Brain Cancer Prognosis

>> okay we're going to get going, we have a switch in the schedule, so we have eva, to go first. and she graduate from yale university and got her medical degree at duke university did an internal medicine residency at bell view new york

university medical center and then she came to nci. her title nonsmall cell lung cancer --eva! >> can everybody hear me? okay, very good. so this is really strange because i see five of you out

there, but i know that there are others ulcwhere so,--elsewhere so online, okay. so anyway, we'll go forward. so terry asked me to talk about nonsmall cell lung cancer. this is a very broad talk, it covers everything from treatment to prevention which is m area

to early detection, so, bear with me, as i go through all of this. so i've been doing this talk now for a long time, from the start, which i'm i don't remember when that was and here's the number, every year as the population grows we still get more new

cases of lung cancer. it's not really--although numbers for a hundred thousand may be not going up as rapidly, the absolute numbers are still going upper as are the deaths. lung canc ser a major cause of morbidity and mortality more so than anything, it's the leading

cause of cancer death in america, it's a huge cause of mortality world wide, more than breast, prostate and colon cancer, deaths put together in america and all those deaths for a hundred thousand are decreasing the upload numbers are going up and up.

okay. survival on the other hand while it has improved in the past 50-60 years, it hasn't improved that much. so, 16% five years survival. and that's because you don't usually find lung cancer early and we will get into that a

little bit. so here is--this is from american cancer society, publications and web site. after people started smoking 30, 40, 50 years later, you see lung cancer peaked and as the rates of smoking have gone down, lung cancer is coming down, of

course, in women, women started sming later, and so, the peak occurred later. it was more in the 1990s, 2000s and it is starting to come down but you know, still a major cause of problems in the healthcare universe. so this is one of my favorite

pictures, it's an old ct scan, okay? there's the spine, there's the heart, these are the lungs, and this shows you what causes lung cancer, so, this little thing is cancer itself. here you have some fibrosis, scarring, copd, okay?

chronic obligations obstructive pulmonary disease or emphysema and over here you have the pack of cigarettes and there have you it, the main could you says of--causes of lung cancer. okay, the risk factor is about 85% of lung cancer is due toxic effects back o exposure and that

is tobacco exposure and although active is much more important. prior arrow digestive malignancy most of which are associate wide smoking are again a risk factor as is emphysema, chronic obstructive pulmonary disease, this is a whole slew of exposure, one of course is

asbestos, think about asbestos exposeddure in mesothelioma and that's what you find on tv, but in fact, the main cancer that is associate wide lung cancer with excessive exposure is actually good old lung cancer, radon various industrial exposures, oil refining, ship

building, et cetera, mining, arsenic, et cetera. so genetic predispositions, what's been most notable in the literature recently are these acetyl co line receptor on chromosome 15 q and what's interesting and helps to tie things together is that the same

abnormalities or the same snps are actually asusceptible for copd, scow the more likely to smoke if have you certain versions of the genes, and you're more likely to get lung cancer, and copd. putting it all together. so we're going to start very

globally in terms of clinical present septorsation, what you do with people, how you treat them and then move forward, so to speak in choosing prevention, so--yes ma'am. >> even closer, okay. >> [indiscernible]. >> is that better?

it is. sounds the same to me. okay, i'll go like this. that sounds better. okay, so, we treat lung cancer very much in an anatomic fashion. if it's small, early stage, one or two, small size, t-1, the

survival is better, but moment it starts to spread to the lymphnodes, so now we're talking stage two. that's where we are. stage two, then the survival plummets into the 30s, five year--over all survival and once you get to the media steinal

lymphnodes or to systemic disease, long-term survival is extremely poor. pathology, okay. at the most common form of lung cancer is actually adeno carc nome aabout 40% of lung cancer. it tebds to pick up peripherally, for the

micrograph, the next most common is squeamish cell carcinoma it used to be and the change in how people smoke and squamous, both are associate wide tobacco exposure okay congressman cancer tends to be adeno case noma and others. --and a very small percentage of

cases of nonsmall cell of carcinoid which are much better behaved, much less genetically abnormal. small cell lung cancer is the other large category that is associated with tobacco exposure, 20% of all cases. it's a systemic disease lecture

on this show i'm really not going to discuss it here, so today's it's all about nonsmall cell lung cancer. so if we're going to tackem this big lung care problem, how are we going to do it? we can do it through prevention? we can do it through detecting

it early. we can do it through stage one detection where it's much more curable o we're can treat it. ideally you treat because you know you have the disease, you don't need-to bring any larger group into the healthcare system but as i will show you, although

a lot of work has been done recently, it still has major limitations. and the whole process of lung carcinogenesis is a long process. it starts with tobacco exposure which for most people starts in their teens.

people don't get lung cancer in their teens, they get it in their 40s, 50s, 60s, the older you get, so it's really this long lag time when you get a serious of accumulating abnormal rites, both histologically, and what we talk about it as metaplacia, dysplasia of grades

and molecularly, so if you can intervene, you're in this early process, before you get an overt cancer that crosses throughlet basement membrane to invade, you potentially could do much better than treating an already metra static cancer. so of course we'll start with

the treatment, nevertheless where you can't cure people so treatment is very much based on the stage, lonely stage, surgery, early stage with lymphnodes, surgery plus a little more adjuvant chemo. it does better you get a five-10% decrease risk of death

wuthe nevertheless, the lopping term survival is not that great. once you get to regional spread, you talk about combined modality and usually chemo therapy and radiation because you want systemic exposures as well as local excitation of the tumor sometimes with surgery and then

once you get passed that where the cells escape friday that regional central focus, then you talk about chemee therapy and the use of radiation as necessary for local control occasionally for the resection of isolated metastasis, such as in the brain.

and they're also using all about very little surgery what's really exciting in the past few years is our understanding that nonsmall cell which needs to when i was a fellow, i don't want to say how many years ago was nonsmall cell verses small cell, and the developmental

endocrinology nition was histologic and rely because small cell was chemo response and i have nonshall cell was not and now we've finally learned that not all small cell is the same. in fact, these genetic subgroups act different, behave

differently and now we have tools toxic effects have them be different. so the ones you've heard about is the egfr signaling pathways. it's very important growth factor, halfway, involved in the many normal processes, all cells have egfr and molecules that

bind to it, to set into motion a number of different survival and proliferative pathways. but what's really become very exciting is understanding that there are certain inhibitors, small tyrosine kinase inhibitors, small molecules which have very profound effect

on certain percentage of people with nonsmall cell, namely 10% and it's the people who tend to be never smokers off the female often asian and they have egfr mutations and that that's a series of discreet agents through axon 21 and these people respond beautifully usually to

although these responses are as oppose to chemo whether the responsibility is there, so much better than chemo blue still not cures. but still not cures. but well, actually now, there are there are three main egfrs and actually the main one in the

there, we'll show recently approved and these drugs are now approved for first line, second line, third line treatment of stage four nonsmall lung cancer. this is just one study that shows these responses show typically in all comers the leading survival at best is

about 10 to 12 monthss, people the stage four lung cancer this, is a study showing in people who had egfr mutations who were treated either with chemo therapy or this case, just have it, or the first of these drugs to go on the market and what you see is that the progression free

survival, pss, in other words, the time between diagnosis and in progress according to cat scan is very significantly increased. instead of five months with chemo, more like 10 which we see here and overall survival in these people isot order of 30-36

months, okay? in this study, there was cross over, so they don't progress, you don't see them meeting survival but these numbers are substantially better for people who don't have egfr rotations. this has revolutionized how we look at people with lung cancer.

