American Breast Cancer Society

our first speaker is farah zia she was a clinical fellow, and now she's the medical officer in the division of cancer treatment and diagnosis. and she also sees patients at the ccr., building ten, with the women's malignant branch. heartlet is overview of breast

cancer -- her title is overview of breast cancer. >> thank you, terry. so let's get started. in order to appreciate how much progress we've made in breast cancer research and treatment we need to step back and look at how things were yesterday.

let's see. in 1975, the incidence rate for female breast cancer in the united states was 105 new cases diagnosed for every 100,000 women in the population. the mortality rate was 31 deaths for every hundred thousand women.

from 1975 to 1977, of those diagnosed with breast cancer, about 75% survived their disease, about five years. what you said female african americans, 32% survival rate. in 1975, mastectomy was the only accepted surgical option for the treatment of breast cancer.

at that time, only one randomized trial of mammography trial of breast cancer screening had been completed. several other trials and the joint nih american cancer society breast cancer detection project were just beginning. the clinical investigation of

colombia bin nation -- combination chemotherapy chemotherapy, using multiple drugs and hormonal therapy was in the earliest stages. in the mid 1970s, clinical evaluation of the drug tamoxifen, as a hormonenal treatment of breast cancer was

just beginning. in the 70s there was no gene associated with increased risk so let's look at today. for the years 2009 to 2013, the incidence rate for female breast cancer was 125 new cases so that number has gone up. mortality rate was 21.5 deaths

for every 100,000 woman. that number has come down. the five year survival rate was 92 persist for whites, 81.5% for african americans. although poth numbers have increased, there is a proportional disparity that remains.

for the year 2016, it is estimated that there will be 61,000 cases of in sutu disease which is a non invasive cancer. 246 cases of invasive disease, and 40,450 deaths. all of these statistics are from the sear program at nci dccps. fewer than 1-6 women diagnosed

with breast cancer will die of the disease in the united states u.s. in 2016. so things to note from these statistics. the breast cancer incidence rate in the u.s. has been steadily increasing since the 1970s, as you can see, which can be

attributed to the strong push for screening ammonk women over the age of 40, resulting in better early detection. the breast cancer death rate in the u.s. has been declining steadily since 1989 where when it peaked at a rate of 33 deaths for every 100,000 women.

and the survival rate has been studily increasing. and that is attributed to both getter early detection and better forms of treatment. so let's look at incidence and mortality by race. this is the -- once again, sear data.

you will notice that for white females, the incidence has always been higher than african americans. however, unfortunately, mortality for african americans has been -- has always been higher. and then also going back to the

incidence, you can see that in -- from the 70s to the mid 80s, there is a steep rise in the incidence. then it plateaued. and the mortality, there is now a decline. but again, the disparity exists. so what is breast cancer?

it's cancer that forms in the tissues of the breast, usually the ducts and lobals, occurring in men and women, and although male breast cancer is rare it does occur. this is a structure of the breast. the breast is composed mainly of

fatty tissue which contains a network of lobbies, tickets conduct the gallons, lobules and lobbies. and about 90% of all breast cancers start in the duct or the lobes of the breast. specifically it's about 80% in the duct, and about 10% in the

lobules. so taking a look at risk factors. so precisely what the reasons are, why a woman develops breast cancer are stillion known. we do know, it's a combination of a bone genetic makeup and environmental, life style

factors, hormones seem to have an important role. research as shown a link between estrogen levels and risk of developing breast cancer. looking more specifically at risk factors, age is definitely a very important risk factor. as you can see, about 80% of

breast cancers actually occur in post menopausal women. also, women who have had a history of a prior breast cancer are at a higher risk of developing another new breast cancer. either in the same breast or in the contralateral breast.

also, women who have a history of hyperplasia, which is not a malignant but access of cells lining the duct, or carcinoma in suitu, a cancer ha remains the boundaries of the duct itself. they're at higher risk for developing insaysive cancer. we also know that exposure to

access endogenous hormones in the form of early men ark, late menopause, hormone replacement therapy or no pregnancy, or age greater than 35 at first child, these are all risk factors, causing women to have more exposer to estrogen in lifetime. history of breast biopsy.

