Biopsy For Prostate Cancer

>> from the library ofcongress in washington, d.c. [ silence ] >> well, good morning all. >> morning. >> it's a nice crowd here. thank you for coming and, for thissession we're having with dr. platz on prostate cancer update. we've done several sessionsin our lcpahso brown bags with the men's health month,and i see several of the people

who participated inthat here today as well. and, so, this is bring us fullcircle now that we're in the month in which we observe prostate cancerawareness and try to find answers. we have an expert with us fromthe johns hopkins bloomberg school of public health, and as we werechatting here, she was telling me about the many departments. and, so, she's deputy chair ofthe department of epidemiology, and she'll share with you whereall the inspiration comes from even as she shares some of the newupdates and things that are going

on in the world of prostate cancers. something that is important,not just for the men in room, which i'm glad to see several, but,also, the women who live with them, care for them, and are relatedto them in one way or the other. so, it is my pleasure, as chiefof health services, to welcome all of you here, and, particularly, towelcome dr. platz and my colleague, that's tomoko steen fromscience and technology with whom we're partnering,will introduce dr. platz to you. thank you.

[ applause ] >> thank you so much, dr. charles. it is my pleasure to introducemy dear friend, elizabeth platz. she's already introduced theprofessor at epidemiology at the johns hopkins university. she studies association of genetics,in epigenetics, factors as well as saturating markers of under[inaudible] city, information, and oxidation with prostate cancer. she has study is notonly doing science,

but she wants to educatepeople as well as the, looking at the associationlifestyle. so, her team has sociologists,not only epidemiologists, and this is a good example ofthe transformation of medicine. we started about two yearsago, dr. jim watson came and we had a pernell discussionon transformation of medicine and this is a continuation of thediscussion on the transformation of medicine, and we also had gotmicros talk earlier this year. before further due, i'dlike to introduce dr. platz.

>> so, thanks to all ofyou for being here today, and i'm very excitedto speak with you about my favorite cancer,prostate cancer. so, it might seem odd that awoman studies this disease. i study this disease because we knowvery little about what causes it. and, so, my whole goal in my careerhas been to try to find risk factors that men can do somethingabout to reduce their risk of developing the diseaseand for men who already have prostate cancer,things they might be able to do

in their lives to reducethe risk of poor outcome, including dying of the disease. okay. i do want to indicate to you that i do not have anyfinancial conflicts of interest. all right. september is national prostatecancer awareness month. you all know that blue is thecolor for prostate cancer. how many of you know where yourprostate is, if you're a man? okay. you probably, as youget older, you become more

and more aware of your prostate. some men do experiencesymptoms related to benign prostatic hyperplasia. it's an organ that isinvolved in reproduction. it's an important organ to have,although, clearly, once it's removed in the setting of prostate cancer, a man can live a perfectlyfunctional life. okay? the lifetime risk ofdeveloping this disease is one in seven, and thisis the lifetime risk.

this is not the risk at any moment. so, it is a commoncancer, and, right now, there are about 2.8 million menwho are living with this disease. okay? so, they've survived,they've had their treatment or they've undergoneactive surveillance, and they're living with it. okay? so, today, i wantto cover four areas. the first is the national andglobal burden of this disease. i want to tell you a littlebit about what we know

about what may cause this diseaseand what we might be able to do to prevent poor outcome, nottreatment, but modifiable factors, what men may be able to change tochange their risk of poor outcome. and i also want to show you alittle bit about ongoing research on both the causes of this disease and about prognosticfactors for poor outcome. and, then, we'll talk verybriefly about the controversies in the early detectionof prostate cancer. okay? so let's start withthe burden of this disease.

so, i do want to tell you wherewe get our national statistics. so, you'll hear statisticsin the news about prostate cancerbeing number one. so, where does thatinformation come from? well, the american cancer societyproduces summery statistics for us every year, which isvery helpful for people like me. it's important for peoplewho do program development for programs related to educationabout screening and so on. those data come from thenational cancer institutes,

specifically the seerprogram, the u.s. surveillance, epidemiology, and results program. and, so, we do have in thiscountry a system of surveillance. seer started in 1973. it doesn't cover the entirecountry, but it does include about 28 percent of the country. places are sampled, cancerregistries are sampled to have diversity, and, then,through statistical means, national estimates are produced.

and, then, how do thedata get to seer? well, we do have a u.s.cancer surveillance system. in most states, diagnosingphysicians and mythologists arerequired to report to their hospital cancer registryas well as private mythology labs. these enter the state and regionalcancer registries, and, then, they enter the seer programor the national program of cancer registries,which is a cdc program. and, then, in addition,information on death

from cancer comes fromvital statistics. okay? so, let's look at where youcan look up details on the burden of cancers, and so, again, oneplace to look is the seer program. they have a lovelywebsite that's interactive. you can look at the cdc'swebsite, the national program of cancer registries, or you canlook at the annual cancer facts and figures produced by theamerican cancer society, and, so, we can make these slides. i guess they'll be available on theweb so you can go to these links

if you'd like to reviewthese statistics. okay. so, new cases in the us. so, as i mentioned, prostatecancer's the number one diagnosed cancer in u.s. men. okay? aside from non-melanomaskin cancer, which does not get recordedin cancer registries. okay. this amounts to aboutmore than 220 thousand men. this is an estimate for this year,2015, and for those of you who care about incidence rates, it's about a138 per 100 thousand men per year.

okay? let's look at the trendsin this cancer over time. so, this graph shows 1975 to 2011. again, these are slides fromthe american cancer society from cancer facts and figures. here's prostate cancer incidence. okay? you can see in thelate 1980's, early 1990's, we had this tremendous peak. this is due to psa-basedprostate cancer screening. okay? and, then, youmay know that in 2012,

the u.s. preventive services taskforce changed the recommendation. so, now, the recommendation is against prostate cancer screeningusing the psa test, and at the end, i'll talk about whythey made that decision and why it may be controversial. okay. i also want to show you thefive-year cancer survival rates [inaudible] prostate cancerback in the mid to late 1970's. the rate was about 68 percentfive-year survival, and, then, by the late 1980's, it was upto about 83 percent, and, then,

today, it's at a 100 percent. and we'll talk about why it'snow a 100 percent, and, again, the 68 percent and the 83percent were before the psa era. okay? and for prostate cancer death, death in prostate cancer'sthe number two cause of cancer death in u.s. men. just under 30 thousand men willdie of this disease this year, this is the estimate, and,again, for those who like rates, the mortality rate is about 21per 100 thousand men per year,

and you can see it's two behindlung and bronchial cancer. okay? which far exceedsthe proportion of deaths from prostate cancer. okay. let's look at thetrends in death from cancer. let me point out first, lung cancer. lung and bronchus, you can see thistremendous rise and then decline. this is related, almostfully, to smoking. okay? and changes insmoking status over time. this is one of the most impressivecancer-related public health

benefits that we have had. okay? it's very remarkable. now, let's look at prostate cancer. we can see this bumpthat's concurrent in time with the peak incidents. you would think that screeningshould reduce mortality, we did have that peak. that's probably due to betterattribution of cause of death because it's probablywhat's going on there.

