Cancer Of The Breast

>> from the library ofcongress in washington, d.c. >> sandra charles: okay. good morning, everyone. and welcome to this jointhealth forum with science and technology division ofthe office of health services. i'm very pleased with thiscollaboration with tomoko. and this is the second oftwo that we've done recently. and thank you all for coming. today in honor not only butparticularly because this is october

and breast cancer awareness month,we're having an esteemed speaker who will be introduced bytomoko -- dr. tomoko steen. and we're waiting with baited breath to hear all the newdevelopments from you, dr. clark. but it is -- it is a subject that wecertainly address on an annual basis because it is still the numbertwo cancer killer in women -- and probably the number oneskin cancer -- that exists. so i just wanted to welcome you andto thank you for coming and hope that you will get as muchout of it as we hope you will

and that we expect you will. and without further ado, i'mgoing to ask tomoko to come to the podium to introducedr. clark. thank you very much. >> tomoko steen: it'smy great pleasure to introduce today'sspeaker, dr. clark. and he's a senior colleague of mine. today's talk is actually part ofthe translational medicine series. we started with two nobel laureates.

jim watson and carol greider came. in fact, professor clark was oneof the speaker was supposed to be, but the schedule didn't work out. and i'm so glad to be able toinvite him back for this topic. professor clark is dean for research at the georgetown universitymedical school and also professor of oncology and core directorof the breast cancer program at lombardi comprehensivecancer center. and he has done thedsc and phd at belfast,

actually, university of queens. and also, he has done thepost-doctorate research at national cancer institute. and i have bio in the back. you can pick it up. and we have books onthe -- basic books -- on breast cancer back there. so if you have time,please take a look. and before i go intothe further ado,

i should just introduceprofessor clark. thank you so much. [ applause ] >> robert clarke: well, thank you. this is a great pleasure to be here. this is maybe going to be a littledifferent today because i'm going to tell you about some ofthe research that's going on and how we've begunto think differently about understandingseveral different cancers

but with a specificfocus on breast cancer. okay. hands up if you know anyone-- yourself, anyone in your family or extended family -- who's everhad a diagnosis of breast cancer. look around. sadly, one in eight women -- i'll show you the statisticsin a minute -- are likely to experience a diagnosisof breast cancer in their lifetime. we made significant stridesin managing this disease, and we do cure breast cancers; wejust don't cure all breast cancers.

we will get there. we probably won't getthere as quickly as everyone would like,but we will get there. today i'm going to give you an -- some insight into what thestatistics are for breast cancer, who it affects, some of the drugs. i'm going to focus mostly on estrogen-receptor-positivebreast cancer, which is about 70 percentof all breast cancers.

and tell you how we've begunto take a different approach to understanding the biology of thatdisease and how that might lead us to think very differently abouthow we might treat that disease and ideally, eventually prevent it. now, i've got a little bitof science in my slides. and probably none of you arebreast cancer researchers. so please, if you don'tunderstand anything, just stop me. i'm happy to take time and try and explains thingsto you as we go along.

and if you see somethingthat you don't understand, just make sure that we don't panic. really, this is here to haveas much conversation with you as i am to tell you what we do. i'd like you to learn somethingabout breast cancer from today and go away with itand think about it. because we all have aresponsibility to do something in whatever way we can toeradicate this terrible disease. so don't panic if you see data, orcrazy words, or symbols that sound

like they're a foreign language. they probably are a foreignlanguage; they're science language. but most of those youdon't actually need to follow; just follow the idea. don't worry about thedetails because it's the sort of big picture i'd like toget across to you today. so let's start by tellingyou something about cancer, just in case you don't know much. you might think that it'sa disease of modern man

in some way, shape, or form. but actually, it is one ofthose unfortunate side effects of just being alive. we've known cancer has beenaround for a very long time. and the dinosaurs got cancer. you might wonder howdo we know that? well, some cancers come up in bone. and so when fossils are createdand the bones are left behind, you can actually see evidencethat there was a cancer

in the bone, in that fossil. and it turns out this is an exampleof one, apparently, who knew -- hadrosaurus were particularlycancer-prone. dinosaurs -- plants canget types of cancer. so it's really a verycommon disease. and i'm going to explainto you what cancer is. and then we're going to go andlook at what breast cancer is. and i'm going to tell yousomething about some of the drugs that we have and how they work.

and then we'll talk about howbreast cancer cells get [inaudible]. and that's the key. we need to understand how tostop cancer cells escaping from when we're trying to kill them. so most cancers --all cancers, really -- start with the dna mutation, achange in the sequence of dna that encodes for eventuallythe proteins that allow our cellsto perform functions. and sometimes when those changesoccur, it screws up the ability

of the cells to function properly, and in some cases itleads to cancer. so it starts with a mutation. that doesn't mean that we knowwhat that mutational event is in all cancers -- we clearly don't. in some cases breastcancer would be an example. we know that some are caused bya mutation in a particular gene, like the brca1 geneor the brca2 gene, which you've probably heard of.

those are mutations thatlead to more mutations that build up in the cell. and eventually we get cancer. and -- and to all intents andpurposes, cancer is really a disease where there's loss ofthe normal controls on proliferation and growth. it's where cells start tomake copies of themselves when they shouldn't and in places where they don't belong is probablythe easiest way to think of it.

there's a wonderful set ofslides which i have borrowed from that the nationalcancer institute has made that you can download and look at inyour own time if you want to see -- learn a little bit morejust about the basics. but this is -- and this isone of them, for example. so you can see at the topyou've got these cells that go through normal divisionbecause as cells get old or die, they need to be replaced. or if they get damaged,they need to be replaced.