a lot of women will get benign lesions. they'll have cysts benign. women who have multi biopsies for these lesions are at higher risk of developing a breast cancer at some time in their lifetime. women who have had -- there is

many case reports. radiation exposure due to radiation for something like hodgkin's lymphoma, there is a lot of case reports of women who develop breast cancer after that. also, if the doctor tells you you have dense breast tissue,

often times that is actually a risk factor we see for breast cancer later in life. also, sometimes the grams are hard to read and things can be missed. life style factors including alcohol, lack of exercise, and obesity can also be risk

obesity, women who are obese have more exposure to estrogen. from the fat you actually make more estrogen. so that is the reason why that is a risk factor. okay. so let's look at family history. that's also very important.

if your mother, sister or daughter has developed breast cancer, before menopause, you are 3 times more likely to develop the disease. also, if it's a close elative, a cousin, aunt, or grandmother, you're at increased risk as well.

so breast cancer has been linked to mutations in specific genes. brca 1, that is related to familial brewers and ovarian cancer. -- breast and ovariance brca 1 is of associated with more aggressive forms of cancer, such as triple negative cancers which are hard to treat.

and more aggressive. brca 23 is associated with male and female breast cancer. the p53, tumor suppresser gene and rb1 also are a risk factor for breast cancer. p53 is one of those mutations, a very aggressive type of breast and women who have both p53

and -- together, their lifetime risk of breast cancer is 83%, unfortunately. then, of course, we have her 2neu, b23, mutations are a risk factor for breast cancer. so october is early detection breast cancer awareness month. i wanted to talk about early

detection. mammograms. so these are an picture of your breasts. they can be use as screening tools in women experiencing no symptoms. they can also be used to detect and diagnose breast cancer in

women with symptoms. it reduces mortality 26% in women angelad 50-74. 17% in women 40-49. guideline it has caused a lot of-- but it has caused a lot of controversy based on different studies. guideline these are the current guidelines that have been put

out by the american cancer society. so for women who are at average rate, age 40-44, women have the choice to begin annual risks and benefits should be considered. the reason for the controversial eruptions, especially the

females in this audience probably have kept up with it, but it's sometimes things are detected that are not really a cancer, and it will put women through a lot of unnecessary tests, and procedures that they may not need. so that is why all of these

recommendations have been considered and reconsidered, and these are the current ones. so from age 45 to have 54, annual mammograms are definitely recommended. for age 55 and older, the acs recommended switching to biannual mammograms, or the

patient has the choice to continue annual schedule based on the risks and benefits and discussion with their physician. so screening should continue as long as a woman is in good health and life expectancy is ten years or more. so women who are said to be

higher than average risk, the acs recommends an annual mri and mammogram. so who are the women at higher risk? these are women who have a lifetime risk of breast cancer of greater than or equal to 20-25% using risk assessment

tools that are available based on family history. and usually this history is taken from either your own physician, or a geneticist if there is a recent one. also women who have a known brca 1 or 2 gene mutation. they're at increased risk.

first degree relative with a brca 1 or 2 gene mutation, but you have not had testing yourself, again, this is a recommendation for you. if you've had radiation to the chest between ages 10-30. so what about mri for early detection?

so while mri is more sensitive, meaning it is more often positive in disease, it also has a higher false positive rate. this may lead to unnecessary biopsies and other procedures. so that is the reason, again, that it, you know -- recommended for some, not others.

the american cancer society recommends the against the use of mri for women who's cancer risk is less than 15%. forwoman who have a moderately increased lifetime risk of breast cancer, 15-20 first are not really sure. there is not enough evidence to

make a recommendation for or against, that's a decision a patient will make with their physician. but keep in mind if an mri is used, it should be used in addition to a mammogram, not in place of a mammogram. so what about clinical breast

exams and self-breast comes? there is no solid clinical trial evidence that a physical breast come done guy healthcare professional or the women themselves provides any clear benefit in early detection. or reduces breast cancer cancer mortality.