and, then, this decline isprobably due to two things. one, it could be due to psa-basedprostate cancer screening, so detecting early, treating early, and it's also likely due thebetter treatments for men who have advance disease atthe time of diagnosis or who, after the diagnosis,develop advanced disease and then need to be treated. okay? so, let's talkabout the global burden of prostate cancerusing statistics again

from the american cancer society. for these data, they receiveinformation from globocan, which is part of theinternational agency for research on cancer, so iarc. and iarc uses a variety of means toobtain estimates of incidence rates and mortality rates forcountries around the world, including excellent registriesin some places, and, then, in other places where therearen't registries, registries are of poor quality, iarcuses information

from neighboring regions. okay? and they impute. okay. so, per the americancancer society, this cancer, prostate cancer's the secondmost common cancer globally. there are about 1.1million cases in 2012. a big chunk of them, themajority of these cases, are diagnosed in countrieslike the u.s., so economically-developed countries. and the incidence of thiscancer varies tremendously

across the globe, and this is inpart due to differences in the use of psa-based prostatecancer screening. okay? also per theamerican cancer society, about 300 thousand deathslikely occurred in 2012, again, these are estimates. it is the fifth leading causeof cancer death in men globally and the highest prostatecancer death rates are found in the caribbean. okay? and it's unclear why this is.

so, one of the most importantdifferences in the global burden of cancer is based oneconomic development. so, when we look ateconomically developed countries, prostate cancer is, again,number one, and, then, for death, it's about number three. when we look at economicallydeveloping countries, you can see prostate canceris much lower on the list. now, number one, forincidents and mortality in economic developingcountries is lung cancer.

okay? so the global burdenof lung cancer related to smoking is a disaster, actually. okay? all right. so, now let's move on to knownand suspected risk factors for prostate cancer and pooroutcomes in men with the disease. okay? so, what causesprostate cancer? so, although we've beenstudying this disease for decades and decades, very little's knownconclusively about what causes it. what we do know is age, soolder age, the family history

of prostate cancer, meaning a fatheror brother who has the disease, and in the u.s., beingof african ancestry. so being an african americanman in the u.s. is associated with a substantially higher risk of developing the diseaseand dying of it. okay? none of these threefactors is modifiable. okay? that's a problem, right? the goal of public healthis to prevent disease. if we don't know what the riskfactors are, we cannot come

up with strategies for primaryprevention, and, of course, we prefer primary prevention. we don't want peopleto get disease, right? we don't want to just pickit up early and treat it. we don't want men toget this disease. okay? and, then, to showyou some statistics. so, the blue line is prostatecancer incidence over age. so the x access is age,and so you can see the risk of prostate cancer increasesdramatically over decades of age.

this is not a disease of young men. okay? the typical age ofdiagnosis is about 66. you can also see thatthe risk of death from prostate cancer increasesnotably with age, and the median age at death from prostate cancer in menwith the disease is 80 years old. okay? so you might askme what's going on here. why is it as men continue to age, it looks like there'sa decline in the risk? well, you have to remember theseare not the same men ageing

through time. these are birth cohorts of men. okay? and, so, the men out here,the men who are currently 75 and older are from birthcohorts some time ago. we've had changes inexposure factors. so either exposure at all or theprevalence of exposure over time, and so what we're probably seeinghere is not suddenly is susceptible or a lack of susceptiblesusceptibility in these men. it's probably that thenature of the exposures

that they experienceproduced a lower risk of developing the disease. it's something very important toknow when we look at age effects. okay? so, now let's look atthe second known-risk factor for prostate cancer, family history. so, this is work that we did alittle bit ago looking at a group of men who are calledhealth professionals. it's a cohort. the health professionalsfollow-up study.

so, large cohort of men whowere other than doctors, they were already enrolledin a clinical trial, and, so, the men who were recruitedwere dentists and veterinarians,optometrists, and so on. and, so, like all the otherstudies that have been conducted in this large cohort study,what we observed was that men who had a family history, again, afather or brother with the disease, had twice the risk of dyingof this cancer, and, also, had an increased risk ofdeveloping the cancer.

it turns out, we see this for mostcancers, not just prostate cancer. people who have a family history,again, father, brother, mother, sister, have about twice the risk. we have no idea why that is, butit's consistent across cancers. okay? so, this type offinding led to this search for inherited geneticdifferences among men that might explain the familyhistory finding, and, so, i would call this a mad dash. it started very slowly, individualchanges in the genome were targeted

to evaluate associations withprostate cancer, and, then, the field of genetics exploded. and as the field exploded, the waysthat we studied, the relationship between genetic variation, again,this is inherited genetic variation, in prostate cancer risk evolved. so we moved from these, looking atsingle variants into these studies, looking across the entire genome. so these are calledgenome-wide association studies. and the news exploded when,suddenly, multiple places

across the genome were found to beassociated with prostate cancer. it was one of our firstbreakthroughs, but, as it turns out, the associations are tiny. they're minuscule for anyindividual snp, we call it, single nucleotide polymorphism, and most of these polymorphismsweren't even in genes. they're in regions between genes, and so we don't actuallyknow why they're related, even though they've been validated.

they've been found overand over in many studies. okay? you can see this was 2008that these types of studies, these genome-wide associationstudies, began to be published. so, we're now up to about a 100risk snps that have been found, and, again, most are not in genes and we do not know whatthey're function is. okay? most of those100 snps are associated with prostate cancer overallbut not with aggressive disease. so, now, the hunt is forsnps that'll will tell us

which men may be at riskof developing disease that will cause premature mortality. okay? so one of our goals is to findfactors that influence the prognosis of the disease and the developmentof disease with a poor prognosis. okay? another area that's stillgenetics, instead of looking for common variationacross the genome, the field has moved towardlooking for rare variants. now, one of the most excitingfindings recently was, i -- a series of variants in a genecalled hoxb13 were identified

in a small number of families tobe associated with prostate cancer. these are families that havefamilial prostate cancer. and, so, the searchcontinued into groups of men who don't have familial disease. they have sporadic disease, andwhat was found is that these snps in this gene tend tobe in men who were from northern european heritage. okay? so it seems to be particularsubpopulation who are enriched for this, for changes in this gene,

and then increased riskof prostate cancer. okay? and, so, other rarevariants are being sought. the third known-risk factor forprostate cancer, as i mentioned, is being a man of africanheritage in the u.s. okay? so, when you look atthis, these are u.s. data, so we've got non-hispanic whitemen, non-hispanic black men, asian pacific islander, americanindian, alaskan native, and, then, hispanic and latino, andyou can see, this is death. you can see that u.s. black men havemore than twice the risk of dying

of prostate cancer, andwe have no idea why. okay? this is a major differencein risk and we have no idea why. okay? in the studiesthat we've done, again, this is the healthprofessional's follow-up state, this is actually a harvard cohort. the men are all over thecountry, the cohort itself, the pi is up at harvard, andmy colleague is giovannucci, who's the first author this paper. and, so, what we did is welooked at the men in this cohort,

we looked by race, and we askedthe question: if we adjust it for all the factors that may differbetween black men and white men with respect to diet and lifestyle, could we explain away theracial disparity in this cancer? and the answer is we couldn't. in fact, the associationis the same. it's about twice the risk. okay. so black men, twicethe risk of white men. and this cohort is highsocio economic status cohort.