first thing you do is you look for egfr mutations and if you find them, it's not the type of mutation, and that's how you treat. this is eml four fusion genes in is another abnormality in nonsmall lung cell cancer, much less common than egfr mutation

and somewhat less than five% of all comers although there is a description of a distinct histology that look like rings, never nevertheless, anybody could have it. and once again, those are a very specific inhibitor and second

line inhibitors could document in this case where once again have you a 60% response rate and a lot of stable disease and people have much longer survival treated this drug then with the chemo therapy alone. so, eml four l, egfr are front line tests that we see, on all

people with stage four nonsmall cell, adeno carcinoma and if they have the abnormality, that's what they get for people with chemo therapy. the rest of the mutations are much, much, more common. i'd like to present the this as somebody we awe here at nci.

and this is a gentleman, again another smoker who had her-two, ampplification, a different part of the egf family, egfr family. this is breast cancer primarily but had in case, the gentleman had with one of our studies, where he had been treated initially with chemo therapy,

had very transient response, was treated with erlat nib, the main egf erne hickilator on the market and did not respond at all and then went to an experimental drug which is not only an egfr inhibitor but an her-two inhibitor and he had a quite nice response which you

see right here. this is his tumor, one of many, and this is what happened after he had this drug that was a her two and egfr inhibitor but it was again a transient response but what was noted is that his serum tw levels were elevated. so we went and looked at a

tumor, and looked to see what egfr amplification, in fact, he so we started him on herceptin. drugs that are used for breast cancer, amplified two breast cancer and actually he had no response but when we added the norle which was a standard chemo agent which usually has the

duration of response of about three months. he had a complete response, had one little thing that grew out in his neck, he took it out and we put it back on these drugs and he had another lengthy response and this gentleman has now been alive for seven years

with metastatic lung cancer, going from one her-two targeted therapy to another. actually, an outlier, without a question, but a remarkable case. and this is what we're upto now in lung cancer. this is roth one. again, a tyrosine kinase,

rearrangement, that occurs, in a very small number of nonsmall cell lung cancers, but these people also respond to redot nib, so we're starting to whittle away at these small subgroups of people which had its own challenges because you have to look for these

abnormalities molecularly, but when you find them, the responses e traumatic they're much longer lasting, than chemo, and they're very satisfying obviously to the patient, but even to the physician. but that's really adeno carcinoma the majority, of lung

cancer. what about squamous cell tcga and that was--the cancer genome atlas, project, as well as number of sequencing studies have identified certain abnormalities such as the fibroblast growth factor receptor one that are abnormal.

this ddr two is another tyrosine kinase mutations in schemish cell carcinomas, and there are experimental drugs, that are effective against these two. so we don't have quite the level of evidence but at least we have a place to go with some of these abnormalities in terms of

clinical trials. so that's really the status of the genetic approaches to with perhaps in its infancy right now, but--and it's not an area that i can speak to very well, but what is really exciting is immunotherapy or nonsmell cell.

and in particular, the check point inhibitors, pdone which is ingvolved in t-cell vacivation and in limiting antitumor immunity and inhibitors which are primarily molecules, or antibodies have shown remarkable efficacy and when i say remarkable, if you look at the

response rates which are on the order of 10 to 25 response rates perhaps that's not as exciting on the face of it, but what's really exciting it the duration of these responses. so on this one trial which uses merck compound which is an ntpd-one antibody, this response

rate yielded a median survival of 51 weeks which is--which is a year. but what's really remarkable is that for the people who respond, the majority respond long-term and not necessarily even while they're still on drugs. so there are people who are

years out from this kind of therapy. in other words they pale to the survival curve and that's a first for nonsmall cell lung cancer because with egfr inhibitors for egfr mutated nonsmall cell, there was no tail, people die, that tail goes

to zero. so this is really the forefront and hopefully, this will make a difference although again, it appears to be in a minority of people, it may leave people whose tumor expresses the liaison gabbed for this pdone-receptor and namely

pdl one. okay that,'s it for treatment. it's an overview, it doesn't do it justice, i apologize but at least you heard the buzz word. and those are egfr, pd-one, so on. the rest of the time i'm going to talk about other approaches

limiting morbidity from lung cancer namely prevention and a little bit on early detection. so again in the sequence of events, the early faces of lung canc nonapoptotic genesis are before the tumor,--the abnormal cells develop enough abnormalities that cause them to

go to the basic membrane and out into the blood stream or lymphatics. can we do something to prevent that process from occurring? let's startot the root cause, smoking. what happen fist you stop well, it's not as simple as one

would seem. if you stop smoking five years later, your risk of lung cancer has been pirseptorsiblely gone down. this study here, which is a long health study looked at 14 and half year follow up, median follow up in a group of people

who either quit, intermittently quit or continue to smoke. and you see the causes of death and what you see for the people that quit, you basically have a and have the number of deaths from lung cancer and people who quit intermittently and continue to smoke who are the one who is

have the highest mortality for lung cancer. looking at the same kind of differences in survival due to various diseases, the five year follow up did not show any differences in lung cancer. so it take ace long time before you start to see the difference.

it's not that you shouldn't quit smoking but you take home the benefits for lung cancer until years down the line where you see them quite quickly for cardiovascular disease, and the ability to breathe better and nonprogression of emseema and things like that.

but quitting smoking is not as hard as everything else, it's not. and the adult smoking rates are on 22-23%. very high. so, is there something else we can do and that's where chemo prevention comes in.

so chemo prevention is defined as the use of strategies. natural, synthetic immune, street edgys to suppress or reverse the process. and on reversing the a presssoever a long period of time. so regressing the preexisting

lesions, preventing and it really starts with the fact that if we could treat metastatic cancer we wouldn't want to focus on those early time frames because not everybody who smokes goes on to develop lung cancer and we want to focus on those people already, except that we

can't cure metastatic cancer we can extend survival but even with the best therapies i showed you, we don't do that l. people aren't cured. and we do have dat from other diseases that canc ser preventible in fact, in breast cancer in particular, using

tamoxifen or relocks fin or inhibitors. one can prevent second primary's contra lateral primarys or first cancers, and people who are at high risk. we can model this in animals with carcinogen exposure and genetic abnormalities and we

also know that it takes a long time to develop all those that would lead to inveighsive lung cancer so we have a population at risk we have a long time for intervention and we know it can be done at least, can be modeled and other diseases preclinically.

