist who was the first head of IARC, realized that the opportunities
for an international agency like IARC was in epidemiology,
and he quickly assembled a really quite powerful group of scientists
who were either epidemiologist, or statisticians or interested in the field.
And there were some laboratory people, but they were all plugged in to epidemiology.
So, I thought he had it quite a bit.
And, who else?
Well, all of the statisticians, I mean David mentioned Bradford-Hill, Jerry Cornfield,
Nate Mantel, also, a bit of a character with his eccentric style, but a brilliant statistician
who worked closely with epidemiologists.
So that's--
>> I thought I was very fond of Arthur Upton who was one of the directors here.
He created the fellowships-- career awards actually-in preventive oncology.
I was one of the early recipients of it and I--
and Arthur did some wonderful work in radiation carcinogenesis.
I thought he was understanding of epidemiology and I know he was supportive of work that I did
and I'm sure others felt the same way.
You know, in terms of support from our discipline,
I want to mention Oxford University Press.
When you think of a publishing house that has promoted a body of literature
that reflects the best and brightest of our colleagues, Oxford has been right
out there and-- in a positional leadership, and I enjoyed working at that time
with Jeffrey House, when I did for some of the earlier editions of our textbook,
and found him to be very helpful in facilitating awareness by Oxford of the importance
of epidemiology or more recently as Jones and Bartlett,
but I really think Oxford University Press deserves some recognition by us.
>> You know most of them didn't answer the second part
of the question, but they are much too--
>> What was the second question?
>> What was it?
Who were the most non-helpful?
>> Non-helpful?
My parents told me early on, never to say bad things about anybody.
But I'll say there may be a category of people, some of them are generalists,
some of them are lawyers [laughter], a pure scientist who has magnified the limitations
of epidemiology, which we're all familiar with, while downplaying the strengths of the field
and the successes that we've had.
And of course, there are scientists
that have been closely aligned with special interest groups.
Some of the scientists, are very capable individuals, and--
but in general, the truth always comes out.
>> Okay, so David, what epidemiologic findings, either yours or others,
were the most surprising to you at that time and why?
>> Let's see.
I think Joe alluded to one: H. pylori.
That was such a dramatic shift in a paradigm of our thinking about duodenal ulcer
and across the linkages with neoplasia as well, that was truly remarkable and then perhaps,
some of that inverse relationships with adenocarcinoma
in the esophagus which I'd like to see clarified.
So, what Warren and Marshall achieved is such a superb example of believe
in what you believe is right and correct, and also anticipate that you will be vilified
by the profession, not embraced by them, because you are going contrary to dogma,
and to have that ability to advance concept that was very, very dramatic.
I think some of the complexity relationships
with steroid hormones has been very interesting to us, you know.
It's like good cop, bad cop, to some of the associations
with combination the OCs have been dramatically protective for endometrial and ovarian cancer,
particularly the endometrioid, carcinomas of the ovary.
And at the same time, the action on the-- on breast ductal epithelium is quite different
than on the endometrial epithelium.
Some of the conflicting findings about folates has been very interesting, you know,
because of its multiplicity of effects both on nucleotide synthesis,
as well as on the methylation pathways and the timing as to
when you introduce supplemental folates, as to whether individuals may have precursor lesions
that we may find paradoxically that there are adverse effects
and not always the beneficial effects.
Those are three examples that come to my mind.
>> I would say one of the most surprising findings --
was the DES story, the discovery of vaginal adenocarcinomas in the daughters
of women exposed to DES during pregnancy.
This really I think caused the paradigm shift in our thinking about the importance
of early life exposures, and-- which is one of the toughest areas to study,
but I think this really set the stage for a research into the role of early life exposures,
not just for a child with a cancer but for adult cancers as well.
And then another surprise I think in my mind, there always-- seemed to be a resistance to--
at NIH and elsewhere, to looking at the role of obesity in cancer.
And now we just see a remarkable array of tumors associated with obesity.
And I think that was a real surprise.
I think the geographic patterns, the discovery of tremendous international variation
in cancer risk, and even our cancer maps which showed clustering of cancers around the country
and patterns that stimulated a lot of epidemiologic research.
Burkitt lymphoma, I think Burkitt's discovery of a belt of lymphoma
across Africa was a striking finding, and also paved in a way for a lot of research.
>> Okay, Joe, this is kind of a 3-part question, there's a been a whole lot of changes
in our discipline over the time that you have been in it.