due to this lack of evidence, regular clinical breast exams and breast start exams are not part of the guidelines. all women should be familiar with how their breasts look, feel, and report any changes to their physician as soon as possible.

therefore, in order to do this, i think all women need to really know how to do a self-breast exam. so it's basically, first you want to look, then you want to feel. so you look once a month, it's good few stand in front of a

mirror with shoulders straight, arms on hips, with arms over your head and you're looking for things like changes in size, shape, color, dipling, puckering of the presses. the nipple is inverted or any discharge on the nipple. so it's definitely -- a

physician would recommend. just because it's not part of the guidelines, doesn't mean you shouldn't be aware of what's going on with your body. you want to do the same thing, examine your breast . and up right position, but you want to do it

rather looking, you're going to be feel this time. so you want to use a firm smooth touch with your fingers flat and together, in circular motions. you cover the whole breast. doesn't matter how youdite. you want to feel and feel the whole area.

so abnormal signs and symptoms that need to be observed, change in breast size, pain, tenderness, redness, change in nipple position, scaling around the nipples, sore on the breast that does not heal. puckering, dimpling, retraction, nipple discharge, lump or knot,

or retracted nipple. a bioscience is necessary to astain whether a lesion is benign or cancerous, involves remove a sample of breast tissue. there are several methods. the most appropriate method depends on certain

characteristics of the lesion, the sides, location, appearance. so the easiest the fna, done when the lesion is palpable, and you can feel it and stick a fine gauge needle. you withdraw some fluids or some [indiscernible]. if it's fluid, fluid is straw

colored, usually it's benign. if it's bloody, more than likely it's malignant. the other stream, the core need, veryume, abbi, those are done for non palpable breast lesions deeper in the breast. they're all radio graphically assisted and only difference is

that they're using different kinds of the -- the tools are a little bit different. you have an open surgical method done in the o.r. so let's look at the types of breast cancer. a pathologists will review the biopsy.

this slide is showing you the continuum from normal cells to invasive cells. invasive cancer. and depicting a duct but similar thing that happens in the so these are the choices that a pathologist has to choose from. the normal duct is lined with

normal epithelial cells. you can have higher planninga, too many cells but they're still normal. you can have atypical hyper planningia, too many cells lining the duct or the lobules. but if the cells look atypical, not malignant.

then you can have carcinoma in situ, ductal or lobeular. that's where there is too many cells, some off them are malignant. then you can have ductal carcinoma in situ with microinvasion. and as you can see, there are

here -- there micro-- what did i do? microinvasion of small area where the malignant cells are going through the membrane, and then invasive ductal cancer, you have too many cells, malignant, and they are coming out of the basement membrane at multiple

locations going into the stroma of the breast tissue. another thing is inflammatory breast cancer, so it's not completely pathological designation. you have to have a finance that will show invasive cancer. more of a clinical diagnosis.

but malignant cells infiltrate and clog the lymphatic vessels into skin of the breast. dermal lymphatics, the blockage causes red swollen skin. that's the classic sign in inflammatory breast cancer. so guidelines, minimum criteria for diagnosis of inflammatory

breast cancer includes a biopsy from the effected breast showing invasive carcinoma, and the clinical part off the bat is history of rapid on set of he had dema, and peau deorange appearance, the look of an orange people. also abnormal warmth to the

you don't necessarily have to feel a lump. so -- these symptoms should be less than 6 months. and very important, err themia that covers at least a third of the breast. this is a example of you see no specific lump but you

see peau deorange, the our knowledge peel effect that you see -- orange peel effect. these are other pictures. you can see how varied these clinical presentations are. it's for this reason that a lot of times, things like this, sometimes they get mistaken for

infection and mass tights, and get misdiagnosed. women can go for 6 months on antibodies thinking it is own an infection, when it is a breast so i have seen this happen many times in my career. here you can see the arthema that's covering the breast.