so what do we want to do, when do we want to intervene. so again, if have you an effect intervention, late stage, you don't need to go to the early stage and early stage cancer is much more curable than late stage. stage one, where it says much

better survival than stage four, one would hypothesize that precuriousors would be more curable than inveighsive cancers. they're not as genetically complex. could we prevent carcinogen damage yes, but then you need to

start with your premature but that would be the ideal thing. these are all issues. tells you when you want to focus. the other is the toxicity. you can give highly toxic agents in the short-term for instance when somebody has metastatic

cancer but you can't do that in the long-term. long-term you need to go with very nontoxic therapies. so that it's obviously a limitation and an important one. and you also have to think about things in terms of public health, the target population,

to treat is huge. right? if you take all the curspent former smokers, that's over 90 million people in the united states at any one time. obviously, we can't treat everybody, we don't want to treat everybody, but we

certainly can't afford tow treat everybody. so these are all issues that one needs to think about when can contemplating how to bring prevention to large population. it's all about the risk and benefit. risk preventing cancer, benefit

for promoting mortality. and the side effects. the adverse events that increase morbidity and mortality from other diseases. so in other words we prevent lung cancer but if people die from heart attacks you're not doing anything good, right?

and tolerability, and you can give people side effects that we don't consider medically important but at first you want to take the drugs because they're having the diarrhea let's say, it's not going to work. so all of these things go into

the risk balance. okay, so, again just a little bit of background. how do we idebtify new agentss for lung cancer, mechanism where we know it. hp vaccine is the best example of mechanism, hpv, human papilloma virus absolutely

required for cervical carcinogenesis. when you know the molecular pathogenesis, that's great. one cancer is a bit more complicated than that. so we go to preclinical models, animal models as well as invitro systems and i will show ,ou

examples of ones that have leditous clinical trials, we go to observational epiteemiology, cohort and case control studies which show us what agents may be of potentially utility, i'll show you examples of that. and we also look at secondary end point from clinical trials

that were perhaps formed for best exampling of drugs coming into the cancer prevention a this route is for tam ox o fen, which is used for adjuvant treatment, breast cancer and found to prevent second primary, et cetera, et cetera showing examples of where we go with

that with lung cancer. so i'm going to tell you three very short stories of areas that we've been focusing on, for lung cancer prevention. one of them is the inflammation story. long history okay, animal data going bag going 40 years or more

so showing that cortico steroids which are ain't inflammatorys can prevent skin cancer in animals that work was extended to look at lung cancer models again inanimals using [indiscernible] and inhaled steroids and there's even epidemiologically data and khi

is reasonable using that in people who have emphysema, copd who are treat wide cortico steroids as oppose to other agents that the incidence of lung cancer is much deyeased is is--devised and data. this is called final carbonnite which produces adenoma and adeno

carcinomas in the you get 80% decrease in the number of tumors that these numbers have and if you look at what these tumors look like histologically they are likely to be adenomas than carcinomas in the untreated animals. these are untreated.

these are treated. so you have the numbers of carcinomas. so you're reducing the notal number and preventing the progression. so we work with stephen lamb and the british cancer agency to do a trial. this is how you do

lung cancer trials. so we took so look at their sputum to see if there's any tip yarks idebtified about half of them having a tippia, asked them to have whereon coscopy to look inside the branchial air ways and identified people who went on study who had bronchial

dysplasia and who had six months of either an inhaled cortico steroid, approved for asthma or a placebo and then a underwent a second whereon coscopy. and they also had helical cts to look at their peripheral lung. can't you see whether they have

any nodules. this was again based on the animal studies and everything i just showed you. so what we found was that actually there was no effect, a very disappointing large effort, phase two study, what you see is that about a third of the people

undergo spontaneous regression of these dysplasias, there was no difference in the progression rate in other words, new areas of displassia, and when we look at the ct detected lung nodules, there was a decrease in the number of nodules. so that gave rise to the second

study and this is the first of its kind where we rooked at people who had ct detected lung nodules. these folks were undergoing ct screening. they were part of a screening trial at the european institutes of oncology in italy and large

pi program at this point. it was asked what happened to the nodules. so the primary end point was nodule size, no histology, because threes are small nodules there and what we found is that in the solid nodules the capacities there was a

difference made people much more likely to regress, have smaller nodules or disappearance of nod ultimatelyooses if they had the inhaled steroid but over all there was no difference. over all there was no difference because of the mar jorrity of nod ultimatelys, that simply

don't change in the size at all. now, we think that it's the nonsolid nodules that are the important ones, it's--chose accumulating data in the literature that a proportion of these nonsolid nodules, you see one right there, this is what it can look like, under the

microscope, a reasonable proportion, quart tore a half or more, are either early bronchial carcinomas, or a typical adenoma hyper plassia which are pree cursors on lung cancer, so the new studies are all going to be looking at only these types of nodules, you didn't think we had

the power to look in the study because we didn't have enough people for these small nodules. so the inflammatory--the inflammation story is not quite over. you know? the two trials that we did were not positive, but they show us

where to go and they may not be positive because indeed, maybe the right hypothesis or the hypothesis is incorrect, it may be that we're simply not getting enough of the inhaled steroid peripherally enough to get it--to get an effect, there are a number of logistical issues

which are the best nodules to go after, but we think they are these ground capacities need to learn more and follow these people long-term and to do studies with perhaps nor effective agents. so is there a more effective agent?

maybe. this is aspirin. we go back to where everything starts. so, this is--this is from a series of studies by pierre rosetta welin the u. k. looking at long-term follow up of people treated with aspirin or not, or

placebo, mainly for cardiovascular reasons. and what you see is that after five years, the two curve display. there's about a 30% decrease in lung cancer deaths, not just incidence, but death, after five years of aspirin treatment.

it's actually only adeno it's not squamous cell carcinoma. but this tells us that again, these were intervention trials. these is not plain hepatitis ey this, is a series of intervention trials so this tells us that if you're dying at

five years your incidence, your diagnosis would be three-four years prior to that. your stage one lung cancer would be four years prior to that. your premalignancy would be even earlier, so this gives us a window for doing studies with aspirin, looking at those ground

capacities and that's exactly where we're going right now. let me tell you a second story, this say drug called my o nostoll-like receptor, okay? it is actually a food constituent. it's been a glucose isorine., it's actually source of several

secondary messengers and signaling molecules moleculesand it's relatively benign, it's something we call grass by fda technology where--that's a dezignation that's regard as safe, you don't even have to do the paperwork for an ind toxic effects use this drug in

studies. based on the work for others, there's a lot of preclinical data for this drug alone or in combination with inhaled steroids in this any case going back to stephen lamb we did a phase one study regression rate of dysplasia after two months of

mononicotol was 91 percent compared to historical control. this is not a controlled study. interestingly enough, there was ologist a slight decrease in blood pressure that was not to abnormal levels but from high normal or just slightly supernormal to more normal