so that's a natural process. it's when that goes wrong,the control of that goes wrong and you start to get growth ofcells where they don't belong and it's no longer controlled. we usually call that a tumor. and they come in different types. they can grow in one placeand never go anywhere else. they can grow in one place andstop and do very little harm. but the ones that are thebiggest problem and for

which breast cancer isan example are the ones that find -- that don't stop. they don't stop in the breast, theydon't stop in that little place where they started and they startto spread into the tissues around. and eventually they get intothe lymphatic system or they get into the blood system andthey go all over the body. and some of them will stick in otherparts of the body and start to grow, and then you have what'scalled metastasis. and you get spread of thedisease all over the body.

that's usually the processthat kills most women and the small number of menwho get breast cancer -- that process of metastaticadvanced disease. so this is what the breast looks like when you're thinkingconceptually. the breast, of course, is therefor one purpose only: to make milk. and we wouldn't be hereas a species if it wasn't for the milk that'sproduced in the breast. and it's made in these little --in these glandular structures here.

and then it gets expelledinto these ducts and comes out through the nipple. that's the natural process that occurs during what wecall lactation, after pregnancy when a woman has given birth andshe's making milk in her breast. sometimes things go wrong inthe cells that line these ducts and you end up with a tumor. and you can see here this is anexample of a tumor that began here in this quadrant of the breast.

sometimes it will getout of the breast and end up in the lymph nodes. so there are lymph nodesthat are in the breast -- that's often call thesentinel lymph node. and there are the onesturned the arm in the axilla. and so often when there's aninitial diagnosis of breast cancer, the doctor will want to knowis it in the lymph nodes, either in the breastor under the arm? and that tells us a lot about --just knowing whether it's there

or not tells us a lotabout the potential of that cancer to bein other places. so often if it's confinedto the breast, if we can have the surgeon removeit and the radiotherapist come along and use radiation to burnaway and kill any few cells that the surgeon might havemissed, that woman can go away and she's completely cured. but many women, as you know,that cancer has already got to the lymph nodes, or to thelungs, or the brain, or the skin,

or the bone, or the liver. and then it's a much moredifficult disease to manage. that's the sort of naturalhistory of breast cancer from sort of soup to nuts, if you like. so let's look at the incidenceand what we think we know about what might causebreast cancer. so in this top panelhere where it says -- that circle there wherethere's known risk factors, most women who get a diagnosis

of breast cancer have noneof these risk factors. we really still don't know whatcauses most breast cancers. we know a number of thingsthat can increase the risk on a population basis. if we look at those women,we can say that if some of these women have beenexposed to one, two, or threes of these things,their risk is higher. but that doesn't mean that we cango in and say "and you're the one that has a risk that's twice as highas anybody else's as an individual

and you're the onethat's going to cancer." we can't do that. we can only look at large numbersof women and draw broad conclusions or associations about what is thiscorrelated with breast cancer? the panel below, the circle, shows you the age distributionof breast cancer. it doesn't arise inchildren before puberty. it's extremely rarein very young women. the incidence increases with age.

and the highest increase, as you cansee from that middle chart is just around the age when womengo through menopause. so it is a disease ofgetting older primarily but not obviously exclusively. in terms of why is that thecase, the longer you live, the more cells you have in your bodythat go through and are replaced. and every time you replace acell, there's a risk that the dna, when it's replicated,makes a mistake and you get a mutationby chance increases.

so it simply increases with age. that's one explanation, probablythe simplest way to think of it. it's obviously a littlemore complicated. but that's enough for nowto get the basic idea. if you inherit a brca1or brca2 mutation -- and that's shown inthe table at the top -- the risk of gettingbreast cancer is very high. but the percentage of women who inherit mutated brca1 isfortunately relatively low.

so it doesn't explain mostbreast cancers, it explains some. and we can find out from looking atfamily histories who's likely to be in a family that might havea brca1 or brca2 mutation, and we can identify those womenif someone in their family, for example, comes in with a breastcancer and we take their history. and we can sequence from justa buccal swab, for example. we can sequence that brca1 gene anddetermine whether they're in one of these breast cancer families. and that allows that family to makedecisions as to who else wishes

to get screened and what they mightwant to do with that information. we have a whole set of folkscalled genetic counselors who will help people workthrough that issue if they find that they're a carrier of this. one in eight womenis a lot; men, too. but not so many largely because theamount of breast tissue in men is -- is much smaller than it is in women. there's a small residual pieceof breast epithelial tissue where the cancers arisepresent in men also.

but it's so small that -- and thereare so few cell there are relative to women, that probably explainswhy the risk is much lower to men. on the panel that shows the youngwoman, breast cancer is caused by the interaction ofgenes in the environment. it's wonderful to say that. so what genes and whatin the environment? that's a mess -- that's much harderto figure out because we have tens of thousands of genes, and theymake proteins in different forms. we don't always knowwhat they all do.

and then there are so manythings in the environment. and if it's a combination ofthings and not a single exposure in the environment, it's very,very, very difficult to figure out what it is that'sactually causing that. so for most people,we really have no idea if they didn't inherita brca1 or brca2. we only know a few things thatmight elevate their risk slightly, and almost none of those areanything you can do anything about. so if someone says, "ijust got a diagnosis.

why me, what did i do wrong?" you probably didn'tdo anything wrong. almost certainly you did nothing. there is very little that you can do to fundamentally changeyour lifetime risk. there are things thatyou can do that are -- that can moderate that to somedegree: exercise, a healthy diet -- the things that everybody keepstelling you to do for a whole series of different diseases can help

to keep your risk ofbreast cancer down. but fundamentally, what is it that'sdriving it is still something we still have a long wayto go to understand. and these are the numbers. now, they are kind of scary. 230,000 new cases every year. 230,000 families will get adiagnosis of breast cancer because it's not just the woman whogets it, everybody in the family and friends in thecircle are affected

by a diagnosis of any cancer. that's 231,000. and at 41,000 women dyingof breast cancer every year, that averages out if you just do thesimple math of one every 13 minutes. we will be here 45minutes to an hour. so you get a sense of the impactthat has across the country. it is still a terrible,devastating disease. and it has tremendous implicationsfor public health, for economics -- not just the human condition.