let's look at staging. once the diagnosis is established the physician needs to stage the cancer. staging is a way of describing a cancer, such as the size of a tumor, and if or where it has spread. it's one of the most important

tools doctors have to determine a patient's prognosis and determine treatment strategy. stage sero, carcinoma in suitu. it's actually pretty curable. stage 1, you can look at the specifics on your own. but stage, stage 2a, stage 2b, and stage 3a are known as early

stage breast cancer because they involve the breast and the lymph nodes but they have not invaded toify other areas -- to any other areas. stage 3b and c are called locally advanced breast cancer because it is involves the breast,rism nodes you, but it

has spread to the chest wall for both of those. stage have is metastatic breast at this point it has spread to dissites in the body. the more common sites for breast cancer to travel to, the bones, liver, chest wall or the brain. so this depicts wavingly cancer

cells escaping into the lymphatics. so the malignant cells, once they break through the basement membrane and get through the duct are or the lobule wall, they can gain access to the limp system and travel to other parts of the body and metastases.

the lymph node will filter out all sorts of things, foreign particles including cancer cells. a lot of the cancer cells will get stuck in the lymph nodes. travels to other parts of the body. the structure of the breast

lymph nodes, the ones important in breast cancer, we have the super clavicular nodes, and you have your entire ax larry chain, and then you have the media spinal nodes and internal mammary nodes. these are the ones that are important.

so ax larry lymph node dissection a traditional procedure. it involves removing 10-30 lymph nodes in the armpit closest to the tumor. the benefits are that all of the lymph nodes can be examined for also, it's a reliable

determination, whether cancer is spreading t drawback to a full axillary infection, there is a lot of post surgical complications. patients experience lymph he had dema, swelling of the effected arm on the side where the noticeds have been dissected.

because of the lymphedema, they can have infection, nerve damage there surgery, from digging out all the lymph nodes in there inch so -- they look for alternatives. and a lot of research is done on the sentinel noticed, the first node or first few nodes.

if cancer is traveling the lymph system, the sentinel nodes are most likely to contain them. the sentinel node biopsy is a method to determine in axillary nodes contain cancer with fewer based on stage 3 clinical trial data it's -- during surgery, they inject some dye, blue or

99, and they do this near the tumor or under the nipple. the tracer and dye will travel through the lymph node and the sentinel node or nodes are the first that receive the drainage. so the surgeon will either remove that one node or the cluster of two or three.

if the sentinel nodes are clean, chances are that the others are clean as well. how do we manage, once we get the results? what is the management of a sentinel lymph node? the easiest, if you have a negative biopsy.

noticed dissection is not indicated. if their are micromets in 1-3 signal lymph nodes it's unclear and the decision is individualized. it depends on other clinical parameters that the physician is seeing at the time.

if there are greater than 3 path logically involved nodes, the recommendation is to proceed with axillary. holy advanced breast cancer -- locally advanced breast cancer, it's recommended hear. based on phase 3 clinical trial in 2011, or 2012, we know

patients that have early stage breast cancer who have a negative clinical exam, node come, but they do have a positive central node biopsy, however, if their treatment plan includes surgery and radiation, we do know that if full ax larry node is not necessary.

so this is based on phase 3 trial data. so the treatments for breast cancer involve multi modalities, surgery, radiation ichemotherapy, hormone therapy and targeted therapy. clinicians have a lot to think about making the decision.

some of the basic clinical factors they look at, if the patient is preor most menopausal, what stage the disease is at, the histological grade of the tumor, the hormone receptor status, her 2 new, whether it's amplified or not. and there is some histologic

types, actual actually, that have a favorable outcome. these are the listed here. i didn't talk about that earlier, the most common is ductal and lobeular. these are favorable histologies. also consider making a treatment condition, patient's age and

general health. whether or not the patient has known mutations. so just general treatment steps in early stage disease for both dcis and early stage invasive breast cancer, doctors generally recommend surgery to remove the tumor.