levels. so it's an exciting concept that perhaps this would be a drug that would also have other good efcts. now what stephen did which is quite novel is that he also did bronchial brushings from the normal epithelium so he looked

at all the areas of displassia and then he brushed the normal bronch i, and some to the brushings to sphera at boston university who did gene expression analysis and would all be found looking at these samples and many others is that people with bronchial dysplasia

have upregulation of pithree kinase signaling and the people who responded by reversing their dysplasia in this small cohort, also reversed the pi-three kinase signaling, so now we have a potential model molecularsignature that we can look at both as the people at risk, pithree kinase

activation, and dysplasia, and as a read out,as end end point. those who reverse one reverse the other. that's the hypothesis. so, we now have potentially, a way of moving forward, which is really a new clinical trial model which we will be putting

into place. so, questions, did pi-three kinase activation truly identify the at-risk smoker who is have the dysplasia? it would be easier to get those brushings than to be looking for dysplasia and there's dat awe get similar changes in the

bronch i as you do in the nays nasal epithemual so this is all an evolving area. but it also gives us a new clinical trials model where we look the pathway analysis, preand post treatment and in fact we're looking at that with aspirin, no less.

so, this--there is an ongoing study which is duplicating this but in a much larger number looking at disblasia but doing the pathway analysis from the normal bonkial epithelium and the study has 85 people on it before i give the results of next year.

wog last year to talk about in terms of prevention which is the prevention activator receptor gamma. pkr gamma. another target we're looking at. this is the one in use today in data and work that really started with cell line studies

and then progress to animal studies we were able to show that lung cancer cell lines differentiation invitro, there are animal models i show you an arrow dynamic model where you have much decrease incidence of tumors again and decrease in about a 33% disciplinary crease

in cancer incidence and diabetics that use this agent compared thors those who don't use this. and higher decrease in head and neck cancer. all part of that arrow digestive tract. so, the way we move forward with

this drug was to do a small phase two study and allly we didn't do it in lung cancer we did in oral luke plackia, and it's a precoursor to oral cancer and again it's part of that arrow digestive exposed field. and at this small study, 22 patients, we found that 80% ever

clinical response rate, with a huge decrease in size of these lesions, a little bit hard to see, but there is this whitish plaque on the tongue which is not there after treatment. so the current ongoing study is again the double blind placenta vehicle o control and this is

all head and neck cancer, and preclinical studies, again, looking asterisks mice, and limitations, huge decrease in tumor burden and preventive study setting and we're only and trial and and people with wronkial dysplasia and so in the next couple of years we'll have

data. okay, i'm going to finish up with just a little bit on early detection, because that's actually an area that has ruly blossomed in the last couple of years because i just and this is negatives, and they are that squeaming is not freebie.

it sounds easy, but it's not that's because these various biases that are part of the the time bias where you diagnose the disease earlier but you don't postpone death. you know the person has the disease earlier. length bias, where you're more

likely to diagnose the indolin disease and the rapidly progressive disease that actually kills the person so you're looking only at the better prognosis and outcome disease and of course,uber diagnosis where you find lesions that are clinically unimportant

that would not be diagnosed otherwise and there's now data that actually all of this occurs in lung cancer, too, the bestuber diagnosis and prostate cancer or if you do autopsiess on men who are 70, the majority am have prost indicate cancer and not to say it's not an

important disease, but it's very common. a lot of overdiagnosis. so, early detection has been a big focus of bnci, but others for a long time and chest x-ray was studied, sputum sitology was studied. many years ago, 30, 40, years

ago with no good results, really the definitive test, definitive screening trial was performed, fairly recently, of the plco prostate lung, colo-rectal and avorrian cancer screening trial, 2011 showing once and for all the chest extra does not detect lung cancer effectively.

what does, helical ct, so this is a national lung screening trial which is a very large study, 53,000 smokers current and former, specific criteria, 30 pack here, quit less than 15 years ago, age 55-74, people were randomized to either getting a lealical ct,--helical

ct or spiral ct or chest extraly times and the results were a very large number of positive tests, okay? but a 20% reduction in lung cancer mortality and even a reduction, statistically significant 6.7% reduction in all cause mortality.

so this is the first time that we have the ability to actually prevent lung cancer death, okay? here's just the slides to show it or the graffs to show it. so,--graphs to show it. so very exciting results. how you get that to 90 million curspent former smokers is a

challenge. so how do we move forward. a lot of modeling data. there will be more data coming from from european studies but here is one beautiful example, looking at when you have a nodule, at the first screening ct, and there's plenty of people

who go to lung cancer and this is again from the british columbia group, who look at, and develop a risk calculator, which one you will find at this web site and based on things such as age, sex, family history, emphysema, by history, and then the nodule size type location

there's very accurate way to predict who's going to develop canc netherlands and next three years. and so, this is one potential way to figure out who really needs to be screened repeatedly. who should go into chemo prevention trials and so all

these things are coming together. we are now at a point where we can actually identify people who have early lung cancer, identify those who need subsequent screen fist at a don't have lung cancer, obviously more validation is still needed but

this is a pretty validated study right here. so,--so way to move forward, looking at the various pathways, some of which i described looking at the pi-three kinase et cetera and so i think we really now noised to move forward okay so in summary, a

lot of progress has been made, not just an understanding the model molecular biology of lung cancer but actually in making it both in terms of treatment, and decision medicine as it's typically called with clearly increase the survival in advance

stage patients, but not you're, immunotherapy, with just at the beginning of and that's going to make a huge difference. and fanatic who need tews be screened and how long they need to screened and with these new tools and targets we can now really do chemo prevention

studies moving forward, so that's it, thank you for your attention questions i'm happy to answer. [ applause ] >> [indiscernible] >> lung cancer? it's not. it's actually commonly

disregulated thway. tually a fair number of mutations in the squamous cancers, so head and neck cancer, frequently have pi-three cain ace mutations or p-10 abnormalities, esophageal and so as bronchial dysplasia is a squamous precursor.

but we're into those either. >> so and they are still going, a group is doing that. if you buy that my o nostoll-like receptor is a pi-three kinase inhibitor, actually invitro, it does appear to be that. >> there are a number of groups.

i know there are. there's a group that its pithree carc ntore, so the number is yes. >> thank you. >> okay our next speaker is kurt harris, he's with the laboratory of carcinogenesis, he does clinical training and internal

medicine at ucla, and these receive numerous awards one of which is the nci outstanding mentor award and his talk today is interweaving the spreads of the p53, microrna and the chemistry of cancer and aging. >> that's what i'm going to talk about?