they're all health professionals,so there's not noise, let's say, due to differences insocio economic status. okay? so that's held constant here. so, in general, we already knowwhat causes about 50 percent of cancers in the u.s. okay? and i always like audienceparticipation. do you know what those risk factorsare for most cancers in the u.s.? >> smoking. >> so, modifiable.

left's do modifiable. smoking. >> sun. >> so, sun for skin cancer,and, so, we tend not to talk about skin cancer basal,squamous cell skin cancer because cancer registriesdon't collect it. so, yes, you're absolutely right. for skin cancer, sun exposureis the key risk factor. so, can you think of oneother important risk factor

for cancer, modifiable? so diet, and what's related to diet? >> exercise. >> exercise, and what else? >> obesity. he said obesity. yes, thank you. so, obesity. so, these factors, smoking,number one, and two,

the complex of obesity, inactivity,and poor diet account for more than 50 percent of allcancers in the u.s. okay? and for some reason, as a public,we cannot think, for some reason, we can't think about if weremoved this constellation, how much better off we'd all be. okay? so, i just wantus all to know this. this was informationthat we knew about. okay? this is work from 1981. we knew this in 1981.

okay? we continue to know this. okay? so, this is mycolleague, graham colditz. he's at wash u in saint louis, andhe had a piece not too long ago in science translationalmedicine saying, we know this. we need to do something about it. okay? of course, smokingwe've done a lot about, but i'm going to tell you ina minute, about 20 percent of americans still smoke. okay? it's not equalacross the country.

there are pockets. okay? and it's not equalacross socio economic status. okay? so things weneed to think about. so, these risk factors, smoking,and the constellation of obesity, inactivity, and poor dietare also risk factors, not just for other cancers, right,but for most chronic diseases. heart disease, stroke, diabetes. okay? so these are excellent targets for intervention andalso for prevention.

so, preventing the risk factors. okay? but are these riskfactors for prostate cancer because it'd be greatif they would be. well, oddly, no. at least not at face value. okay? so, what we've realized,when we look at these factors and prostate cancer as anoutcome, we don't see association, but once we start looking atthe fine details of the disease, for example, by whether thedisease resulted in death, then,

we start seeing associations. okay? so we have tolook very carefully. we can't just say in general. so, let me just take a momentand tell you why we now focus on lethal disease and by "lethal",we mean death from the disease, and we also mean thedevelopment of metastasis. okay. so, advanced stage disease. that disease is more likelyto reduce the life span. okay? so those are the statuseswe care about in our studies.

okay? so, i need totell you that we now over detect prostate cancerbecause of the psa test. okay? it's very clearwe over detect it. meaning, we pick up disease thatwould have never caused harm. okay? we also pick updisease that would cause harm, the problem is we can'tdifferentiate. okay? so we pick upa lot of disease. you saw that peak in prostatecancer incidence rates because of this test.

okay? so, which means, in ourstudies since the early 1990's, the majority of cases thatwe're studying are psa detected. well, it turns out, when we startdividing by disease that's detected by psa and disease that's not,the risk factors aren't the same. so we don't want tocontaminate our studies by studying total prostate cancer because we may not be findingthe right risk factors. so, again, lethal prostate cancer,we mean metastatic at diagnosis, we mean metastaticduring follow-up time,

and we mean death from the disease. and, so, in our studiesnow, we have shifted to using lethal as the outcome. okay? i know that soundshorrible to say it that way. we use lethal as the outcome,but you follow what i mean. that we specificallylook at the subset in men who have experiencedlethal prostate cancer. okay? and, then, just to showyou, this is sort of a model of the natural historyof prostate cancer,

so normal prostate epithelial,meaning that the cells that are the secretorycells of the prostate. it's unclear exactly what theintermediate disease states are that then result inprostate adenocarcinoma. we think there may besomething called focal atrophy. we know there's something calledprostatic intraepithelial neoplasia that seems to be onroute to the disease, but we're not a 100 percent sure. okay? and, then, most men withthis disease die of something else

when they eventually die. okay? but there's a subset ofmen who develop metastasis, and the typical treatment for men when they have metastatic disease issomething called androgen ablation. so, basically, knockingdown androgens, either chemically or surgically. and these men, unfortunately, a subset of them will developsomething called hormone refractory disease, and many of those men willgo on to die of prostate cancer,

and so it's these men thatwe want to be able to study. we want to find riskfactors for this disease and be able to intervene on them. okay? and, then, what about menwho already have prostate cancer? what are the rightoutcomes in those men? well, i should tell you, abouta third of men who were treated for a local disease, meaning at thetime that they're being treated, the disease is confinedto the prostate. it hasn't spread yet.

about a third of thosemen will actual recur. disease will come backafter surgical treatment. okay? and, so, how do we knowthat the disease has recurred? well, after a man has had aprostatectomy, a few months later, his psa should dropto not detectable. the prostate is the source ofpsa, so once the prostate is out and enough of time has gone bythat the half-life of psa would, you have to have that clearance, the concentration shouldbe not detectable.

okay? so, men who recur, if theyhave biochemical recurrence, it means their psalevel has gone back up. not all of those men, though, willgo on to develop metastasis and die. okay? so, we tend not to usenow recurrence based on psa as the sole outcomethat we want to study. we actually do want to study,again, the worse prognosis, which is progression tomen's metastasis and death. okay? so in our associationstudies, our epidemiology studies, that's the outcomethat we want to study.

so, now, for these, the nextslides, i'm going to only show you, or mostly focus on, lethaldisease, and, then, also, progression to death from prostatecancer in men with the disease. let's start with cigarette smoking. so, we know it's a majorrisk factor for lung cancer, head and neck cancer, kidney,pancreas, among many other cancers. is it a risk factorfor prostate cancer? and let me show you, i told you, but let me show you,americans still smoke.

so, we've got tremendousdeclines in smoking. this is 1965 immediately afterthe u.s. surgeon general's report on smoking and health. people did quit. we'd see the tremendous declines. okay? but we're stillabout at 20 percent. i can tell you in mystate, in maryland -- how many of you are marylanders? okay. just a few of you.