So basically, as one from your position, what's been the biggest change and similarly,
what's been the best change and what's been the worst change?
Now, they may all end up being the same thing or they could be different?
[Laughter].
>> I think the biggest and idea-- the biggest change has been the application of biomarkers
and the move toward the genomic and molecular epidemiology.
This-- and this really required us to think more broadly about how
to conduct epidemiologic research and has led
to large scale collaborative studies in the form of consortium.
The realization that cohort studies had so much to offer and the Nurses' Health Study is
of course the primary example and I would say the best accomplishments and-- what was the--
>> Best change.
>> The best change-- well, I think of the new methods developed by statisticians
and methodologist in epidemiology has made our discipline more rigorous and robust.
>> How about the worst change?
>> I think probably the temptation to magnify the risks that resulted
from subset analyses has led to a lot of blind alleys.
And again, underscoring the importance of very large scale studies
and replication before publication.
>> David?
>> All right.
So I'm going to agree with Joe's insights.
I think one development that's positive and challenging for us are the importance
of interdisciplinary or trans-disciplinary research that compels us
at an epidemiology discipline to see the necessity of being able to communicate with
and collaborate with a variety of disciplines and appropriately testing hypotheses.
What this requires of us is that we are well grounded in fundamentals of these methodologies
and understand the language of our collegial disciplines,
so that we are not a second class player,
but we're really a co-equal in collaborative research.
And I think, some of the directions that we are going in now and the teachings at schools
to public health and training here within DCEG are important that we not be viewed
in a pejorative way as a kind of "soft, social discipline," but rather as a co-equal
in doing rigorous scientific research and risk assessment.
I think also the development of population registries is very important.
We know of notable examples of these because they not only are the basic requirements
for looking at trends in evaluating our-- the efficacy of cancer control interventions,
but also service and resource for the population-based studies
that are essential for us in studying associations.
On the neg-- the worst change.
It's hard to put as a worst change, the one thing I'm concerned about--
this was something that Abe Lillenfeld was concerned about too is
that we may fragment a bit into different multiple societies that we're associated with.
And that we would find pharmacoepidemiologist, neuroepidemiologist,
and all of different subdisciplines within the overall discipline.
We're all epidemiologists, and we have a methodology that we all employ,
though the content of it may change.
But by doing the fragmentation, we dilute, I think,
our visibility within the scientific community at times.
So that's concern I have about that.
>> Okay I'm going to-- while you're still warmed up, David [laughter], do you--
what do you view as your own biggest professional success, and the flip side of that,
what do you view as your own biggest professional disappointment.
>> The textbook?
[Laughter] With my buddy.
That certainly is a source of enormous pride.
Can I go beyond one?
>> Yes, please do.
That's a given, you don't have to-- [Laughter]
>> I felt very good about bringing the graduate summer session program
to the University of Michigan.
I had been teaching up in the University of Minnesota during the summer.
And when I became chair, I had the support of the dean of our school, and not the support
of my faculty, I must say, to bring the summer program and the invited faculty to teach.
And you know, over these years, we have trained and instructed and enhanced the skills
of hundreds and hundreds of health professionals.
And I'm really very proud of that achievement,
and even some of my faculty would agree with me here.
It was a plus for the department of epidemiology.
The issue for them was: why do we have these foreign agents come here to our school
and teaching, why should we embrace the program on our own when that was ridiculous
because the content of the curriculum there is so vast and so broad, we want to bring the best
in to teach and provide even new insights to our students,
who within our own program take advantage of the summer program as well,
so I think most have bought into it now, but I'm very proud of that.
And you know Joe is very much involved with me in the creation of the American Society
of Preventive Oncology, which was a recognition of preventive oncology as a discipline
that crosses all the different component parts of epidemiologic science,
and to bring together young investigators into a rather small club in there to have displays
of their research, to have panel discussions.
It was certainly a move that has continued to this very day,
and I think has been a plus for cancer epidemiology.
>> How about disappointment?
>> Oh, disappointment, I think it gets down usually to turf battles,
cross-institutional collaborations.
I felt this in New York City certainly, and even in Ann Arbor, Michigan when I would try to reach
out to Detroit, to institutions there, there was always that uneasy alliances.
So that if I wanted to develop a broad consortium of us, there was as resistance
between our community and the Detroit community and of course that's
where the SEER program was based.
And it was not easy to have access to that resource.