we have to do much better than this. now, interestingly, 70 percent of those breast cancers express aprotein called the estrogen receptor that binds to thatfemale hormone, estrogen. and we've known thatif you target that, it has a significant survivaladvantage for most women. and we've known thatsince the 1970s. first ever molecular-targetedtherapy for any cancer wastamoxifen for breast cancer.

this shows you that theimportance of the estrogen receptor because it shows you that estrogen-receptor-positivebreast cancers and estrogen-receptor-negativebreast cancers don't behave exactly the same -- the risk of thebreast cancer coming back, which is distant recurrencein that panel, or the risk of dyingfrom breast cancer. the patterns of those over a woman'slifetime are really quite different -- the risk of dying

from estrogen-receptor-negativebreast cancer is higher than estrogen-receptor-positivebreast cancer in those first fewyears after diagnosis. after that, the riskbegins to get higher for estrogen-receptor-positivebreast cancers. and although those women often livemuch longer, they actually have on an annualized basis a higher risk of their breast cancer comingback the longer they live than those women who'vemanaged to survive that long

with an estrogen-receptor-negativebreast cancer. so i said we can target the estrogenreceptors; how do we do that? and there are basicallytwo types of drugs that target the estrogen receptor. there are some that will bind thereand stop estrogen binding there, and those are called anti-estrogens. and tamoxifen is agood example of that. and that's shown inthis bottom part. so it's binding tothe estrogen receptor

so estrogen can't sit there. the other group arecalled aromatase inhibitors because they block the abilityof the body to make the estrogen that would otherwise bindto the estrogen receptor. so you either block thatprotein that binds estrogen or you block the abilityto make estrogen. and those two classes of drugshave been really quite remarkable in reducing the risk ofrecurrence and death. and that's shown here onthis particular table,

that the risk of recurrence is cutalmost in half and the risk of dying from breast cancer if you haveestrogen-receptor-positive disease is reduced by about a third. that's pretty good, butit's not good enough. because you can see on thatgraph at the bottom here, you can see the survivaladvantage -- the overall survival advantageof taking tamoxifen -- but you can still seethat there are plenty of women who don't survive still.

now, they may live longer andstill die of their breast cancer, which is a fabulous thing to do. if you think about that top bar forthe incidence is highest for women who are 50 to 60 years of age, if you could shifttheir survival 20 years, they get to see theirkids get married, they get to see their grandkidsgrow up, they may even get to see their grandkidsgraduate from college. even if they stilldied of breast cancer,

you have transformed the lives ofthose women and their families. if we could make breast cancera disease you died with, not a disease you died from, that alone would beutterly transformational. so think in your own mindshow you might choose -- how you might choose to define cure. because if you die of somethingelse and you've had a good life, if we can get that far,that's a great step. curing it is the last step, then.

so think of it that way. so here's some basic questions. i'm going to speed up alittle bit, and don't worry if you get left behind because i'llbring it all together at the end. and if you want me to -- if youget lost and you need to stop, just raise your handand ask your question. so i posed a couple ofquestions here because i'm going to give you some insightinto how we think about them. so i showed you thatrecurrence curve

that some breast cancers cancome back 20 years later after -- when there's been noevidence of disease at all and you think you're done. are the ones that come back 20years out the same as the ones that come back threeor four years out? we don't know. so maybe we can beginto ask that question. and if we could findthat they were different and we could understandhow they were different,

maybe we could take all those onesthat recur in three and four years or five years and not havethem recur for 20 years. and for many women, as i just said, you'd basically have curedbreast cancer for those women because they die of oldage or something else. so how you define where the finishline is frames the way you ask the questions, the questions that youask and where you look for answers. so if we could learn something fromany differences that we might find, we might have different ways ofidentifying the women who were

at the highest risk of recurringearly and we might be able to finds ways of not having themrecur at all or having them recur to light that they wouldhopefully never experience -- have that experience. and the fundamental questionis: if these drugs are so good for a high proportionof those women, why aren't they good for everybody? what is it about some breast cancersthat have that estrogen receptor that we know we can hit andhit hard that still come back?

because if we couldunderstand that difference, we might be able to cure them. so there's some of thequestions i want to sort of pose. and then i want to show youwhere research is going. we can talk about the clinicalimplications in the questions. i can only answer those vaguelybecause i'm not a clinician. so let's ask the question,first of all, are all tamoxifen failures the same? are those tumors thatcome back late different

from those that come back early? and there are differentways to do this. and i wanted to give you asense of the power of the tools that we have today that we didn'thave 10 or 15 or 20 years ago. because this is where the hope is. we can ask and answerquestions in ways today that when i was a student gettingmy phd i could not have dreamt of in one lifetime. and i'm not done yet.

don't think so. i've seen a transformation andan acceleration of our ability to create knowledge and turn thatknowledge into actionable outcomes that can make a differencein the lives of people. i've seen that change so quickly andso much in the last 10 or 15 years. this is the most excitingtime to be trying to find a cure for breast cancer. and it's the best time we'veever had to get there quickly. so this is an example ofone of those technologies.