in my state, in our state, among those who do not havea high school education, so people who have less than a highschool education, about 30 percent of them smoke, even thoughin the state of maryland as a whole, only 15 percent smoke. okay? so, we do have socioeconomic disparities in smoking. okay? so it's something to knowand something we need to act on. okay. so, the u.s. surgeon general'sreport, there's been an update. the 50th anniversary update waspublished january last year,

and here's what it saysabout prostate cancer. the evidence is suggestive of nocausal relationship between smoking and risk of incidentprostate cancer. okay? there's no association. and, so, for a very long time, that's what was inthe public domain. smoking is not related to prostatecancer, so let's not worry about it. but here's what's now been found. okay? so, once the evidencehas been put together,

we now see that the evidence issuggestive of a higher risk of death from prostate cancer insmokers than in nonsmokers. so the worse outcome, smokingis a risk factor for it. okay? and, then, in menwho have prostate cancer, smoking is associated with disease that has a worse phenotypefor prognosis. so at the tissue level, the gradeis poorer, and it's a risk factor for progression of the diseasein men with the disease. okay? so we now believe,based on evidence,

that smoking is a risk factor forthe most aggressive prostate cancer. okay? so, this is apoint for prevention. so, let me show yousome research, again, that we did in healthprofessionals follow-up study. there's no relationship. okay? so, this is the relative risk. says the risk comparing risk ofthe development of prostate cancer in men who smoke compared tothe risk in men who don't smoke. okay? so. here's the neversmokers, here are the men

who quit a long time ago, and hereare the current or recent quitters. no relationships. but for fatal prostatecancer, here's what we see. the blue bar, again, is men who arecurrent smokers or recent quitters, meaning, they quitwithin the last ten years. these men have a 40 percent higherrisk of developing prostate cancer that will lead to their death. okay? and here's another study. this is an americancancer society cohort.

cancer prevention study, too. it's the same findings. same rough strength of association. and, then, taking a look atmen who have prostate cancer, looking at risk of poor outcome. okay? so, here, the never smokers, these are men who havea prostatectomy. former smokers, meaning, they quita long time ago, no increased risk. but men who were current smokers,meaning, they continue to smoke

after their diagnosis and treatmentfor prostate cancer, they have more than twice the risk of recurrence. okay? and if we look at this asaccumulative incidence, okay, over across about seven years, 34percent of the smokers recurred, and 12, 14 percent of the neversmokers and the former smokers, those who quit a longtime ago, recurred. big difference in thelikelihood of recurrence. okay? so, public health action. so, i'm here to tell you aboutresearch, but i'm also here

to tell you what we needto do to make a difference. okay? so, what's thepublic health action? so, we need to continue to promotesmoking cessation and prevention of starting smoking, right,in the general population. so, we know we need to do that,but we need to continue to do it and we have to make suremen know this too, right? and this is including to preventfatal prostate cancer among a very long list of otherpoor health outcomes. and, then, from men who havethis disease and actually

for all cancer patients, we have to make sure we promotesmoking cessation. okay? traditionally,we didn't do this. okay? once someone wasdiagnosed with cancer, we just -- that was no longeron the to-do list. okay? but there are now guidelines that were just released earlier thisyear that say we need to promote, in a very active way at the timeof diagnosis, smoking cessation in cancer patients, and, again,

this should include menwith prostate cancer. okay? let's move on tothe epidemic of obesity. so, we know that this is a majorconcern in the u.s. and globally. we all know -- everybodyknow your body mass index? okay. more or less where you fall? okay. so, overweight is 25 pluskilograms per meter square, obesity is 30 plus kilograms permeter square, and bmi is calculated as your weight in kilograms dividedby the square of height in meters. okay? so, obesity is a risk factor

for many common cancersincluding colon cancer. okay? and we do know that we'vehad this massive rise in obesity in the u.s. startingin the mid 1970's. i mean, this is a horriblerise, right? we're up to about 66percent of american, adults, being overweight or obese. same thing's happening globally. not to the same extent,but it's on the rise. okay? so, are these typesof factors, so, obesity,

physical inactivity, riskfactors, for prostate cancer? well, for years, we thoughtthere was no relationship. okay? this is a consensusdocument, 2007. this is specific to prostate cancer, and you can see here a limitedno conclusion body fatness. okay? so we thought therewas nothing going on. well, now, in 2014, that sameorganization updated the evidence base, meaning, theylooked at the literature, they found all the new studies, theyput it together in a systematic way,

and, now, what they find isthat there's strong evidence that being overweight orobese increased the risk of advanced prostate cancer. so, the way we usedto our studies, again, was to look at totalprostate cancer, but that psa problemmessed up our studies. okay? so, once we startedsorting out the men with most the aggressive disease,finding them within our cohorts, and, then, studyingthese common risk factors

for many chronic diseases,now we see associations. the evidence base is there. okay? and, then, just toshow you example studies. this was a landmark study. it was back in 2003. this is the cancerprevention study ii out in the american cancersociety, and you can see as body mass indexincreases, the risk of death from prostate cancer increases.

okay. here's another cohort. this is by my colleagues atthe national cancer institute. this is the nih-arpdiet and health study. the darker blue bars arefor fatal prostate cancer, the lighter blue bars are for theincidence of the disease, and, again, as you may see,as bmi goes up, the risk of fatal prostatecancer goes up. and, then, something tonote as obesity goes up, the risk of incidencedisease goes down.

and, so, when all this wascombined together in the past, it made it look likethere was no association. so, it wasn't until wepartitioned into advanced or fatal and incident disease, whichwas primarily psa based, that psa detected basedisease, then, we were able to seethese associations. okay? and what about menwho have prostate cancer? this is, again, work by my group. cory joshu was the first author.

so, these are men whounderwent treatment for prostate cancer,they had early disease. if they maintained their weight fromthe years before their diagnosis to after their diagnosis,that was the reference group, and when we said maintain,it was a five-pound range. men who gained weight,some more than five pounds, the typical weight gainwas about 11 pounds from five years beforeto one year after. we see that these men havetwice the risk of recurring.

okay? men who lost weight,there weren't that many. [ laughs ] but there's a hint ofan inverse association. how about physical inactivity? well, it doesn't appear to beassociated with prostate cancer, but it does appear to beassociated with outcomes in men who have the disease. so, this is from mycolleagues, again, it's the health professionalsfollow-up study.

stacy kenfield is the first author. and, so, this is levelof physical activities, so you can see number of med hours. this is just a measureof both frequency and intensity of the activity. so, increasing med hours isincreasing activity and intensity, and what we can see is, as thenumber of med hours increases, in general, the risk ofpoor outcome decreases. so, how we're going to intervene?