to ensure that the entire tumor is rye moved, the surgeon will remove a small area of tissue around the tumor. it's like a wide margin, dissection. although surgery aims to remove all the visible cancer it's known that many times

microscopic cells can be left behind either in the breast cancer or else where. so the next step in the management of early stage breast cancer is to roar the risk of recurrence by trying to eliminate any remaining cancer cells that have been left behind

after the surgery. so this called adgovant treatment, including radiation, chemotherapy, hormone therapy, and or targeted therapy. so it lowers the risk of recurrence, essential does not eliminate it. stage and molecular features

also determine the need for systematic adgovant chemo and the choice of modalities used. so the selection of therapy is most appropriately based upon knowledge of an individual patient's risk of tumor recurrence. balanced against the short and

long term risks of adgovant treatment. chemotherapy is not without its risks. and so we have to think about who needs chemotherapy. patients that don't need it, we don't want to give them the toxic chemotherapies.

so this approach allows clinicians to determine if the gains anticipated from treatment are reasonable for the situation. that they're seeing. the treatment options should be evaluated in terms of both patients and tumor

characteristics. so this is just a chart showing some of the chemotherapies that we use in breast cancer. in the adjuvant setting or early stage setting where we are trying to cure the cancer, the most common ones that we use, the ones that have the most

effect, are the [indiscernible], doxorubicin, dose atoxal, and some of these other ones, they're very effective. and we use a lot in the past. we still do. some of these other ones, we use in the metastatic setting. so the chemotherapy, we know

that when we combine chemos together in the adjuvant setting, trying to cure patients, we -- studies have shown that when we combine and use chemotherapy, it's differently an absolute risk reduction of relapse and in the evidence setting we use

combination treatments. and we see the most benefit in the group that is less than in the age group less than 50. you can see the graph showing a recurrence and mortality. and there is definitely a bigger gap or a benefit in the age less than 50 group as compared to the

patients 50-69. some of the examples of combination regiments that we use in the adjuvant setting. all of these agents can be used in different -- there is so many different dosing and schedules. some are given every 3 weeks. some of them we do dose in

patients every 2 weeks, patients that have more aggressive forms of cancer. we want to keep hitting -- we don't give them too much time to recover. we give it every 2 weeks, sometimes every 3 weeks. so the dosing varyies.

these are the common ones. when patients are hormone receptor positive, we also have the option or we definitely should use adjuvant hormonal therapy after the chemotherapy. some of the things we have these days, so this is tamoxifen. and that is -- you can give that

in preand post mep pausal patients. it is a selective estrogen receptor modulator. it acts as an agtagnist in the breast, partial agonist in the endometrium. women at risk of endometrial small risk.

not large. then there is aromatase inhibitors. this is used in post menopausal it blocks aromatase enzyme that converts other hormones to estrogen, mostly in the adrenal gland. that is where most estrogen is

made in post menopausal another ai is lettro zole. we can also suppress estrogen through using these last two drugs. so this graph basically shows that tamoxifen definitely gives a benefit to women for a and this is five years of.

in hormone receptor positive and for recurrence, you can see that the benefit is mostly seen in the first five years. and for mortality, the benefit is seen later. now we're follow tamoxifen with aromatase inhibitor. there definitely is recurrence

and survival benefits for that as well. it's the red line, tamoxifen alone. the green is tamoxifen plus an ai. this is recurrence. and this is mortality on this side.

so we also have to make choices with targeted therapies. some of the things we have to choose from, there is herceptin used in patients who are her 2 positive. antibody that binds selectively to the her 2 protein. activity is on the outside of

the cell. if indication is adjuvant therapy along with chemotherapy chemotherapy in her2 positive breast cancer, also neo adjuvant therapy in her 2 positive cancer used unmeted static positive breast cancers as well. pertuz man is another example of

a targeted therapy. it's a mono colonial antibody that bines to the extra cellular domain. inhibits ligands. and causes [indiscernible] through the pi3k/akt pathway. and then the tykerb is anointer, another targeted therapy.