this is a technology thatallows us to take a little piece of a breast cancer andlook at the expression of every gene that'spresent in that piece of breast cancer, in one experiment. so we can measure theexpression of 40,000 genes in one biopsy from a woman. and if we can do that,maybe we can take biopsies from those breast cancers that willrecur early and those breast cancers that will recur lateand see if they're --

there's a pattern of geneexpression in those two groups that predefines the outcome. so we'd be able to say whichwomen are at the greatest risk of recurring early or late. and then we could askwhy are they different and why do they recur earlyor late and learn something about the biology thatwould allow us to look at new ways of creating treatment. so this is a study we didwhere we did exactly that.

we took breast cancersat the time of diagnosis but we had 20 yearsof clinical follow up. so we knew which women recurredearly and which women recurred late. and we measured those 40,000 genesin each one of those breast cancers. what this shows you -- andyou don't need to be able to understand the figures -- if you look at thosetwo boxes in the middle, you'll see there's a redline and a blue line. so the red line --each line tells you

when that breast cancerrecurred or that woman died. and you can see that the red lineis very separate from the blue line and that the ones on the red line,their breast cancers came back early and the ones in the blueline came back late. so we were able to find a patternof genes that told us whether that breast cancer was goingto come back early or late. and we were able tofind another set of data that was completelyunrelated to our data -- this is a scientific question --

and show that we coulddo the same thing. so for us that told us -- ourinterpretation was, "well, there's something different aboutbreast cancers that recur early and breast cancers that recur late." now, i'm not going totell you that this is -- this tool is usefulfor telling any one of you whether your breastcancer is early or late because we didn't askthe question that way. we asked the questionjust to learn something

about the biology this disease. so they're -- theyseem to be different. that's good, we'velearned something. so now we do what scientistsdo: we set up hypotheses. we say, "well, if that's the case,what might explain those differences and how would we ask that question?" and so we took a differentapproach from others. we had 40,000 measurementson 140-odd breast cancers. and we said well, what dowe really want to know?

and rather than lookfor a single gene -- because we've alwayslooks for single genes. and the best we could come up withwas brca1 and brca2 mutations, and that's less than tenpercent of all breast cancers. we have to think differently. and you think about whatthe estrogen receptor is, it controls everything thatbreast cancer cell wants to do. you take it away andthose breast cancers -- some of those breast cancers die.

it's that important. so what is different between thosewhere you take it away and they die and those you take itaway and they don't? so we're going to thinkof this as a system. it's called systems biology. we're not going to try andfind one gene or one pathway; what we're going to try and find out is how does these breastcancers cells work as a system? how do they coordinateeverything that they do

to survive the stressof these drugs? and will that teach usabout breast cancer? and so we have a network idea,that concept doesn't matter so much as where do you start looking? i mean, you've got 40,000genes, you've got all the things that cells do; what doyou really care about? well, this is cancer. so we thought let's not ask allof the questions that we could, let's just figure out what it isthat makes cancer cells live or die.

because if they die, we can go home. we're not there yet, obviously. but if they don't die and they nevermade another copy of themselves, that would be a casewhere a woman would die with breast cancer,not of breast cancer. so then we want to know ifthey're going to not die, what is it that makesthem make copies of themselves when they shouldn't? so let's not worryabout the other things,

which are very importantquestions to ask about biology -- how do they spread, how do theyget to other parts of the body, how do they live in otherparts of the body -- let's just ask alive or deador making a copy of yourself or not as a place to start. and so that's -- so that thenconstrains where you look in all of the data for the answers because you already knowwhat some of these genes do. and they're not doing what --

anything that's relatedto being alive or dead for the cell or making a copy of it. so you can ignore those. and it gives you a place to look ina focused sense to learn something about the biology of the disease. so i've shown you that we could -- we had all these breast patients'tumors and we had all of this data. and we knew we could separate them. so now we knew which wereearly and which were late.

so what are the genes thatare different between early and late recurrences, anddo they teach us anything about why they may be different? so you don't need to knowthe tools that we used. we used a computational modelingtool that basically said, "okay, we know the estrogenreceptor's important. let's first find out all the genesthat we think or all the proteins that we think are likely tointeract with the estrogen receptor or interact with theprotein that interacts

with the estrogen receptor." and we're not worriedabout anything else. we'll just say what is thatclose in that data space to the estrogen receptor? and then when we find those, andthere's about 50 of them that are -- no matter how you look -- keep coming up as being relatedto the estrogen receptor. we just look and see howare those expressions for those genes different betweenbreast cancers that come back early

and breast cancersthat come back late? so once you have a focusedquestion and you have the data, you can begin to lookat it in different ways and answer a very focusedspecific question. so that's kind of what we did,and this is what we found. basically, it's complicated. you kind of knew thatwas coming, didn't you? but also, it suggests thatthere are some relationships -- some genes that are always up,some genes that are always down,

some that are alwaysprobably talking to each other no matter whatestrogen-receptor-positive breast cancer you look in that would allowus to think if we were to knock down one of those,maybe we could stop all of this signaling from happening. so now it gives us a place tothink about where we want to look to understand the biology andhow we might want to think about where we want tolook for new treatments. this is one way of looking at onedataset -- one or two datasets.

so that's something. now we've got some genes andsome ways they talk to each other that tell us that wethink that's important. so what if we take cells-- breast cancer cells -- and culture that we know can bekilled by an endocrine therapy like tamoxifen or [inaudible]and we treat them and we look at the changes in genes atexpression over time and we can look at how genes -- howcells respond quickly when they're first treatedwith an anti-estrogen.

and that gives us another model. and what we find is some ofthe things that are present in the patient's tumorsthat seem to talk to each other alsohappening very quickly -- literally within hours of thosebreast cancer cells being exposed to the drugs. so are any of thosehard wired later? does the cancer learn how touse some of those to survive? so yeah, they kind of do.