again, public health action. it's perhaps even harder tocombat obesity and inactivity, and i didn't reallytalk about poor diet. that's a very complex literature. it's hard to intervene. and you know when we take publichealth action and we try to do it through policy and laws,it's a tough sell, right? it's been a tough sell,but we've done it. we've taxed smoking, andwe've had tremendous declines

in smoking prevalence. you saw what happened in new yorkcity with the attempt at taxation of sugar-sweetened beverages, andit failed because of the pushback. okay? so, maybe we need to go back into the primary caresetting, right? so, instead of doing it asbig population messages, maybe we need to do it inthe primary care setting and have health systems tryto deal with this problem. okay? and, maybe, take advantage

of the affordable care actprovisions for prevention. okay? and, so, the affordable careact mandates that any service, preventive service, that the u.s.preventive service is task force says has a recommendationof b or a needs to be covered without copayment. okay? so no cost to the person. and, so, there is arecommendation for adults that all adults shouldbe screened for obesity, and if individuals are obese,they should be counseled.

and, so, what do we need? we need research desperatelyon weight loss and inactivity for cancer patients. okay? so, we need to know whetherchanging weight after the diagnosis, increasing activityafter the diagnosis, would produce betteroutcomes in those patients. okay? and we need to know whento start these interventions. is it at the time of diagnosis,or is it sometime later? okay. after some survivorship time.

okay? so, now let mesummarize the risk factors for lethal prostate cancer. so, we looked at non-modifiablefactors, older age, being african american orhaving african ancestry, and family history. for genetics, we now know there areabout a 100 risk alleles associated with prostate cancer,although i should point out, this is not for lethal disease,it's for total prostate cancer, and there's a search forsnps for lethal disease.

and, then, modifiable factors,most that have been found, these few that i showed you,are primarily for lethal or fatal disease, and, again,smoking, obesity, and inactivity. and i highlight obesity because it'sgoing to be so hard to intervene on. we have to figure out what to do. so, now let me give you examplesof the research that's being done on risk factors andalso prognostic tools, and i'm going to show youwork from my own group. so, what do i study?

so, i'm going to give you twoexamples for prostate cancer. one is a cause and oneis a prognostic marker, inflammation and telomere length. and, then, i'm going to show youtelomere length as a risk factor for prostate cancer,looking at obesity as it influences telomere length. okay? and, then, i alsowant to point out something that tomoko eluded to, which istranslational research is critical, and that's what we're trying to do.

translational research is best done by bringing multipledisciplines together. and, so, here's part of our team. so, we're basic, clinical, and population scientists workingtogether; so, angelo de marzo, a pathologist; alan meeker, a cancerbiologist, and, more specifically, a telomere biologist; me, anepidemiologist; my colleague, vasan yegnasubramanianwho's a cancer biologist; and will nelson who'sa medical oncologist

and also a cancer biologist, and we all come togetherto do our research. okay. so, let me startwith inflammation. so, what we hypothesize is thatchronic or recurrent inflammation in the prostate itself,so not systemic, but actually in theprostate, increases the risk of the developmentof prostate cancer. and we made this hypothesisbased on other cancers. we know that inflammation canserve as both an initiator

and a promoter of cancer. okay? so, think about h.pylori and stomach cancer. okay. so, it's a bacteriumthat increases inflammation of the stomach mucosum, it'sknown to be a risk factor and a cause of stomach cancer. okay? if this hypothesis is truefor prostate cancer, we would expect that inflammation would beincredibly common in the prostate because prostate cancer's so common. all right?

so, do we know whether or not inflammation iscommon in the prostate? well, we do know thatbased on studies where tissue was removedfor indication. so, tissue removed when men havebenign prostatic hyperplasia, tissue removed at the time ofbiopsy if a man has an elevated psa, and tissue removed at the timeof surgery for prostate cancer. but what we don't know isin men without indication, whether inflammation iscommon in the prostate

because all these states tendto be related to psa level, and it turns out thatinflammation increases psa. okay? so, it's sort of circular. so, what we needed was aplace where we could look where men don't have an indicationfor biopsy, and we'd be able to get an unbiased estimate ofinflammation in the prostate. well, turns out, there was a bigtrial conducted called the prostate cancer prevention trial. it tested the drug finasteride,which inhibits the enzyme

that catalyzes the conversionof the male hormone testosterone to another male hormone,dihydrotestosterone. and, in that trial, theybiopsied the men at the end. irrespective of whetherthey had an elevated psa. irrespective if they had anabnormal digital rectal exam. and, so, we were able to collaboratewith the investigators on this trial to use that biopsy tissue. the majority of the men,at tend of the trial, did not have indication for biopsy.

they were just biopsyas part of the trial. okay? so here are the resultsof that study that we did. okay? first thing to note is that78 percent of the controls, meaning, men who were not diagnosed withprostate cancer in the trial, had at least one biopsycore with inflammation. about six to ten we are taken,we sampled three per man, so men who had at least one ofthose three with inflammation in it, that was our measure. okay? and 78 percent of thecontrols had inflammation

and at least one core. okay? so, it turns outinflammation is incredibly common in the prostate. it probably needs to be there. you probably need inflammatorycells to clear any type of agent like e. coli that mightenter the prostate. prostates actually openedto the outside environment and will give you process. it's open to the outsideenvironment,

so you need immunecells present to be able to clear those types of agents. so it makes sense. okay? and, then, somethingelse to note is the prevalence of inflammation in the cases. so total prostate cancer, about86 percent had one or more core with inflammation and highgrade disease, which, again, is a prognostic markerby 88 percent. okay? and, then, if we looked atthe association between having

at least one biopsy core withinflammation and the risk of prostate cancer, we foundabout an 80 percent increased risk from men who at least one corepositive and even higher risk, so it's about twice the risk formen who had at least one core, and the outcome herewas high grade disease. okay? so, again, thesewere end of study biopsies. we actually measured inflammationin those biopsies that were used to rule in or out thediagnosis of prostate cancer. okay? so, what that means is ourstudy was not temporally correct,

so we didn't have biopsiesfrom men who, biopsies from men and then follow them forward forthe development of prostate cancer because not a lot of men wouldvolunteer for a study like that where you, like, 100thousand men say we're going to biopsy your prostate andthen follow forward in time. so, the study we didwas not prospective. it's not temporally correct. but turns out through collaboration, we were able to link theprostate cancer prevention trial

with another trial conductedby the same cooperative group. the next trial was called select. some of the men in the pcptparticipated in select, which means their end of studybiopsy became the baseline biopsy for that next trial, therefore, we were able to conduct aprospective cohort study of inflammation in theetiology of prostate cancer. okay? so, here's the design. so, again, the end of study biopsy

from pcpt became thebaseline biopsy in select. we assessed inflammation in thatbiopsy and then determined who ended up with prostate cancer later. okay? and this work is supportedby the department of defense. and here's what we saw. we saw men who at least,had at least one biopsy core of inflammation in the controlswith 72 percent, so very similar to what we had seen before. and you can see that in, withvarious outcomes, low grade disease,

men with low psa but with prostatecancer, men with low grade disease, and suppose to say, and low psa,all these men had a higher risk if they had at least onecore with inflammation. now, what's missing fromhere is high grade disease, and it's because very few men inthe study had high grade disease. they had to make it through thepcpt and then enter select, and, so, once they got into this trial,we had served depleted the men of more aggressive disease. okay? never the less, we dosee evidence of an association.