That's been my greatest difficulty, I think it takes a lot of skill in diplomacy,
and endless patience, and trust, and I was not always successful.
>> And Joseph, your success and you're-- success--
>> Success, I think with the help of a lot of people, developing a truly robust research
and training program in cancer epidemiology, one that really covers the waterfront
of etiologic risk factors that embraces both genetic and environmental factors.
And having a working hypothesis that cancer may be a relationship,
or an interaction between genetic and environmental factors.
I think this has probably been my biggest accomplishment.
The-- I want to say word about the books.
I think-- the history of the book.
David had a book that he published,
it was called "Cancer Epidemiology and Prevention" as I recall.
And I took the lead in a book called "Persons at High Risk of Cancer".
David's book was oriented toward cancer sites, mine was oriented toward risk factors.
And the book that I edited came out of a workshop that we had in Capers Game, Florida.
Those were the good old days.
[Laughter] Really, you could go anywhere and nobody would bat an eyelash.
That was a lot of fun.
But David came up to me after that and said, "Why don't we join forces
and put out a very comprehensive book that covers all aspects, that covers descriptive epi,
the-- talks about risk factors, talks about every cancer site known to man,
and then end with a chapter on-- or a section on prevention-- opportunities for prevention."
And so I just wanted to give that little history, because I think it was interesting.
And working with David has been a-- that was an enormous pleasure.
>> How about your biggest disappointment?
>> I think some of the-- what David says a little bit different.
There was-- there's always competition, there's always turf battles,
but they didn't bother me of that much.
It was the access to data resources that bothered me.
I could never quite understand why even within the Cancer Institute, getting access to--
the clinical trials data seem to be self-threatening to certain clinicians,
particularly in the Cooperative Groups.
It was a bit frustrating at times.
The-- I have a whole cabinet full of correspondence with groups
that had clinical trial out, but didn't want to share it.
Perhaps out of fear that we'd find something wrong with the drugs.
And I don't know, but I think our program had some difficulty there that was frustrating.
>> Okay and so, kind of a one leading question here, but epidemiology is one
of the few scientific disciplines where studies and/or the results can frequently threaten the
vested interest resulting in obstruction and critical attacks
by such interest and their consultants.
Joseph, in your career, what have been the most noteworthy such experiences--
>> Well the--
>> -- and how did you handle them?
>> I think the occupational studies program really takes the cake.
[Laughter] My first experience was a study of smelter workers exposed--
heavily exposed to arsenic, and the finding of excess lung cancer.
And this was a challenged by various groups.
And in fact, there were two other studies that preceded mine that had--
that had found no evidence of an excess of lung cancer.
So this was challenge, but challenged unsuccessfully.
It finally came down to a hearing at the-- at OSHA and at the--
it became very clear at the hearing that arsenic is a carcinogen,
despite the fact that there was no strong experimental evidence for this,
but based on epidemiology alone, the evidence was quite compelling.
But we've had so much experience
with the controversial studies that-- there's just no shortage.
I listed some here, breast implants, abortion and breast cancer,
number of pharmaceutical agents, radiation, electromagnetic fields and cellphones,
fluoridation of water, you can probably think of others that were very sensitive topics
that led to a lot of disagreement.
And-- but that's the nature of the game, and the truth always winsout.
>> Yeah, David and I were talking about this as we were coming in too,
and before I turn the question over to him.
It really is true that almost everywhere else in medical research,
findings are universally met with joy and advancing biology.
In ours and in toxicology, every finding we have assailed somebody's vested interest.
So there is always somebody who is very upset, and I think it's just kind of a nature
of what we do, but what has your-- been your experience?
>> That's very interesting.
I-- you have to know your institutional setting when you carry
out these industrial studies that are funded by industry.
And I had two parts to my career, one was at a private cancer center, and it became clear
to me, when you do a study and it's sponsored by a private industry,
be sure you look at the membership of your Board of Trustees, if they are averse
to what you're doing, they're going to be more powerful than you are as an investigator.
And indeed that was an example where I was quite sobered by the antagonism to a study
that I was doing of petroleum refinery workers, which was an enormous study
and could've had enormous potential.
But unfortunately, was not viewed to the same enthusiasm that I had for doing that study.
But then when I shifted to a public university and I did these studies with Dow Corning,
and did studies of breast implants, Joe mentioned as well as chemical solvents,
I had the umbrella and protection of my university, and my senior position there.