>> the world traveler is here, i think. >> actually what jerry is talking about is i got back last night from china. so it's sometime early in the morning for me. but they came over to keep me awake and yell at me every once

in a while. handouts that have been handing out or over there, outside. all right, let's--this thing works, right? great. i'm going to break this talk up into four parts. first is going to be about aging

and senescence and it's a barrier, tumor suppressor bear queer to cancer. second will be will be a very short presentation about stem cells and why we're just interested in cells and then we'll get into practical stuff about screening and micrornas

and their role in prognosis and and response to therapy and the inflammation in their too, and their disease and combination biomarkers and and micro rnas, and biomarkers and prognosis. so very briefly, senescence is caused by mechanisms, first is telomere shortening that

occurring during replication of cells that's the traditional way. and that leads to dna damage response that can be measured by gamma h-two x and thal case you can see the telomeres and you can see the response at telomere.

in the chromosome. and there's evidence that actually dna repair occurs more slowly at telomeres that does the rest of the genome. it goes through dna damage signaling cascade in primarily p53, leading to sen illegalsen senescensel, there's many

markers of senescence, and it's a blueness that can you see, just a change in the ph, the second kind is dna damage that occur at any placeot chromosome or the dna and this also goes through p53 and the third is onca gene overexpression such as krase model, this goes primarily

through an rb pathway. so these are the people in the lab whoof been studying this and they've been--they've been investigating two different hypothesis. first p53 isoforms which i'll define are physiological regulators of replicated

senescence, both invitro and in all of us sitting here in this room. the second is that these isoforms are deregulated during carcinogen sis, particularly lung carcin o genesis and also in tumor associated t-cells. so about 90% genes have

isoforms, p53 actually has 13 and their function as not been well studied, but two of these have been investigated by us and a few others, one is the p53 beta which has a spliced terminus and the second is delta 133. in which a third atg site is

used as a start site and so the first time in 33 amino acids are missing and it acts as a dominant neg95. this one acts primarily as a co-transactivator of p-53. so initial kind of experiment was to take the traditional fibroblast experiment, grow the

cells, young cells, ones are rapidly growing verses those that stop growing or sin estimate thad spent ask the question, what's the level of expression of these eye o farms, p53 beta appears only in senescent cells and the delta 133, is detectable only on the

young growing cells, these cells already have markers of senescence, activation of p53 and high levels of p21 which is down stream from p53. this is a logical way or the best way in my opinion of asking the question: whether or not a gene has function that's

relevant is to knock is it down or knock it out. we use two different sirnaas here and as can you see there's less delta 133 here and this is using a of antibody and can you barely see it here. when we knock down the delta 133, the dominant negative form

binds p53 and inhibits p53 function and can you see an increase in the p21 and you can see these blue cells that are for more fried egg like and you can quantitate them. so what happen fist you do the orsit experiment and that's overexpress the co transact

vacator form of p53? which enhances p53 function, for the delta 133, the dominant negative so this is a short-term assay just looking at cell growth over expression of beta, compared to the vector control inhibits cell growth and induces senescence as quantitated here

with the sen illegalsen cess ensel--senescence, and overexpression of the delta 133 looks like it enhances growth, has no effect on senescence and then when you do the experiment of daking cells that are near senescence, and only have two or three or four poputions to go

before they become senescept, you overexpress the delta 133 and you get about another 10 or 12 population doublings, well, that's all cell culture trying to do experiments what happens in us. and as we age, i thought about this in terms of what cells

through this experiment with, finally came up with cdeight positive t-cells. and there have been previous work with this population of t-cells in senescence in a sense that there's already some evidence that as people age there's a change in these cells,

and they became what's--people what's called immuno senescent and i'll go very briefly through this data rich slide, and what happens is that cd28 and cell surface co-stimulating protein goes down with increasing age and the lack of this, the cells without it, goes up, one might

examine, telomeres are shorter, in the cda, cd28 minus cells. and they have an increase of senescence associated protein and another sin estimate thadent marker is hp one gamma that's heteroprotein or one gamma and it increases quantitatish is here, and if you isolate these

cells they have an increase in ileight and ilsix in cxcr one and cxcr two. and all of these have been associated with senescence of fibroblast and and ilsix and ileight actually are growth factors for human lung cancer cells, and in fact many lung

cancer prosduce ilsix and ileight as autocrip growth factors. so, so what happens in the with levels of the co transactivator or the dominant negfive form of p53, so the dominant negative form of p53 goes down and co transactivator form of p53 goes

up. in the cells that are immuno sin estimate thadent. we did the same similar kinds of experiments with cultured cells from lymphocytes from--that is lymphocytes from people and they're getting sen illegalsennent markers and they

have dna damage at telomeres. even though, and then lastly, these are western analysis showing that the p53 delta 133 p53 goes down, the cells growing in culture become being senescence and the p53 beta goes up as you can see here. this is a knock down to delta

133, the cells grow and they become a senescent phenotype and you can overexpress p53 beta skels if they do grow. excuse me. so i will get that construct in a moment and let's summarize this and eye will tell you how it relates to cancer.

so p53 isoforms, there are 13 different ones we identify the glycoprotein for the p53 beta, the delta 133s, control primarily by eto havagesy and it goes do you think in the senescence its or fibroblast or lymphocytes in vivo and invitro. p53 beta goes up, in benign

tumors that same thing happens, those tumors, have a lot of sin estimate thadent cells but when they bottom carcinomas, you get the switch in the isoforms and this occurs also in lung cancer, so, the dominant negative form goes up and p53 beta goes down and t-cells associate wide

tumors that can have an enhancing effect as well as a antitumor effect and the enhancing effect occurs primarily at the cells are producing ileight, and ilsix in our sin estimate thadent so as i mentioned whether this is high in humans they have a high

senescent cells and this is from very people's data including our own and in animal honduras model fist you restore p53 at levels that are relatively physiological as oppose to super high levels that you actually don't see in tumors. and frequently occurs by sin

estimate thadent and not by apoptosis. so there are a number of mechanism for in benign cells i listed at least two here that's ppoo beta and they decrease the p53 negative form and increase in p16 in the rb pathway, dna damaging agents and free

radicals such as nitric oxide and telomere shortening and uncapping and two barriers and p53 network and the rb network that are barriers too and can bypass, can you bi-pass the sin estimate thadent barrier by decrease in the p53 beta increasing the delta one flea,

but more effective than that is the p53 mutation, and dna methylation of p16 which usually leads to some decrease in expression will by-pass this mechanism but more effective than that is the p16 mutation. so let's move on to lung cancer stem cell topic and why am i

telling you about this whether stem cells, normal stem cells are important in homeostasis in which the stem cells divide and one of the two partners in the stem cell and the other will differentiate in the lung would be ameek o silliary differentiation but you should

certain circumstances which i'll tell but in a moment. you can get some metric differentiation so a stem cell becomes two differentiated cells and in cells that are regenerating and in some cases cancer, you end up with symmetric cell renewing so that

you end up with two stem cells. and that's we--the hypothesis we're explore nothing tumor genesis, that this cancer related genes particularly cancer driver genes push this option towards symmetric cell renewal. well you can measure asymmetric

cell division and cell division and let's first look at asymmetric semester i vision. these are brdr and you go through two cell replications and then you isolate the anaphase cell asks then what one finds is that budr is in one of the two halves of the cell.