and, so, this is the number of coreswith inflammation, so some cores and then all cores, wesee a dose response. when we see dose response, wetend to believe more strongly that there's an association verseswhen we don't see dose response. okay. so what's theclinical import of this work? well, if there's truly a causalassociation between inflammation and prostate cancer,which we're hoping it is, we need to do moreresearch, we may be able to identify what itis about inflammation.

which cells, which immunecells are increasing the risk? okay? so, additionalwork we could do. what's not going to happen isinflammation will never be able to be used as a tool toscreen for this disease, and the reason is it wouldhave a poor predictive value. so many men have inflammationof prostate, so i have no predictivecapacity whatsoever. okay? but from understandingthe etiology of the disease, we think that these findingshave potential import.

okay? so, move on to thenext topic, telomere length. so, telomeres are theends of the chromosomes. okay? they help maintain the patencyof the structure of the chromosome. they're known telomere shortenwith each round of replication, and it's well recognized thatcancers have shorter telomeres than normal tissuefrom the same organ. so, we wanted to addressthe question of whether there'sprognostic information in knowing telomerelength in prostate cancer.

okay? again, this is amultidisciplinary study done with our colleagues incancer biology, and, so, what we did is we had tissuefrom men with prostate cancer, we had the cancer, and we had thecells near by called stromal cell. the stromal cells, the supportivecells, so fiberglass, for example, smooth muscle cells,all from the prostate. we determined telomere lengthusing a technique called fish. we were able to imaginethe telomeres, we were able to determinethe length of the telomeres,

and what we found was itwasn't how short the telomeres in the cancers were because theywere short across the board, it was how variable telomere lengthwas from cancer cell to cancer cell. the greater the variability,the higher the risk of death from prostate cancer in those men. okay? when we looked at thecells near by, the stromal cells, it turned out how short thetelomeres were was associated with risk of deathfrom prostate cancer. and when we put those two parameterstogether, those two measurements,

what we found is that menwho had the combinations, purple bar is the combinationof more variable telomere length from cell to cell in the cancercombined with shorter telomeres in the stromal cells was associatedwith 14 times the risk of dying of prostate cancer in men withprostate cancer, even after taking into account the otherknown prognostic markers. okay? this is a verystrong association. we tend not to see associationslike this in epidemiology. okay? the men in the middlebars had other combinations.

so, they either had variablein the cancer and not short, or they had not variable and short. okay? and, so, you can see that theyalso have increased risk but not on the order of the menwith the poor combination. okay? all right? and, so, let's talk alittle bit about these men. so, the men who had this betterprofile, so the men who are here, they had less variability in theirtelomere length from cancer cell to cancer cell, and they had longertelomeres in their stromal cells.

only one man died ofhis prostate cancer. okay? we had expected about sixmen to have died in this cohort. the time for this one man,from diagnosis to death, was a very long time, 16.5years, which is longer than the usual life expectancy fora man who had gleason 9 disease. okay? this is the very,usually considered to be a very aggressive disease, buthe managed to live a very long time. and, again, there was only one manwho died of his disease in those who had what we considered tobe the good combination, so.

less variable and longer. among the men who had more variabletelomere length in the cancer and shorter telomeresin the stromal, 20 men died, about 10 were expected. okay? so that's a big,big difference. and the median time from diagnosis to prostate cancer deathwas about eight years. so, you can see this oneman live twice as long as would have been expected.

okay? so, we don'tthink this was an error. okay? so we -- if our biomarkerworks very well, you'd expect no men in that group to havedied of prostate cancer. well, this one man who did diedof it, his time to death was twice that in the group that wethink has the worse prognosis. so, now let's tie togetherthe findings that we just, i just showed you fortelomere length with some of the modifiable factors. and, so, this is work done incollaboration with cory joshu.

she's an assistantprofessor in our department, and this work was supported,her specific work, was supported by theprostate cancer foundation. okay? so, i've told you thather work showed that men who gained weight had ahigher risk of recurring after treatment for prostate cancer. so, what she wanted todo was to look at obesity in relation to telomere length. okay? we didn't want to look intocancer because the cancer we know

because of its higher proliferationrate and just greater damage that goes on in cancer cells. it didn't make senseto look at obesity and telomere length in the cancers. instead, what she did, shelooked in stromal cells. so, recall that we sawthat short telomeres in the stromal wererelated to an increased risk of prostate cancer death. so, when she looked in thestromal, men who were overweight,

were obese, had shorter telomeres. okay? so, men who were overweighthad shorter telomeres, overweight or obese, than men whowere of normal weight. okay? this is in the stromal cells. okay. when she lookedat physical activity, men who had low physical activitylevels had shorter telomeres. okay. this had not been seen before. no one has been able to look in thetissue, in the target organ itself. okay? so, now we're seeingevidence of potential impact

of these modifiable risk factorsfor prostate cancer, lethal disease, and poor outcome, and their effectspotentially on telomere length. okay? when she combined obesity andinactivity, she found that the men who were obese and had lowphysical activity levels had about 21 percent shorter telomeresin their stromal cells than men who were normal weightand were active. okay? and she alsolooked at smoking. so, remember her finding. men who continue to smoke havea higher risk of recurrence.

she looked at smokingand telomere length. there was no relationship withhow short the telomeres were, but in the stromal andactually in the cancer as well, she noted relationships betweensmoking and how variable, and just to remind you,more variability is worse. okay? so, variables worse. so, these are the men who are neversmokers, and, so, they have -- here's never smokers again. these are people who quit recentlyand these are the current smokers.

so variabilities worse. this is more variable. so, that's some few examples of contemporary researchon prostate cancers. the last thing i want to dois discuss the controversies in the early detectionof prostate cancer, and by early detection,we mean screening. and, so, i want to remindyou what screening is. so, screening is the use of a test

in individuals whoare a symptomatic. okay? and that test is used todetermine whether a person is likely or not likely to havethe disease or state. it doesn't mean they have it. okay? so, it's that they're likelyto have, more likely to have it. okay? and, then, usually, there'sdiagnostic workup that needs to happen to actually determinewhether or not disease is present. okay? so that's what screening is. and, so, we know that psa-basedprostate cancer screening has had a

profound effect on prostate cancerin the u.s. and in other places where this type of screeninghas been used more widely. okay? and, so, recall this figurefrom the american cancer society. we saw this huge bump in theincidence, the number of new cases, because of psa-basedprostate cancer screening. and i also showed you the changein survival, five-year survival, and this change is because ofpsa-based prostate cancer screening. but i should also tellyou that a lot of this 100 percentsurvivorship consists of men,

if their cancer hadn'tbeen detected, it would have neverkilled him, killed them. okay? because we've overdetected the disease. okay. and the pre-psa era, thesewere men who were symptomatic at diagnosis or had clinicallypalpable disease of their prostate. okay? so, we have to considerthe benefits verses the harms. we, as a public health disciplineand also the medical community, convinced everybody that screening for cancer is necessarilygood, right?