the indication is for [indiscernible] and other chemotherapy to treat advanced stage breast cancer that has stopped responding to other medications. more targeted therapies, there is an explosion of drugs in this area.

this is an emptor inhibitor, blocking energy cells need to grow. indication is for hormone receptor positive breast cancer that has stopped responding to arim dex and fe marra. i brans, an inhibitor, given accelerated approval to be used

with fe marra to treat locally advanced burner, hormone receptor positive. her 2 negative and untreated with hormones. then this is a herceptin plus emtansine. it's approved to treat her 2 positive metastatic breast

cancer cancer previously treat. how do clinicians make decision about which patients to treat with chemotherapy and what to choose? so in the past, i guess we basically used clingal indicates to make our decisions. those were the parameters of

tumor size, molecular subtype, pathology, grade wish noticedal stating -- nodal status, age, co-morbidity, and performance statusment we still look at these things. now we're in the genomics era and we use gene expression profiling to predict the

response to particular agents. and we also use gene expression profiling of the primary tumor to predict and treat only patients who are most likely to recur and who will, therefore, benefit from chemotherapy. so the study that started all the interest in breast cancer

genomics was the landmark study published in pnas in 2001. and so basically, they showed that gene expression patterns in breast cancer carcinomas distinguish tumor subpathway -- so they defined 67 subtypes. all -- 6 subtypes. they all had different outlooks.

they identified subtype a, bc, in normal breast cancer life. basal life, and erb b2 positive. they did survival analysis on the subcohort of patients with who were uniformly treated in a perspective study. that showed significantly different outcomes for the

patients that were belonging to the various subgroups. and these kaplan meier curves, showed a subtype appears to have the best prognosis. and the basal has the worst progress -- basal 46 like has the worst prognosis. the study was ground braking,

setting the stage for larger studies where additional breast cancer tumors would be examined and combined with detailed clinical information. this would provide a means for identifying express motifs that represent important clinical phenotypes, such as resistance

to therapies, sensitivity, tumor invasiveness and metastatic potential. the last can caid has witnessed the emergence of genomic profiling with risk of prognostication and predict benefit of adjuvant cheap therapies.

the activity level of the genes effects the behavior of the cancer, and gives information on how likely the cancer is to grow rapidly. these tests can be very helpful in clinical decision making. joke and genetic testing are very different.

genomic testing done on a sample of glowed, sariva and other tissue and can be detect mutations in genes. for example, brca 1 and 2 is genetic testing. so the the study is an assay, a tissue based genomics test, using microarray technology to

analysis a patient's breast tumor biology to predict weather existing cancer has the ability to met considers. this was approved by the fda in 2007. the assay was approved, however, validation studies and different breast cancer cohorts had to be

done subsequently. and there has been a lot of so this 70 gene profile is validated as an indicator for breast cancer prognosis in patients that fulfill the following criteria. stage 1 or 2, invasive carcinoma.

tumor size is less than or equal to five centimeters, lymph node negative. er positive or negative. this is who this assay applies to. so researchers wanted to answer one of the most important clinical questions in breast

cancer management. who is at risk for reoccurrence? this is the question that we've all -- we try to answer in making clinical decisions. which patients need chemotherapy. in the past we were using pathological data and clinical

and now we have genomics. so adjuvant treatment recommendations for chemotherapy chemotherapy are primarily based on this question. so the research hypothesized, breast cancer is a heterogeneous cities. gene expression should be

different in high risk breast cancer tumors that may benefit from chemo verses those low risk and would likely see any benefit from adjuvant chemo. so the -- for the tumor specimens from 78 untreated breast cancer patients with known outcomes were analyzed.

the entire human genomic was interrogated to determine which genes were most predictive of recurrence at five years. the 70 most significant genes predictive were identified, and it's important to note that the mama print interrogates all 7 genomic pathways of the cascade.