so this is an anaphase cell that will divide right here and the stem cell is here in that will be the difrepresentiated cell so you will actually measure this in cells of culture either, more nontumor genic bronchial epithelial cells or noncell lines or colon cancer cell lines

and cell lines are okay for doing certain manipulations but they're quite far from the original tumors and when we looked at primary lung cancers, and put the cells in culture, we found that the frequency of assim retinal locationic cell division was quite high compare

to cell lines and but with increasing selection in culture, the asymmetric cell division rate went down and we were proposing or or we do propose an increase in symmetric cell renewallal instead of asymmetric. you can measure stemness by

different ways and lung cancer as well as other cancer types, but one cellular marker is cd31-33. and and you can measure it as shown here and here and then when you do the experiment with the budr, you can find budr and the--the stemness marker going

together but what's happening to this particular portion, the cell that would be the progenitor cell from the--from the stem cell. well you could use markers at differentiation such as surfactin marketer, and as you can see here, the prosevere

mitral actin marker is only on the side of the nonstem cell, nonbudr or the pro or the pan cytokeratin is also only on the side of the nonstem cell partner. so the p53 or other agents modify this. p53 comes in many different

flavors, these are all somatic mutations and one can do experiments, taking p53 null cells and overexpressing different mutants and some mutants have been shown to have gain of function and some have just loss of function and so we used a docksin inducible system

that can you see with docks there's expression of these various--of these various mutant forms of p53, and the degree of asymmetric cell division goes down at least with two of these in both cell types here. and if you remove the docks in the mutant protein goes away,

frequency goes up and so that could control saying that it's being modulated why the mute ant p53, so how do you measure the other half of it, and that's the metric self-renewal and so as long as quake field has made a construct called spore six that is activated by octet four and

sox two. these are stemless transcription factors and you can measure that by p a xatox analysis and by fluoresce spent when you isolate cells from cell lines of this case, initially, and isolate cells that are positive for spore six, they form tumor seers

in culture where the ones which are negative rarely do. so we're currently testing the hypothesis that mutant p53 increases self-renewings, and self-renewing--cell--cancer cell division so anarobles and i recently wrote a commentary that appeared a few days ago,

actually on a paper that was measuring reportedly cancer to stem cell effects or the effects of a primate specific microrna on symmetric and asymmetric cell division. so in this cartoon, this is a normal stem cell and as i mentioned earlier, normal stem

cell primarily goes through asymmetric cell division with one cell being the stem cell so it persists in the other partner becomes or the progenitor becomes differentiated because because of so many are differentiation, but under chronic damage or stress, if you

get symmetric differentiation, this can lead to squamous differentiation or under acute damage and express stress, can you get symmetric self-renewal and regeneration for example, basil cell hyper placia. so what is hypothesized is in cancer stem cells, that can

occur through a progenitor cell or much less frequently, from arise progress stem cell itself, is--can consider--occur with self-renewal and assim rhetoric self-renewal and could be objects of targeted therapy and we're propose thanksgiving is driven primarily by cancer

driver genes such as fusion genes in lung cancer. andlet and the mirks row rna thal case, this particular one is involved in terminal, as a side. and so you can make the argument well, why are they primate specific?

well there's even one that appears to be human specific. now what's the evolution of that. what are their particular functions? are there a redundant functions or brand new functions that have evolved.

anyway, so it's another area of micrornas which now, i'll talk about. so, micrornas and cancer interesting story over the last two decades, first of all victor and gary made the inneri shall discovery in studying c.elegans and they found there were small

lung coating rnas, about 26 base pairs that were evolutionary conserved gene expression and so there were basic scientists and their observations actually were not well noticed until about 10 years later in which micro rnas were found in human cells

in tumors they research differentially expressed in cancers and carl and observations at that in chronic lymphocytic leukemia, as and employ and array for using basic research in the clinic and micrornas are evolutionary conserved so can you find

micrornanno, sir plants as well as in humans. so they were oldsmobile recently discovered but very ancient mechanism of epigenic changes, so the rest of the talk will deal primarily micrornas and inflammation, for predicting risk with this prognosis, in

therapeutic outcome. so micro rnas are encoded by genes and paul two, polymerase two, encodes most of them, they form initially a prior micro rna stem loops kind of structure, it's processed in the nucleus in the structure or variety of other proteins in a

complex and rna and please parts of this, so the premicrorna, and then it's exported into the cytoplasm it interacts with a second rnas called dicer which is part of a complex, to cleave this down to 26 base pairs and then, in a are co transactivator third complex, called risk, it

forms a complex message, protein, messagernas. it has too many functions can either repress the translation of the message or cause instability of the message. and that's shown here again, translational regression for message cleavage.

another function is small micro rnas inhibit protein function in this case, rnp function that is important in granue low sight differentiation, the inhibitor actually so this particular microrna inhibits the inhibitor of protein transcription and lastly, micro

rnas can act as liaison gaunts to the receptors including telomere shortening receptors--telomere shortening recoopters, that are innate in humans and seven in mice and activates nf-kappab pathway and you get high levels of ilsix and tnf halfa.

and the third aspects of micro rnas is they can be transforted practices one cell to another and this will be through exostudies of multiple endocrines. so exostudies of multiple endocrines are butted off of cells employ and this is a

normal process that can occur from plants alm the way to humans. and these exostudies of multiple endocrines contain various proteins and micro rnas and in some cases message rnas and then they go out into the body fluid.

and you can isolate them from serum and from plasma. but also from limp and they fuse with other cells and in this case, the immune cells which is a mack o phage and then the co tents of the exostudies of multiple endocrine are released into--exostudies of multiple

endocrine are released into the cytoplasm and can activate the toll receptor and you get a feedback loop. so they're very adaptable moleculeless that many, many different functions. so, we're quite interested in mechanistic biomarkers of cancer

risk, prognose and i guess for risk, you heard the presentation about that, prevention and screening are important and gene environment interaction. prognosis to, treat or not to and most stage one lung cancers which i'll talk about are not

treated other than surgery but 30% of those folks are going to die of recurrent. so, they die of micrometastasis that are not detected. and then there's interesting biology in development of therapeutic targets. so this begins with the study we

did with carlow croche seven years ago now in which we ask a very simple question. which of the micro rnas are highly overexpressed in six major types of healing cancer. breast, colon, lung, pancreas, and stomach at time there were only 280 and carlow had

developed a chip, microarray for measuring these micro rnas and one came up to the top and that was mira 21 and that's what i'll spend most of my time talking about. all six of these, it was highly overexpressed and now miraka 21 is upregulated in many major

cancer tripes and a prognostic biomarker in at least 12 cancer types. so this is low hanging fruit in that we've pursued it. it is an interesting mechanistic data i'll tell but toward the end. the pitcher of nuzumo, who was

in the lab at the impediments time and he test the hypothesis that hike row rnas in associate wide lung diagnose and i guess micro naas are significant between primary lung cancers and the corspannedding noncancerous tissue and the among the different histological

types of lung cancer. and that there was increase miraka 21. miraka 155 and 106 in these lung cancers particularly in stage one lung cancer. again the earliest stage of lung canc they're we tend to focus there were associate wide

diagnosis and prognosis, in the let seven decrease was confirmatory of the study that tog tacka harby did in frank slack in lung cancer in regards to diagnosis as oppose to prognosis. so one important aspect of doing these mechanistic and biomarker

studies is to look at multiple cohorts in different geographical areas. different ethnicity. so we follow that up with examining three different cohorts, one in maryland, one in norway, one in japan. lead to cap lanmeyer plots.