we wanted people to be screened. it was the right thing to do. we made people feel guilty about it, but that was a population-levelrecommendation. we now realize that we'veoverdone it to some extent, and so what we're doing now is we'repulling back and we're thinking about who should betargeted for screening. who is at highest risk and that riskmay change throughout the lifetime. okay? we need to think aboutthis because there's a tradeoff

between the benefits of screeningand the harms of screening. a lot of people say, well, therearen't harms for screening. well, but they are. okay? so, we pick up disease thatmay not be cancer at all, but, yet, are medically worked up,surgically worked up. we pick up cancersthat are indolent, would have never causedpremature mortality, so you're better off notknowing if you have it. and, then, there'sadverse effects related

to the tests themselves includingthe diagnostic workup, and, certainly, the adverse effectsof associative treatments. okay? there's alsopsychological harms, right? there's the worry. if someone turns outto be a false positive, they're told that theirscreening test is positive, but on diagnostic workup they'renegative, imagine the worry. okay? a person will always think, iprobably do have the disease, right? it's a problem.

okay? plus, we have the burdenon our healthcare system, right? there's cost. there's actual cost. okay? so we have tobe cognizant of that. and, then, we have this problemwhere our guidelines keep evolving, but that's important, right? we're updating our evidence base, but the public isn'table to respond to that. okay? and, so, there's confusion.

doctors are confused. they don't know what they'resupposed to do, who are you supposed to screen, when you screen. it's very complicated. and, again, these uncertainties arecoming up because of good reasons. we have new knowledge, newevidence, and we have new tools. but as these come online, therecommendations are not changing across all of our societies,right, our professional societies. they're not changing atthe same time concurrently

in the u.s. preventiveservices task force for the american cancer society,so everyone's getting confused. okay? before the messagewas you should be screened, now the message is changing. okay? so for prostatecancer, the first thing that happened just a few yearsago, was there was a recommendation to stop screening after age 70. okay? and the idea was that the risk of other reasons why peoplemight die, why men might die,

actually become greater in thatage range than for prostate cancer, especially for men who've madeit to 70, they've been screened, the prior probability oflater having prostate cancer, if you've been negative on thosescreens multiple times, is low. and, so, the idea wasjust stop screening. but, then, the most recent change,again, 2012, was to not screen at all using psa-based approach. okay? so, here's the u.s. preventiveservices task force website. here's the recommendation.

okay? they gave it a gradeof d. and, basically, what they did is theylooked at the evidence base. there were two largetrials that were published. one was done in europe. one was done in the u.s. the studyin europe was done in a background of populations that did not havea history of routine screening. the use of the test did result inreduced prostate cancer mortality, but at the expenseof over detection. they're very clear on their paper.

they say that over 14 hundredmen would need to be screened and 48 additional cases ofprostate cancer would need to be treated to prevent one death. okay? so, a lot of menwould need to be medicalized to be able to prevent one death. okay? in the u.s. study,it was done on a background of psa-based prostate cancerscreening, and the prevalence of that uptake increasedduring the trial. so, by the end of the trial,they did not find a relationship

between psa and reductionin mortality. okay? so, we alreadyhave a background. so, these studies are notinconsistent with each other, even though when it got into thenews, they were called inconsistent. it's the nature of what populationwas included in the two studies. okay. so, based onthis evidence base, the u.s. preventive servicestask force found good evidence that psa screening can detectearly stage prostate cancer, so the test does pick up thedisease, but there's mixed

and inconclusive evidence that earlydetection improves health outcomes. okay? and they go on to talk aboutthe harms associated with screening, so at the end, they concludedthat the evidence is insufficient to determine whether thebenefits outweigh the harms for a screened population. of course, men got very upset. okay? so, we had told everybodygo be screened, and then, suddenly, we're going, nope! don't be screened.

okay? and, so, we sawletters, we heard -- i had a man stop me ina hotel hallway once. i was there to give a talk and wasin a hotel, and the man stopped me, he said, "i'm a survivor. you can't let this happen." right? and, so, themessage that a lot of men who were survivors was that,but it saved me, right? i was screened. i had prostate cancer.

i was treated. i'm alive. it saved me. okay. well, i'm sympathetic to this, but we lack the counterfactualstate. we don't know whatwould have happened to this man had henot been screened. okay? undoubtedly,some men are saved. that decline in prostate cancermortality that we showed you,

no doubt is in part due toprostate cancer screening. the problem is we pick up diseasewe shouldn't be picking up. we do not have a toolthat, at this time, will specifically identifythe cases that are likely to result in premature mortality. okay? we do have differencesin guidelines. the american cancer societyguidelines say in men in average risk, screeningshould start at age 50 plus, dre couple with psa, and it shouldbe based on informed decision making

between the individual andthe healthcare provider. man's own values needto come into play. his other health characteristicsneed to come into play. okay? and that decision,is that a joint decision, between that patientand his provider. okay? the aua, based on the europeantrial, has a series of guidelines. they recommend that screening couldoccur for men 55 to 69 years old, and, again, this would be in ashared decision making setting.

okay? so, i do want to remind youthat under the affordable care act, there are certain preventiveservices that are covered withoutcost sharing. you'll notice that psa-basedprostate cancer screening is not here, but others are. so, colorectal cancer screening,mammography, and there's age ranges and timing that are here. so, what do we need urgently,given the burden of this disease, given that men do dieof this disease?

we need to figure out strategies to improve the benefits verses theharms in the screening that we do. and, so, we need newalgorithms for use of our current technology,so who should we screen? which men? when? how often? and so on. so, that's an activearea of research. and we need new screeningtests all together,

whether they're blood based or maybeeven better based on imagining. okay? so, all of thisis under active study. okay? so, i hope thati've met the objectives of talking with you today. we described the burdenof this disease. we talked about known andsuspected risk factors for the disease andincluding outcomes. i gave you examplesof ongoing research, and we discussed the controversies

and i just want to point out,major knowledge gaps where we need to put lots of attention. why do african american men havea higher risk of prostate cancer, especially disease that kills? are there additional modifiablefactors for lethal disease in poor outcome in men with adisease that we can intervene on? and what is the optimal prostatecancer screening approach? okay? going forward, make surewe emphasize as risk factors, not just for prostate cancer, butfor cancer in general, smoking,

obesity, inactivity, poordiet, doing so is going to desilo our diseases, right? so, we tend to say, oh,cardiovascular disease is over there, diabetesis here, cancer's here. well, these are allthe same risk factors. we have to desilo and get the mostbang for our public health dollars. it avoids duplication of efforts andexpenses, creates a seamless model for public health promotion. okay? and we want todeemphasize controversial factors

and quick fixes, like magic pills. okay? we have to be active. we have to maintainour weight, right? we've got to quit smokingor not start. those are things we should workon, and i'd like to acknowledge all of my collaborators from bothjohns hopkins, from harvard. i've had a ton of traineesover the years who have contributed to this work. also want to thankthe pct investigators

who have provided a greatsetting for us to do or our work. and, then, here's our school. we're celebrating our 100anniversary this year. here are my close colleagues,dr. marzo, dr. meeker, dr. joshu, and here's some of the teammembers in my office one day, and we just got a shot of us all. and, of course, the fundersfor the work, nci, dod, the marilyn cigarette restitutionfund at johns hopkins as well as the prostate cancerfoundation for dr. joshu.