And it was so much to the benefit of the industry to our graduate students,
because the data we were collecting served as a research content for their theses.
And we were able to engage at the clinical community.
The breast implant steroid is an extremely fascinating one.
And with the large funding that I had, we had our own, many, many publications come
out looking at connective tissue diseases, as well as some of the chemical solvents.
So it's a little different with cancer, but in the environment of the graduate school
of public health, I was able to do these studies.
And I was protected-- I was protected really in terms of my academic freedom.
So we have to know where you're conducting your research, where your support is.
How solid it is?
And recognize where you're vulnerable or where you have the support really
of the leadership of your institution.
>> Okay, we're going to turn a little bit from reflections on the past to visions
of the future here, from with-- girded by all of this experience.
So then we'll go to Joe here.
Over the next decade, what do you think will be the most important opportunities
in cancer epidemiology to exploit?
>> I think the biggest opportunity is that we have in front of us the opportunity
to incorporate some of the novel technologies that are resulting from conceptual advances,
in molecular and metabolic sciences.
We now have opportunities for developing, and the applying high-throughput technologies
that will help us in exposure assessment, and in the discovery of susceptibility states.
So that's the biggest.
You just wanted opportunities--
>> All right.
Well, the next one is what the biggest challenges that we'll work on?
>> The challenges, really.
The massive data sets that are going to be emerging from these projects,
and the informatics that's required to cope with them, and the, another area,
another big challenge is the policy impediments, the need for protecting privacy
and confidentiality, which is important but also complicates our life
as epidemiologists in terms of access to data.
And another challenge, I think is going to be, and we're all aware of this right now,
is that the development of infrastructure that will allow us to carry
out molecular epidemiology, the specimen collections,
as well as the epidemiologic data that's required.
And then of course, the big challenge, and we're going to rely so much on statistical colleagues
to look at gene-environment interactions, and try to tease out pathways
that are essential to the development of cancer.
Trying to figure out the role of epigenetics in all these, which may provide the answer
to gene-environment interactions.
>> Opportunities and challenges.
>> I certainly agree with what that Joe has listed.
I would also add-- I think there are challenges and opportunities looking at the dynamics
of health disparities, looking at the issues relating to the multi-ethnic and racial studies
that require our discipline to dissect now the determinants of these disparities, and risks,
as well as in survival outcome of mortality.
I think there are global challenges, many opportunities that will be there for you.
And investigating the cancer patterns that are quite different
in different countries throughout the world that really will depend upon trained disciplines,
and professionals, which are lacking, in many of these limited resource countries.
I think there will be challenges in our being able to demonstrate efficacy
of our chemopreventative interventions.
Right now, there, I think expectations
and enthusiasm has exceeded what we've truly been able to demonstrate,
and I think more work will need to be done there.
Now, the other was challenges to overcome, I think Joe mentioned that.
I do think that one of the issues for us is, in the discipline, this came out in the discussion
after my talk, and also in this session with these fellows is,
what is the role of epidemiology in the dissemination
and implementation of our research findings?
Do we not have a responsibility as public health professionals
to pursue translational implications of the findings,
and associations that report in epidemiology?
I think those are some issues.
Okay.
>> I'm staying with you David for the-- referring to the serial natures ofsome studies,
and best of interest that you've come in on before.
In your opinion, are these experiences likely to get better or worse in the future?
And do you have any advice on how it might be managed better?
>> I can answer that very succinctly: they won't get better.
[Laughter] They'll just look a little different.
And I just think we need to stand together as a discipline,
and support each other realizing that, we will be targeted.
And we have to overcome that because of our conviction
that what we're doing is good, and right.
>> I think we have a special challenge at the [National] Cancer Institute,
because what we do gets us recognized much more that work that's emanating from Cancer Centers
of our universities because we're the [National] Cancer Institute,
because we are a national agency, and considered really the nerve center for cancer epidemiology
around the country, if not around the world.
The, one coping mechanism that we have, and we're very fortunate to have is
when we launch a potentially controversial study.
We have created expert advisory panels to review the protocols and the results.
And this has helped us enormously, and as a sort of a safety net.
We need to really, when you conduct a study, we need to be prepared for the long haul,
there's going to be delays, there's going to be obstructions that are placed in front of us,
so we just have to be prepared to conduct the best possible science at all stages,
because that's really what stands the test of time.
>> Okay, there is a supplemental question.