because if you take the same cellsand you treat them long enough so that a small number ofthem survive and they grow out and they're resistant, you can ask "how are the resistant cellsdifferent from the sensitive cells?" so what's now hard wired intothose breast cancer cells that now are surviving the drug? and it's some of same genesthat predict from early or late recurrence thatare changed very early when the breast cancer cellsare exposed to the drugs.

they're now wired in hard. the cells have adaptedtheir biology, their signaling to survive the drug. they've learned. we have to learn how they learn and what they learn tobe able to stop that. so we know what someof these genes do. we know what functions they controlwithin the cell that might allow that cell to survive, or die,or make a copy of itself or not.

and when we start to putthat information together in a systematic way,we begin to get insight into how the breastcancer cell works. and i'm going to tell you oneway that we think it works. so these words on thefar right in red and blue representfunctions that cells perform. so apoptosis is a processof cell death. we like that in cancer cells. we want them to do more of that.

autophagy is a fascinating process. that's the cell's recycle plant. so when bits of itget old or damaged, it eats them up so they can bereplaced and the cell can continue. but it doesn't waste thebyproducts of what's chewed up; it uses those to live off. because nature andlife is very careful in how it uses its resources. so there's something goingon here about alive or dead,

whether that apoptosispiece is switched on or not, and whether that recycleplant is activated. and that upr -- i'm goingto tell you what that is in a minute, so just hold on. it doesn't get too complicated. so what we're really looking for nowis how does the cell coordinate all of the things it needsto do to survive and to make a copy of itself? what all does it need to dojust to do those two things?

we -- that's the piecewe care about. we know that when we givethese drugs, they are -- the cells become stressedand some cells can't deal with that stress and die. this is good. but some cells are stressed bythese drugs and they don't die. and that's bad. and, again, we're trying to understand the differencebetween those two.

now, we know that to doanything the cell needs energy. we all need energy. so one of the places we wantto look is how the cell -- the breast cancer cell --controls its cell metabolism. because without energy,it can't do anything. we've known a lot aboutcell metabolism for a very, very long time. but we've always studied yeast,or simple cells, or normal cells. so a lot of what we know is hownormal cells do what they're

supposed to do. but these are cancer cells -- though don't do whatthey're supposed to do. so can we learn something abouthow they manage their metabolism so they have the energyto make a copy of themselves in order to survive? hang in there. this is going somewhere. i -- i assure you.

so we can do that, too. we can actually measure thousandsof metabolites at the same time on a single specimen, just like wecould measure all of those genes, we can measure allof the metabolites. and we can look at which metabolitesare being used more frequently and which aren't, and thatwill tell us which processes and pathways the cellis preferentially using. and so some of these are listedhere -- glutamate and glutamine. you see, we've actuallyknown for a very long time

that cancer cells are addicted toglucose -- the sugar glucose -- and to the amino acid glutamine. we've actually known that --i'll show you in a minute -- for even longer than i've been here. that's how -- age-wiseon this planet. that's an example of some thingswe've known for a long time. and we haven't knownwhat to do with it. so the bottom tells you some ofthe functions that are changed. and you've got celldeath, that's good.

cancer, well, we knew that. glucose metabolism. so we know that there's so much -- that glucose metabolismmust be important in how these cells become resistantand evade the anti-estrogens. and autophagy, that process,that recycle plant process, that's also been upregulatedin these cells. so now we have to think"how are they connected? and what way does that connectionallow the cells to survive

and make copies of itself?" and i've shown you that we've got-- we can measure the metabolites and we can measure thegenes that are expressed. well, we can take those two typesof data and map them onto each other and begin to understand howthe cells uses its proteins and its genes to regulateits metabolism. now we can begin to understandhow it functions as a system and how it makes choices. and we find things like cellsurvival, we find insulin and igf,

which would regulate glucosemetabolism normally are also changed. they do other things. and we see that some of thehigh-energy metabolites, the levels of those aredifferent in the cancer cells that are sensitive and resistant. so this is what we've knownsince about the 1930s. so when my mom was born,this guy, otto warburg, was figuring out that cancercells are addicted to glucose.

and they metabolize glucose in away that's very, very different from most -- but not all-- but most normal cells. because cancer cells -- cancersdon't have a proper blood supply. so they don't have all of thenutrients and oxygen and all of the bits and piecesthey need as easily. it's not as easy for them to getas it is for the rest of our body. and so there's not enough oxygen. and normally that would change theway cells would metabolize glucose, but cancer cells don't care.

they've learned to do thiswhether there's oxygen -- enough oxygen present or not. and they just love glucose. and glucose gets brokendown naturally to produce the chemical energy thatthe cells use to build everything. one of those moleculesis called atp. and you don't reallyneed to know much more than there's a chemical that's madefrom glucose, that's a byproduct of breaking up glucose, thatthe cells use to store energy,

and then they can releaseit to build other things. so we looked at whether the glucoselevels are changed in sensitive or resistant cells andwhether that's effected by the anti-estrogens. because that's fundamentally -- if we're changing their abilityto make energy, maybe that's one of the reasons they're dying. it turns out that that's the case. but what's really interestingis actually

on that atp graph on the right. you -- i'll tell you what itsays so you don't have to worry about whether you can read it. it says if you give ananti-estrogen to sensitive cells, they take up less glucoseand they have less atp. they don't have enoughenergy to do everything. and what that normally doesis first that causes the cells to stop making copies of themselves. and if you can hold thatglucose levels low enough

and the atp levels low enough,some of those cells will die. they -- they just can'tsurvive any longer. the resistant cells have figuredout how to take up glucose and make that apt even though they stillhave the estrogen receptor present and the drug is present. they don't care anymore. they've figured out away to get around that. so we don't quite knowyet how they've done that, but we know that they've done it.