so hundred% survival. 0 percent survival this is the five year point. so high levels of miraka 21. of the median. for prognosis in each of these cohorts. now has this been found in other cohorts, the answer is yes we

discovered this also in lung cancer cohorts in one, two, three, four, different cohorts and excuse me, in three different cohorts, i just mentioned and colon cancer and four different cohorts, and these are these are studies done by other people and almost all

of these have been replicate bide others and so, there are only 10 out of the 12 or 13 that i've shown on this particular slide. so high levels of miraka 21 are associate wide poor prognosis. now can we combine micrornawise inflammation, and influmation as

you've heard as associate wide increased risk cancer can it can be inherent in hemachromeatose estimate thad and iron overload disease with liver cancer, inflammatory bowel disease or familiar pancreatitis but more trently inflammation is acquired either due to viruses or

bacteria parasidic infections and it's been now recognized that chemical physical and metabolic conditions are--lead to inflammation. including acid reflux and barot's esophagus and reintegrated services creasing esophageal cancer asbestos, it's

many things that actually damages dna and it's also an inflammatory agent. obesity is a chronic inflammatory disease and there's--i think asterisks last count 62 chemical carc sin o gens, no chemical carcin so gens found tobacco smoke, but so it's

a witchs brew of carcinogens. but any of you who have ever taken your first puff of a cigarette realize how toxic and how inflammatory it is. it's only after you become addicted to nicotine can you tolerate smoking. now the world cancer report,

this was about version about four years ago, found that 18% of human cancers over two million a year were related to infection not counting these kinds of inflammatory agents and we've just finished--just finished the draft tallly, the proofs of the next

report--actually the proofs of the next report of which i'm one of the many authors and this is a gross underestimate of the amount of cancer caused by the combination of inflammation and infectn. now i mentioned obesity. and these are two seminole

papers that were published 10 years ago in diswrowrnal of clinical investigation and what these authors found was that in fact there's high level or high concentrations of frequency of macrophages and these macropages are activate indeed the inflammatory loops with predip

lo sights and dip lo satellites. so obesity is a chronic inflammatory disease, increasing the risk of cardiovascular disease, neurological disease and cancer. and the macrophages produce the radicals and it can be oxygen base o nitrogen based free

radicals cadicals a that can damage the dna and the prepare is not basic cision repair to get mutations but the free caderals are also damage proteins, and pro teens that are important in dna repair and there's epigenetic effects of free radicals in terms of

incrising dna--increasing dna methylation and also reintegrated services creases lipid proximityidation producing these highly reactive electrofiles that damage dna and again are repaired by basic cision repair and a lipid proximityidation pathway to cox2

and cox one leads to cell proliferation and certainly support a variety of cancers. all right, so this is jane and erin who asked the question about inflammation and in this case colon cancer we've done the same thing with lung cancer, and what they found was that a

combination of primary pro or antiinflammatory cytokines in the nontumor or adjacent area as well, as the cancer, together are associated with an increased risk of in this case, colan cancer, in the adjacent nontumorrous area, high levels of the antiinflammatory il10,

are found and that's also associated with increase and metastasis in lung cancer and colon cancer. but in the tumor these proinflammatory cytokines are at high levels including il23 which enhances proliferation of th17 cells and produces high levels

of il17 which frequently has a pro tumor progression effect. so what happens if you combine in this case inflammatory risk score with miraka 21 expression, is it a better prognostic classifier alone and indeed that's the case. so, these are cap lanmeyer plots

again. hundred percent survival, five year point, high inflamma torey risk score, primarily proinflammatory cytokines, poor prognosis, same patients, same prognosis, high levels of miraka 21, poor prognosis and but both are low, quite good prognosis,

either one is high and intermediate prognose and i guess both are high, they have poor pooinginoseis. it's important to take these kinds of results to statistical analysis, particularly multiby valid and reliableiability analysis which in which you're

asking the question feach co variant independent of the other as you can see here, these are hazard ratios, and the inflammatory risk score is independent of peer 21 expression and they're both independent of tumor stage. so what's the principle here?

well, each have some value and you get curves that primarily look like this and you get a more robust prognostic classifier. but the rational for saying that this is more robust, is that each one of these whether it's noncoding rnas or coding

rnas as the biomarker, this by some patients. but they misclassify different patients. now that's putting two mechanist and i can statistically independent biomarkers together. what happen fist you put a third with these two.

is that better than two. the answer is yes? , what happen fist you put a fourth biomarker that's mechanistically and statistically independent, is that better than three, we don't know, we're study tag right now. --studying that right now.

is it oldsmobile true in this case this colon? no, it's true in esophageal cancer, it's true in adeno carc nome asquamous cell carc nome aesophagus and true in lung cancer which i'll tell but now and then, kyle's imrowp has done the best in these three that

came from our group. so this is a paper that was published about two months ago, and the hypothesis is the combination of protein coding genes are neckanistically related to lung cancer and noncoding miraka 21 is a better prognostic classifier than

either alone. so let's look at these data in two different components. one is a japanese cohort in which it's primarily stage one a that's less than 3-centimeters, metastasis. so the four gene signature, which has increased xpo one,

increased brca one and alpha one alpha and a dereece in a souper suppressor dlc one. so those are--those are a couple of combination are high, then you have poor prognosis, if miraka 21 is high, poor prognosis, you should put the two together poor prognosis.

sewhat happens in a second cohort, a us norway cohort which is primarily one b, that's greater than 3-centimeters. so you can see the prognosis even worse in these primarily stage one b patient and the 4g signature poor prognosis, miraka 21 poor prognosis and they're

both fearful that if either one is high, then the two together worst prognose and i guess again, multivariant analysis was done on these cases. we went to the literature and we found five other cohorts that had the same four gene signature, that had hadn't been

recognized imu once we analyze those, we found indeed that the--four gene signature protein coding signature was associated with poor prognosis and since we wrote the paper, four new cohorts of stage one adeno carcinoma have come up with essentially quite similar or the

same results, and so we're writing a meta-analysis paper on this and some other aspects of and if you look at hazard ratios of all these studies, and these are relapse free survival in cancer specific survivals those give you more confidence when you're doing these kinds of

studies than just overall survival although i must say in lung cancer over all survival is a pretty good measure because once a person gets lung cancer they die quite quickly usually within a few months i know. so stage one lung cancer cases have curiaative surgery, 30%

will reoccur and die of the disease. stage one a, which say little better prognosis than one b, the big difference is the tumor is a bit smaller here and here. but in both cases, surgery is what is utilized but we know that some of these people are at

low risk in developing recurrences and some of are at so how could we identify those individuals because the high risk individuals might benefit from adgent chemo therapy and certainly will be better population for doing the trials. so prognostic biomarkers

particularly those that are mechanistic and use of gene expression is one way of going. now you've heard in the previous talk about the national lung screening trial. which which has been replicated and in fact, there is a 20% reduced mortality.