genes representing growth, angiogenesis, local invasion, intravasation, survival, extra vasation and adapt shun to the microenvironment and the process starts again. so multiple studies have validated the mama print assay. one of the studies validated

this was the transbig study published in 2006. they found mamma print can stratify patients into a binary risk of classification, high or low, with significant difference in the probability of metastases free survival at ten years. so low risk patients, they have

a 10% chance of recurrence, 10 years, would it any adjuvant high risk patients, they have a 29% chance of recurrence 10 years without any adjuvant so this assay is able to divide patients into these two groups. so there was a study published in 2010 by [indiscernible], that

also demonstrated the predictive value of mama print for adjuvant chemotherapy in early stage so those people who had a high risk, they received endocrine and chemotherapy. and they had a 12% absolute benefit. they had a 88% dissent disease

free survival. for the patients who were low risk, they were treated with endocrine therapy alone. and for the ones that were low risk, they also had an arm that they treated with endocrine indianapolis chemo. and patients treated with

endocrine therapy plus chemo in the low risk group, there was no significant benefit to adding the chemo. why risk patients to toxicity when you can predict which patients will do fine, just as good with chemo and without? so the analysis is another key

study published in 2013 investigating the use of mama print for cheap decision. 97% of the patients have identified as low risk for a disease-free after five years. so that's pretty good 97 perts, mamma print, low range, still disease free after

five years. the study shows it identified 20% more low risk patients as compared to traditional clinical parameters that we used to use. and as you can sigh, even though it was identifying 20% more low risk patients than the way we do it, still, you have 97% disease

free after five years, the low risk mamma print was shown to correlate with a good clinical outcome. the high risk miami result was shown to correlate with a poorer clinical outcome and higher response to chemotherapy important question is what is

the likelihood of distance recurrens in patients with breast cancer who have no involved lymph nodes, and they are hormone receptor positive? these patients prognosis are fully defined by hiso pathological and clinical measures alone.

we're not sure what to do with these patients. so in order to answer that question, this is the oncotype dx assay. it was used to develop a gene assay, they have to validate the assay clinically. so rtpcr method was developed.

250 candidate genes were selected from published literature, todays and experiments based on dna arrays done on fresh frozen tissue. data was analysised from 3 independent clinical studies of breast cancer, involving 447 patients, to test the

relationship between expression of 250 candidate genes and the recurrence of breast cancer. they used the results. 3 studies to select a panel of 167 cancer related genes. and five reference genes. and these are the genes that have the best rtpc performance,

the most robust predictions. and then they designed analogy also based on the levels of -- algorithms wayed on the levels of expression, a score for each tumor sample tested. so they -- to validate the assay, that was the assay, now the assay has to be validated

using clinical data. this is how it's done. so they used paraffin embedded tissue samples from patients who were previously enrolled in the b14 trial, so they're using tumor or tissue from patients who were -- that were from a previous trial, used that to

validate the ability of the 21 gene rt-pcr assay and risk for algorithm to quantify the likelihood of distance recurrence in patients who were node negative, hormone receptor positive, early stage breast cancer who had been periphery treated with at a max fortune.

this was the cohort from the b14 trial. looking at those patients ithey were able to validate the recurrence score. based on that study, they said the queererance score has been validated as being able to quantify the likelihood of

distant recurrence, in this subcore hoard of patients. the patients who were quantified as high risk, according to the oncotype dx assay, they had a higher distant recurrence rate of ten years. the ones who were classified as low risk according to their risk

score, they had a very low recurrence rate at ten years. using data again from the b14. so it correlated very well, and they were able to validate that part of the oncotype dx assay. so this is also -- they were looking at patients from the b14 trial, looking at the benefit of

so the patients from the b14 trial are those that were categorized as given a risk score in the high risk category of greater than 31. you can see they definitely had a great large absolute benefit of chemotherapy. so the -- this one tamoxifen,

the yellow line is tamoxifen alone, the green line is tamoxifen plus chemo. this is years out. and disease free survival is on this side. and so the ones in the lowest category, they had as you can see, very minimal benefit from