but the other thing that told uswe needed to look elsewhere, too, was -- and that's thesepanel on the far right -- the cells that are resistant andare growing happily at the same rate as the cells that are sensitive when they're not beinginhibited actually have less atp. they don't need as much energy. well, why do resistant cells thatare doing all the same things as sensitive, normal cells are,why do they need less energy? that's where the recycleplant begins to come in.

so they're clever little suckers. so what happens in a cell ifyou don't have enough energy? well, the first thing isyou actually can't make and fold your proteins properly. you need proteins through all of thebusiness that the cell needs to do. and cells have acrossevolution been exposed to very different environmentswhen their nutrient supplies and their ability to makeenergy was constrained. but they still survived becausethey developed a system to deal

with not having enoughenergy to fold your proteins. it's called the unfoldedprotein response -- that's that upr thing ipromised i would tell you about. and you can find thisin yeast cells. it's a little different in mammaliancells, but it's not that different. and it's a damnably cleversystem, as you would imagine. because if you don't have enoughenergy to fold your proteins and you want to survive first --survive first, grow second is a sort of priority for the cells.

normal cells have that priorityjust like cancer cells do. and if normal cellsare able to do this, why couldn't cancer cellsuse this as a way to survive? and what the unfoldedprotein response does in this cell is itsays "i have a problem, i can't fold my proteins properly. let's make less protein, figureout how to get everything back to normal, and then we'llmake a copy of ourselves." in the process it doesn'twant to die.

so it also sends a signalout to the mitochondria, the energy piece in the cell. it says "wait a minute,we've got a plan here. i know we have a problem;we're not going to die yet. so signal, just don't die yetbecause we're going to fix this by making less protein, making onlywhat we need until we can survive. and then we're going togo off and we're going to make a copy of ourselves." and that unfolded proteinresponse coordinates all

of those decisions in the cell. and cancer cells canuse the same thing. and that takes us back to thislittle model that i briefly went over and told you noneof the details. because what that modelfrom our study said was that seems to be important here. of all the things that this -- thatthe breast cancer cells could do to survive the drugsthat we're giving them, they seem to be using thisunfolded protein response.

and we learned that one of thecoordinating signals that comes out of from the unfolded proteinresponse also coordinates the recycle plant signal. so now we're beginning to see howwhat the cell's doing doesn't have enough energy, so it's nottrying to make copies of itself. it's going to survive first. to do that, now it doesn't need asmuch energy because it doesn't have to make a whole othercell to make a copy. so it can live with a lot less.

and at the same time it's sending asurvival signal to allow it to see if this is going to work or not. and that's what theseresistant cells do. they upregulate at the sametime that recycle plant because there they were happilygoing along, making lots and lots of copies of themselves,and now they don't need to. so all the machinery that they hadto make another copy is superfluous. they don't need it anymore,so they just recycle it. and they kind of live offtheir little bit of belly fat,

if you like, and finda way to survive even when all their metabolismis being shut down. and some of the cells nextdoor haven't figured this out and they die. and what happens to thebits of the dead cell? the one that's alivejust sucks them in. so they start to feed each other. so they start to feed themselvesoff what they don't need anymore, and they start to pull inthe nutrients from the cells

that are dying next door. so it's really an ecosystem. it's a whole system where all thecells are talking to each other and working togetherand some are dying. and because they're dying,they're actually helping some of the other ones to live. cancer is smart. and this simply shows --so i'm just about done. and i'm going to live you ananalogy of how we've come to this

to understand just these basicprinciples and why that's beginning to change the way we thinkabout dealing with breast cancer and dealing with particularlythis drug-resistant -- type of drug-resistantbreast cancer, which is in effectultimately up to 70 percent of all breast cancer patients. imagine a car. okay. you could have mini,you could have a maserati. i know which one i want.

but it doesn't matter. the basic principles of theinternal combustion engine that drives the mini and drivesthe maserati are exactly the same. what's different is allthe bells and whistles in the maserati that'snot present in the mini. but all the control systems -- thebrakes, the engine, getting the fuel in -- all of those thingsare basically the same. and that's kind of a little bit likelooking at cancer and normal cells. the cancer cells are usingwhat they've inherited

as from their having been oncenormal to allow them to survive when they're cancer andwe're trying to kill them. so think of where we are on a car. so we kind of know whereto put the key in the car. and it's like the estrogen receptor. it's kind of like the switchthat turns on everything. and if we turn it off, noticethat car's not going anywhere. it's stopped. if you leave it stopped longenough, it rusts away and it dies.

so we know where the key goes and weknow that if we mess with that key, we can cure some breastcancer patients. so it's a great place to start. what happens, then, whenthat key gets broken and turned on all the time? so what it's doing is it'sturning on the engine. it's firing up the engine and it'sgoing to make a copy of itself. now, we know how cellsmake copies of themselves. we know how yeast cells do it.

we know how normalcells in humans do it. so that piece of what happens incancer is often very much the same. what's not the same is the signalbetween the switch and the engine. that's what's got changed in cancer. so that now begins to tell uswhere to look to understand in more detail, a muchfiner place to start looking to understand how thesecancer cells are different. we know, like, the car ain't goinganywhere without gas in the engine. and the cancer cellisn't going anywhere

without the same gas in the engine. it makes it differently. it's like it's got a differentgrade of fuel, if you like. but it's still ultimately the sameenergy source, the same fuel source. if we could understand how thatwas different, we might be able to stop it from making its energy. and in a sense that's what theseanti-estrogens do when they work: they turn off the switchand they take away the fuel. the real thing is howdoes that happen?