in their initial study and most of these cancers that were actually cancers were stage one a cancers. and secondly, of the sensitivity of their as assay was quite good of the spiral ct is quite good, the specificity is not so good. in fact, only one out of 10 of

those tumors proved to be cancer as you heard, there are ways of trying to improve that. well one approach that we're taking and we're, grads in the lab and others are to look at proinflammatory cytokines and we've done pspective studies and prognostic studies of

cytokines in serum and have found that ileight and crp are associated with increased risk of lung cancer and in a time of diagnosis, one finds--this is as much as five years prior to diagnosis of cancer, and about the time of diagnosis, you see an increase in ilsix.

both afro americans and european americans and you see, interleukin one beta and il10 increasing. prognosis, you see a variety of different cytokines that are increased lung cancer cases. and some are more related to your being americans and toward

afro americans or toward both are currently writing up a paper now that shows even stage one lung cancer, ilsix and ileight are fairly good predictors of who's going to have recurrents, so, one strategy is to--we have those serum samples, we will soon have the serum samples from

the study with the proper controls to ask the question to test this hypothesis. another approach is unbiased metabalomics, associated markers with risk and diagnosis and therapeutic outcome, and the use of urine as a biofluid goes back quite sometime to pip cognitive

ratties who tasted urine to diagnosis his disease, physicians at that time were rather more bold than we are today and this is just an interesting quote, just an interesting--at least to me, interesting paint tag goes back several sebtrys.

this is a picture of of mia who is working with frank gonzalez as a collaborator, maya if a fellow in the laboratory. this is the stat edgy that we've used for identifying metabolites in cancer patients verses controls to be associated with cancer this is some of her data looking

at four metabolites, three which have been defined this were still--this is a mess, the aspect of the--unknown metabolite, two of these are found in tumors, and this is cap lanmeyer plus. so a hundred% survival. 0 percent survival.

five year, high in all five of them-of four prognosis fits the analysis here so in addition to serum, we have urine samples for the national lung screening trial to ask the question, whether or not metabalomics will help identify those who are at increased risk verses those who

might not be. whrooof the lastly--lastly, miraka 21 has a lot of mechanismanistic data that we and other vs gathered up to bolster import in diagnose and i guess prognosis biomarkers but also as a target to therapy. so gene ampplification, miraka

21 inherent frequently amplified. it can be decreased by dna silencing, and loss of that, leads to higher levels. we show that the egfr and the kraze rase model pathway, lead to activation of p21 and david baltimore and his colleagues

have shown that ilsix and interfiron, activates stat three, transcription factor that activates, miraka 21 with nicole simon and jim and jim mitchell who we published a couple years ago that different kinds of begin o toxic stress lead to increase in miraka 21.

so what's down stream from miraka 21 because i mentioned earlier micro rnas have many targets so those of you who do laboratory studies and you use the antisense and i described some of those, you're very concerned about offtarget effects.

so you do everything possible to prevent any offtarget effects looking at multiple sirnas and so on and on. and mike micro rnas cerebellums volve to have many offtarget effect. then the expression of many protein encoding genes.

not just one, but many. and this--these are some of the genes that miraka 21 targets and i selected these because they've been validated in laboratory studies, and secondly, they're related to cancer in one way or another and this is just a partial list and then the most

recent which i mentioned earlier is that miraka 21 can act as a ligand. so is miraka 21 an onca gene, well, an animal model, the answer is yes, both slacks lab and olson's lab showed in this case is lymphoma and this case is lung cancer, and the slack

lab has shownhlã‘ onca gene addiction, so what is onca gene addiction. so an animal model you'll overexpress an orchga gene. you get a tumor. and then you turn it off. and then the tumor melts away. that's called onca gene

and is there onk o gene addiction for miraka 21. the answer is yes. so is miraka 21, the only onk o gene? the onca gene, that's micro na, miraka one pie pie in similar results including the onk o gene addiction, so, the next question

is, is there a therapeutic target. so antimirakas are being used in the clinic today. the first report was not for cancer, but for epititis c virus. so there was a new england journal report that was

published five months ago showing that an antimiraka that knocked down microrna one, two, three, in the liver is associated with a decrease in hepatitis c replication. and this microrna actually enhances the hepatitis c replication, and previous shown

in monkeys that this was effective and the stage--now there's evidence in hugh mans. a microrna that's a tumor suppressor microrna is in the clinic and stage one hepatocellular carcinoma and also lung metastasis to the liver.

and that study should be reported about two months from now. are there any studies of antimiraka 21 that are in the clinic yet? the answer is not quite yet. but there's a company called reg they's is about ready to--maybe

it's already started a trial of using an antimiraka against which you might imagine is a pretty attractive target for some of the reasons i've talked so let's just preventively summarize this, p53 isoforms for the regulators and cellular senescence, both in cells and

culture and all of us in this room that p53, isoforms deregulated during carcinogenesis and in tumor associated t-cells, and we're studying the hypothesis, now, to show you proliminary data that p53 modulate balance between symmetric and assim retinal

locationic cell division and serum and inflammatory cytokines and lung cancer risk and prognosis including stage one lung cancer that miraka 21 is a mechanistic and oncogenic biomarker many types of cancer and the combination of protein coding and microrna gene

encoding and the classifiers in these cancer types than either alone and then lastly, gene expression and metrics targeted tab o loamic evaluation may approve lung stage cancer and that's hypothesis that we're testing and we'll see what the results will be.

and these are my coworkers, on the left who have been involved in these studies, a number of them have gone to other laboratories established their own laboratories and our collaborators are listed here that i learn a great deal from all these people and i thank

them and i thank you. >> so are you able to alter miraka 21 by regulating [indiscernible] activity? >> that's a very interesting question. we haven't den close experiments at all. carlow croche has.

he has eviden that different kinds of experiments, one is to knock down miraka 21 in the tumor cells that isolate in the exostudies of multiple endocrine in which you have a deficiency of miraka 21 and they don't activate the tollrece. so, those are the kinds of

experiments that have been done. exostudies of multiple endocrines can be measured in serum and plasma. there are a kits can you spend money on that don't work, so don't do that. you have to go through a three sentrification process and it's

pretty--laborious but that does >> okay, well, go get some rest. >> thank you for your patience. >> thank you for your lecture.