so, i thank you all for comingtoday, and i'm open to questions. so, the question was whether hormonereplacement therapy effects the aggressiveness, either the presence of prostate cancer orits aggressiveness. and, so, you know, there issome research that's being done. there are panels that are beingconvened to try to understand, do we have enough informationyet to say one way or another? so, the answer's i don'thave an answer for you yet. yup. so, the question is whetherthere's a coordination of effort

for men who have bph, and, yeah. it depends on what order you want tosay the words, and, so, basically, in large prostate thatproduces symptoms in prostate cancer,and, so, you're right. these men tend to overlap. the region where bphoccurs in the prostate and where the cancer occursare actually different regions. and, so, some people try tostudy whether ones a risk factor for the another.

at this time, it doesn't appear thatone is a risk factor for the other, instead, they seem tohave common risk factors. so, factors that influenceprostate cancer risk also tend to influence the development of bhp. yeah. okay. so the question is: one,asian men have a lower risk of prostate cancer. given that there are differences inwhat men eat and what they do based on where they liveglobally, has anyone tried

to evaluate whether an asian dietwould prevent prostate cancer in western men? okay. and, so, one, i shouldtell you when asian men, these are historical studies, whenasian men move from asia to the us, which is actually hawaii inthese studies, their own risk of prostate cancer goes up. their sons have a risk ofprostate cancer that begins to approach the risk ofwhite men living in hawaii. okay? so, that says it's probablynot genetics that are counting

for this profound racialvariation in prostate cancer risk. it says it's probablydiet, lifestyle, things people have control over. we haven't been able towork out what is different about the asian diet, in asia,and risk of prostate cancer. so, we have not done majortrials of asian style, diet, and other factors, but we've tried to isolate individualfactors in cohort studies. so, prospective studies, nottrials, but observational studies.

we've tried to look at,for example, soy intake. the problem is that people in theu.s., in places like the u.s., don't eat a lot of soyintentionally, right? we have soy that's mixed into ourfoods that we don't even know about, but what men can report thatthey're eating is low levels of soy. so we cannot pick up association. so, it's a great question. we do not know what explainsthe racial differences, the profound racial differences,

including the much lowerrate in asian man in asia. so the question is: beyondlooking at individual factors, have we looked at contextualfactors. so, where people live,their social structures, not just whether healthcare'savailable, but whether people are ableto actually access healthcare, and how this may affect differencesin risk of prostate cancer and, in particular, prostatecancer death. and, so, there are studies thattry to look at contextual factors.

so, these are factors that areabove individual decision making about whether you're goingto smoke and, you know, physical inactivity, for example. and, so, if you don't even havesidewalks in your neighborhood, how are you going to walk? so, it's work like that. and, so, my group has notdone that style of research, but there are researchers whoare working in those areas. thus far, we have not beenable to find those types

of contextual factors, asidefrom looking at disparities in prostate cancer outcomesin equal access care systems like within the va. so, within the va, it doesn'tmatter who you are, there's a system in place to make sure peoplehave equal access to care, and so in those settings,some studies have found that the disparities are reduced,others still see disparities that the disease stilllooks more aggressive, for example, in africanamerican men.

the question is: whois it that we are able to study based on who participates? okay. so, historically, studieswere primarily white men, these are prostate cancer studies,maybe a little more well to do, maybe less opposed to, sort of, the worries about hospitalsystems and so on. we have made, in the field,not just for prostate cancer, but just in health research ingeneral, a concerted effort to reach out to communities that were notrepresented in earlier studies,

and there are now anumber of cohort study. i'm an observational researcher. a number of cohort studiesthat focus on communities that were traditionallyharder to reach, so my colleague bill blot has acohort study called the southern community cohort study. he went into clinics thatare federally subsidized, recruited middle asianolder individuals to participate in those studies.

those studies are primarilyafrican americans, and he has done a reallytremendous job of people, keeping people engaged. but we're still missing outon the poorest of poor, right? we're missing out onpeople who are not connected to health systems, right? and, so, your question aboutwhether it could make a difference in findings, it could, right? it could because it's not just asingle risk factor that we study,

it's a risk factor on abackground of other factors. okay? so, we do needto keep this in mind. we need to make surewe reach everybody. we have to make sure we, notjust recruit people into studies, but maintain equal follow up. it can be quite difficult. there are people whose research,career, is around identifying, recruiting, and retainingpeople in studies. i think it's a very important point.

the nih, when we receive moneyfor our research studies, we report on the diversityof those we've recruited. we've tried to make sure wehave sufficient representation that we can look withinsubgroups of individuals defined by race ethnicity, defined by socioeconomic status where possible. okay. so, it's actuallytwo parts to your question. and, so, he wanted to know what'sthe research on sedentary behavior in risk of prostate cancer,and, then, more specifically, sedentary behavior thatputs pressure, actually,

it may not even be sedentary,it could be active behavior, that puts direct pressureon the prostate. so, with respect to sedentarybehavior, that's a newer area. so, it's more newly recognized thatthere's a separate contribution to risk of cancers in other diseases from being sedentaryverses just not being active because there are somepeople who are not, they're not exercising, right? there's no leisuretime physical activity,

but they're also notseated all day long. and, then, there are people like me. i'm seated all day long, and,then, i go to the gym, right? so, i get my exercise in, buti'm seated most of the day. and, so, again, activity levelseems to have a different and independent contribution fromthe sedentary part of one's day. so, for prostate cancer,that's a new area. so we don't have abody of evidence yet. the question you askedabout behavior activity

and sedentary behaviors that putdistrict pressure on the prostate, it's an interesting question. we certainly know that if a man hasjust ridden a bicycle and then goes and has a psa test, psa will havebeen, i'm going to use the word "expressed", it leaks intocirculation, so there's advice not to do that right before psa test. but there's not, they're not, again, not a body of studiesaddressing that question. it's an interesting one.

yes? >> you can have someprivate conversation. >> great. so, thanks to all of you. >> please join me inthe thanks [inaudible]. >> thank you. >> this has been a presentationof the library of congress. visit us at loc.gov.