now we know where wereally need to focus. and the basic principles,when you think of it that way, are on my last -- ormy almost last slide. so this is how we thinkit works as a system. this is where it all comes together. you give an anti-estrogen, you don'ttake up enough glucose or glutamine, fatty acids, but you don't havewhat you need to have enough energy so you can't fold your proteins. you can't fold your proteins,

that activates theunfolded protein response. and that coordinates the don'tdie yet signal and lets turn up the recycle plant because we'regoing to live off that for a while. and maybe we can pull someother things in from outside. that's the way theintegration works. that's where if you lookat it as a whole system, you see a very differentpicture than if you just try and find a single geneor a single mutation that you think mightdrive the whole system.

because the estrogen receptor getsmutated in some breast cancers. and that's not the whole answer. it's when you look at itas a system that you begin to see now there aresome vulnerabilities. there are vulnerabilities inhow the cell makes metabolism. there are vulnerabilitiesin how it coordinates that -- that recycle plant. we could turn the recycle plant off. we could turn that off, we couldturn down the cell metabolism,

and we might get a completelydifferent response using drugs we've never thought of using before. so that's kind of the bottom line. and what have i told you today? because i've covereda lot of ground, and i've probably talked too long. we have to fix this problem. we have to understandhow this works. the cancer doesn'texist as a single cell;

it exists as an organismalmost within the patient. it talks to itself. the cells talk to themselves. they use the host's responsein some cases to survive. at the same time the host is trying to eliminate the cancerwith its immune system. why hasn't that worked? that's another piece of the puzzlethat we haven't cracked open yet. because if the immune systemworked, you'd never have cancer.

they would just be identifiedas being foreign or screwed up and eaten up and removed. that's obviously anotherpiece of the problem. so there's a lot more to think of. but if you start thinking of it asa system, you begin to ask "well, what is it that's in theimmune system that's talking to the breast cancer?" you know, just thinkwhy does it not work? you look at the questionin a different way

and you see differentparts of the jigsaw puzzle. with one woman dyingevery 13 minutes and 70 percent having thisbiology when they start, we need to do something that'sfundamentally different. we have drugs that work for somewomen, we have drugs that kind of work for some more women, and wehave the same drugs that don't work at all for another group of women. we need to be able to do better. and there are drugs coming alongthat are playing in this area

that are probably goingto be very helpful. we have drugs now that arecalled cdk/46 inhibitors. what that means is they're very good at also turning off theproliferation signal. so when the anti-estrogens comealong and shut down some but not all of the cells, these otherdrugs come can come in and shut the rest of them down. don't know if they'll cause themto die, but maybe if they're shut down long enough, they will.

we have the tools -- and i've givenyou just an example of a couple -- to ask the questionsof these systems in ways we could never do before. our ability to generate datafirst, and then translate that data into knowledge, and from that knowledge identify theactionable items that can lead us to better preventionsand better treatments for breast cancer ismoving very quickly. and it's very exciting.

and the potential for making abig difference is greater now than it's ever been inany time in the past. those are tremendouslyexciting things to think about. we have the ability to do this. now is the time to do it. we shouldn't wait any longer. we shouldn't constrain the resourcesthat we put to curing this disease. this is a time to do the oppositebecause now we have the tools that will allow us toanswer this question

in ways we could neverhave done before. obviously, what i've told youinvolves a lot of different people, and it's nice that i'm ableto put their names up there. and you're welcome to read them. but there are namesthat are not up there. and they will never be up therebecause i'll never know their names. i'm not allowed to knowtheir names, and that's fine. it's fine that i don'tknow their names; it's not fine i can't thank them.

those are the women who gave usthe tissue in the first place, some of them 20-odd years ago. they're not here anymore. some of them died oftheir breast cancers because they had early recurrences. some of them died oftheir breast cancers later because they had late recurrences. some of them probably i'mcertain are still alive. i don't know their names.

i can't thank them. without what they did, wecould not have got even started on this pathway. so the people that are mostimportant are the patients who contributed to thestudy in the first place. thank you. it is so important to hear frompatients what their experiences are, what their needs are,what their questions are. it's not just about the research.

and it's just lovely tosee someone say "i had it in the 1970s and i'm still here." because we'd love that tobe the story for everyone. now, can we take what we'velearned and make it -- and turn it into new treatmentsfor prevention, or to eliminate that cancer, or makeit never come back? i think we can. but we started to think abouthow we do that very differently. so the way that -- i usethe example of a car.

so the same way of thinkingis the way we're beginning to change how we -- ithink we should be thinking about how we combine drugs to geta better effect than just randomly or following a logicallinear way of thinking. this is a non-linearway of thinking. if you really wanted to stopthe car, you'd break the switch, you'd take out the battery,you'd drain the fuel tank. we've kind of thought of differentways of breaking the switch, but we never thought of "okay, ishould really take the battery out

and drain the fuel tankas well because, you know, we want to make darn surethis cell can't function." so in the past -- and itwas a fine way to go -- and the cmf therapy that you got isa good example of we took drawings that had differentmechanisms of action and didn't have the same toxicity sowe could give them in combinations at the highest doses we could. and that's fine, but it's --it's a practical, pragmatic but not hypothesis-driven approach.

it's what you can do. it doesn't mean it'swhat you should do. sometimes it is. sometimes there's better ways. so we're trying tothink differently. so let's not just takethree drugs like that because they have differentmechanisms of action. they're not -- not that different. [inaudible] are similar in a sense.

but we want to now targetdifferent parts of the whole system. the more pieces of the systemthat we can knock out at once, the less chance that cell hasto figure out how to survive. and that's a different wayof thinking because we had to understand what those functionswere, how they might be wired, and talk to each otherbefore we could even think of doing it this way. and we still don't know enoughto do this rationally yet. but it's taking -- it's allowingus to begin to think differently.