I think you're both aware of, but kind of a long lead-in, but for quite some time,
there were actually, and particularly early in mid-career for you, there were strong,
and frequently contrasting views of which were the best chronic disease epidemiology
training programs.
Leading to spirited debates among the graduates of different programs,
and the most contentious was probably Harvard versus Hopkins but there were plenty
of other participants, North Carolina, Berkeley, Michigan, University of Washington, and others.
And similarly, there were strongly held views about the so-called CDC epidemiology,
both strong advocates and most believe that it wasn't epidemiology, it was microbiology.
The level of interest in passion seems to have disappeared.
I rarely hear, spirited arguments about where did you go and why did you go there?
And so historically when young people have consulted you about where to go for training,
what has been your advice, and did that change overtime?
And are there currently any really meaningful distinctions between training programs,
or is it-- and basically is this been a good trend, or a bad trend?
Who wants to start with that?
[Simultaneous talking] [Laughter]
>> Yeah. [Simultaneous talking]
>> Well, it is clear that in the current world that no single department
of epidemiology can encompass all the disciplines that are essential for looking
at all the various disease problems that we have.
That may have been possible years ago, but each department will have one or maybe more
than one area, real focused on emphasis.
And it would be reflected in their faculty in their resources,
in their research that's published.
You think of Harvard, you think of cancer.
In our own school, when I came there as chair,
it was infectious diseases, communicable diseases.
And you could go across the country as this develop, and you could say well,
this is a strange thing, cardiovascular, there is diabetes, metabolic,
and nutritional diseases, cancer, and so on.
But no single department could encompass it all.
We didn't have a budget for it, we couldn't recruit a faculty for it, and it was important
to acknowledge that what we were doing, we were doing well, and we would give some recognition
of the diversity of public health problems that are disease-oriented.
But that our faculty will have certain strengths.
When I came there, we began to really emphasize molecular epidemiology.
But then, I was succeeded subsequently by a social epidemiologist,
and then we built up this enormous capability in social epidemiology,
our current chair is a social epidemiologist, Anna Diez-Roux who is brilliant, and wonderful.
So your strengths change.
And if I want to counsel somebody, there are a number of issues.
One can be pretty personal, geographic.
But there maybe, an area that they have a particular interest in, and you sort of think
about faculty matching with that student's interest, because,
if they go into the doctoral program, they clearly need a sponsor, someone who will support
that student, and compatibility of the particular school,
a department, and that student's interest.
So it's no longer just Harvard, and Hopkins and Michigan.
[Laughter]
>> Now, Joseph?
>> Okay. Well, our experience is really so different, at NIH, rather than departments,
we have institutes, and, that our categorical institutes that focus on a particular disease,
in our case cancer, but there would be benefits.
I think I'm sure of that.
I could collaborate more effectively with our counterparts at other institutes.
And there are some barriers to that that are-- that have to do with the turf,
but they can all be dealt with in one way or another.
But I think if we could work more closely with other, with epidemiologists,
and other scientists located elsewhere, at NIH, I think we would be much better off.
And it would be a much more cost-effective way of conducting business.
>> Okay, just a little follow up to that question, 'cause I, you know,
discussed this with you a little bit, that part of the enthusiasm in the past has been not only
for the departments but for specific individuals, the-- really the Harvard, Hopkins,
was willing to help the command discussion.
And they are actually-- when I look out there today,
I don't see any equivalent person known for training.
Known that you would go to this place because that person is there to train you,
that was how I ended up, were I did my-- I had an adviser at medical school,
who gave me three names at three different institutions to go visit
and it depended a little bit on what you ultimately wanted to study,
but it was, but it was based on the person.
And I actually don't see that right now, people who,
now maybe it's 'cause rotating chairmanships and that, you know, there's no circle legends
that stay at one place, but, so do you think it was just actually so in is it a good
or a bad thing that are we getting.
Is everybody getting good, or, David [inaudible]?
>> Well I think, if you're inquiring about an institution, and the department,
and the mentoring, and the training environment, it's a good thing to speak to the students,
rather than to get faculty names, and to get their feedback about, who's teaching?
Are the faculty teaching?
Or the graduate student assistants teaching?
How are the faculty serving to guide you in your research,
in your training, in your career development?
And that would be I think, an important source of information rather than,
just going by the reputation of a person that, you know, what they published,
and their eminence as an epidemiologist may not also translate to their preeminence as a mentor.
>> Okay.
>> All right.