and sometimes that's just enough. you think differently, you may get to a place you neverthought you'd get to, but it's still the right place. we do know that there aredifferent types of breast cancer. we know that estrogen-receptor-positiveis different from estrogen-receptor-negative. we know that -- that amutation in a gene called her2

or erbb2 also creates a specialgroup of cancers that we can target by targeting that growthfactor with drugs like [inaudible] asyou probably heard. so at the very least there arethose that have the expression of the estrogen receptor,those that have expression that mutated oncogene her2, andthose that don't have any of those. and the other one that goeswith the estrogen receptor is progesterone receptor. and that's why that last group issometimes called triple negative

because it doesn'thave estrogen receptor, progesterone receptor,or mutated her2. at that level there'sat least three. once you start asking "arethere other groups within that?" then is opens a whole can of worms. there are, dependingon what you look for. different mutations you'll findare scatter in a different way. other pathways thatare differentially -- you'll find that as subgroups.

ultimately, for the momentthe type of treatment that a woman gets isdetermined by whether or not there's the estrogenreceptor there to target, whether there's her2there to target, or whether there'sneither of those to target. and that's the triple negative, and they unfortunately have thehighest risk of recurrence and death in the shortest periodof time since diagnosis. and we only have chemotherapyfor the moment for those.

and they seem to be aquite heterogeneous group of breast cancers. and it's about 12 to 15percent of all breast cancers. the subgrouping is usefulin my personal view -- and this is only my personal view -- if it teaches us something about thebiology we didn't know that leads to an actionable outcomethat allows us to identify which women should betreated with which treatments because that's the best for them.

i'm not sure we've quitegot there yet with a lot of these subtype analyses. we can say from some of them that this woman has aworse outcome likely or a really good outcome likely. so some of the tools aregood in a prognostic sense. and the real value thereis who not to treat. because some of the veryearly stage breast cancers like i described [inaudible]if the surgeon gets in

and the radiotherapist gets in andit's gone, you don't need chemo. so why would you take it? it's not a very pleasant experience. so the real goal there is if yourrisk of recurrence is so low, why would we use chemoat this point? so those are the sort of questionsthat we're able to do better at. but which chemo should you get? we're not there yet. what new drugs, what combinations

of current drugs would give us thebest outcome for you versus you? but those are exactly thedirections that we want to go in. we'd like to be able to say that -- that your cancer is unique insome way, and we understand that uniqueness, and thatuniqueness is a vulnerability in that cancer that we can target. that's the sort ofprecision medicine concept -- one way of expressing theprecision medicine concept. i think we will find better ways

of using the existingdrugs first, i suspect. and we'll see the benefits of thatearlier simply because we don't have to wait for a single drug to beapproved and sure that it's safe and efficient, and then identifythe patients that need to be used, and then figure out how to use itin combination with other drugs. that's a long process. and we need to keep doing thatbecause we need to find drugs that hit different parts ofthat machinery more effectively. from where i sit, we havedrugs that target metabolism,

we have drugs thattarget proliferation, we have drugs that target survival. we've never understoodhow we should use those or whether they're the right drugsto hit those targets if we were to. but now we can build mathematicaland computational models with the data that allowus to ask those questions. and those questionsbecause they're done in the computer can be askedhundreds of thousands of times and very quickly andmake predictions

as to what we can do quickly totest first in cell culture models and then in preclinicalanimal models to see -- make sure that those drugcombinations are safe effective and then quickly get them into womenbecause the drugs are approved. that is -- so that'sa drug repurposing or redesigning the cocktails. and there are some reallyinteresting clinical trial designs that i won't go into todaybecause i'm not a clinician and i wouldn't be ableto answer your questions.

but there are some reallypowerful new clinical trial designs that are beginning to try and getat that, where women can come in and they can get treatedwith one set -- series of drugs, and then uponrecurrence they automatically get into a different box. we think that there mightbe ways of better selecting when a woman gets what drugand what box to go into next. and so these new designs whichare using existing drugs -- and a small number -- will probablyteach us how to do that better.

and we'll be ready, then,to do those experiments -- to do those trials in women oncewe've completed the experiments that tell us how to mix and match. so my prediction? i -- i make this as aprediction for my lifetime, and since when it's over, i won'tbe here to know i was wrong, i think we will makeestrogen-receptor-positive a disease that women die with but notof in my lifetime -- i think. triple negative?

a little longer. if there was a brca12 mutation, the risk of ovarian cancer is alsoincreased along with breast cancer. so that's the angelina jolie story. so -- and that's -- i mean, againthat's a very personal choice for every woman in consultation withher physician, removing the breasts and then removing the ovaries toeliminate the risk of both breast and reduce -- greatly reduce -- therisk of breast and ovarian cancers. so, again, that's -- we have -- wehave professionals who help women

and families work through those and find what is theright solution for them. so that's probably therationale for that. so her2 is a protein thatsits in the membrane of -- of breast cancer cells, andit's mutated in a way that means that it's constantlysending a growth signal, a proliferation signal, a survival and proliferation signalto the breast cancers. and they are -- a significantproportion of the ones that have

that are depending on thatsignaling in a way that's similar -- conceptually similar -- to beingdriven by the estrogen receptor, not as being driven by her2. and so if you come along withdrugs that block that signaling, the cells can't surviveand they can't proliferate. and about half of those have noestrogen receptor and about half of those have estrogen receptor. the ones that don't usually geta chemotherapy at some point. the ones that do can stilltake tamoxifen or letrozole --

an anti-estrogen aromataseinhibitor -- and they will sit gel some benefit from that even thoughthey've also got that driver mutationin the her2 gene. so it's an interestingintermediate biology in one sense and a completely differentbiology in another. >> tomoko steen: please joinme thank you for [inaudible]. >> robert clarke: thank you. >> this has been a presentationof the library of congress.

visit us at loc.gov.