The same in training [inaudible] here, if you had only one piece of advice to give
to a young investigator starting his or her career, what would that be?
Since you're warmed up, David who about that?
>> Talked a little bit about that with the fellows yesterday.
I had a really wonderful time speaking to them.
It's not one piece advice.
I have a few pieces of advice.
[ Inaudible Remark ]
[ Laughter ]
>> Some of this relate to their personal attributes, and qualities,
and approach to their discipline.
And I made a point, if you really want to be serious about it,
you better acknowledge the fact that you're going to be a life-long student.
That you're going to have to continue to learn, and continue to master your methodology.
And to experience passion and joy, why you're doing it, if you don't have it,
don't trouble yourself with all the years that you're going to have to compete for funding.
Also to encourage them to be questioning, to be critical observers, and ultimately,
and I mentioned this to them, to think out of the box, to think imaginatively about questions,
because it is framing the questions that's the real issue.
That maybe contrary to dogma, because recognizing what you teach,
what you have been taught today, will be perhaps not accepted decades from now.
So that you have to think of yourself as potentially innovative if you really want
to be very, very successful in the field, that's tough.
Particularly when you're starting out, because young investigators can't be risk-takers,
but as you get more established, I think, you might want to think about that
if something doesn't seem quite right to you, as you mature in your discipline.
That's a bit of advice, are we good?
>> Joseph?
>> Well, I think finding the best possible mentor, that you can have.
The mentor is so important in our field.
The reason that I singled out Rulon Rawson as having such a major influence on my life,
this was the thyroidologist who had sort of an interest in cancer,
and he was able to identify traits in me that I was not aware of or dimly aware of.
And then steered me in the right direction.
So finding a mentor that-- as caring, really caring is important.
And also, finding compatible collaborators who have complementary skills that I found
at working-- I was so reliant upon statisticians like Nathan Mantel
and others that were in our program.
And also carving out a niche where you can gradually evolve and turn
and become an authority in that particular field, I think is important.
We don't-- David and I had the luxury of entering epidemiology
as really general practitioners, since cancer was a black box.
So we-- there were all kinds of opportunities.
And as long as you had the imagination and resources to follow through on them,
and they have-- we have natural experiments going on all around us, high risk populations,
that we need to seize these opportunities to study these groups that they maybe in a country
in need and they maybe in developing countries, often in developing countries.
And the other thing is there's a fear or phobia on most of young people
about large scale collaborations, and consortia, and I wouldn't worry too much about them.
There are always projects that are spinoffs or add-ons or something to,
for a young investigator to carry out in, and they also should be encouraged to--
by their mentors to play a lead role that might result in first-author publications.
>> Okay, I'm going to give our two participants the chance
to have the last word, whatever they want to say.
But before I do that, maybe, I've said-- all right.
One or two burning questions that somebody thinks they'd like them
to address before we start to wind it up.
I'm willing to entertain.
Okay, so you want to keep this on schedule.
That's good.
So David, any last thoughts you want to--
>> I want to thank you, Bob.
I think this required a lot of soul searching [laughter]
and I gave a lot of thought to your questions.
And I was-- and we didn't collaborate, you know.
We did not share what our thoughts were.
So that was also of interest to me [laughter].
I appreciate the large group here who are willing to listen to us.
>> Joseph?
>> Well, Bob I want to thank you for, as the host of David's visit,
and also for organizing this session.
If I knew it was going to be a videocast, I would have never done it.
[ Laughter ]
And so, I think there are just so many opportunities still in cancer.
We don't have, despite the fact that we know so much more than we did 50 years ago,
there's still a lot that remains to be learned.
And I hope-- and I think for those of you who are in a division of epidemiology and genetics,
I think the opportunities are there, and we just have to seize every opportunity
that comes along, not every opportunity, but the best
and most exciting opportunities that are out there.
>> And I just want to thank Joe and David for this, I think, as I said,
I think oral history of our discipline is important.
I am hoping that you all go away with little nuggets that you will remember
for a very long time, and help you when you're on trying
to tell your mentees how this discipline came about, and how it evolved.
And particularly, I want to thank you two for being so forthcoming.
And I recognize the fact that it's hard to do this, and that's why I didn't tell Joe,
it was going to be videotaped, until I discussed yesterday
that he probably wanted to wear a single-colored shirt.
And he said why is that [laughter]?
And I think-- by that time, I think it was too late so, [laughter] anyway,
thank you all very much for that sharing.
[ Applause ]