adding the chemotherapy. so this way we're using the oncotype dx or able to predict for certain sub-- cohorts of patients who is going to benefit and who is not going to benefit from chemotherapy. so again, this is the b14 data. the people -- the patients

character iced as low risk, tumor samples would characterized as low risk, they had a risk for less than 18, and their rate of recurrence at ten years was 6.8%. it follows perfectly, the intermediate, 14.3. the ones who were categorized

high risk, their rate of distant recurrence was 30.5%. so the risks score provided by the oncotype dx assay provides an accurate estimate by risk category. so this test, oncotype dx a prognostic and predictive test a prognostic test in that it

provides information about how likely or unlikely the breast cancer is to come back. it is also a predictive test in that it predicts the likelihood of benefit from chemotherapy or radiation therapy. this makes this test stand out from the others, including the

ma'am print assay. the validation studies have shown oncotype dx is useful for both purposes, a prognostic indicater and predictive indicater. it is genetic profiling test with the greatest research and endorsed by the national

comprehensive network. there are other huge trials looking at oncotype dx prospectively. it's a huge trial, a big deal. in cancer treatment areas. it's a landmark trial, a major initiative few integrate molecular diagnostic testing

into the clinical decision making. so the primary objectives is to determine whether adjuvant hormonal therapy is not inferior to chemotherapy, with mid range oncodx. we're not sure -- high range we know, the low range, fine, no

what to do with the patients in the middle still a question. so they wanted to create a tissue specimen bank. nci used ango type d assigned -- firm -- initial results were recorded for one cohort of i'll tell you that in a minute. this was basically so the

patients who had stage 1 or 2 hormone receptor positive, node negative, her 2 negative, cancer, they were put into risk categories using the risk score there the oncotype dx. in those patients who had a referring score greater than 25 were given chemotherapy and

hormone therapy. those who had a low risk of less than 10, given hormone therapy the ones in the middle, randomized again we don't know what to do with these patients. they were randomized to hormone therapy alone verses chemotherapy and hormone.

this is what we're anxiously awaiting, still. so the key points, taylor expressway examines whether genes frequently associated with recurrence for women with early stage breast cancer can be used to assign patients to the most effective treatment and spare

the toxicity of chemotherapy for women who would not benefit from it. it was one of the first trials to examine a methodology to personalize cancer treatment. so we only have the results for the low risk cohorts and they're not surprising.

the results for the low risk were reported in september, 2015. for women a reoccurrence score, less than 10. recurrence with office 1% at this provides evidence that other women in the future may effectively use hormonal therapy

alone, if the recurrence score is 10 or less. this supports the clinical validity of the 21 gene assay in identifying patients who may be spared the toxicity of adjuvant chemo. so another prospective analysis -- validation of the

oncotype dx is called the -- the southwest oncology group, a huge study, probablyra x trials. this is rx positive node endocrine responsive breast this opened in 2011, evaluating the use of adrevenuant endocrine therapy with or without specifically the trials will

evaluate women with a recurrence scores from the oncotype dx assay that ra less than oequal to 25, mid range, and they also have 1-3 positive nodes. the previous trials patients were flood negative. this is looking at flood positive. up to 3 nodes.

so the 2 questions, this trial is seeking to answer, whether chemotherapy benefits patients with node positive breast cancers who have low to intermeet idea oncotype dx breast cancer results. this seeks to determine when there is optimal cut point for

these patients above which chemotherapy should be those studies opened right now, targeting enrollment of 4,000 i wanted to mention what we have at the nci. as far as molecular analysis for therapy choice. division of cancer treatment and

designation, we have -- diagnosis, we have the nci match trial, a clinical trial that analyzes tumors to determine whether they contain genetic abnormalities for which a targeted drug exists that. means do they have actionable mutations?

and they're assigning treatments based on the abnormality. so biopsy specimens are sent to one of four laboratories where they're analyzed from more than 4,000 variants across 143 genes. a broad based screening effort is necessary to efficiently identify new patients and assign

appropriate therapies. the trial is designed to add and drop treatments over time. each treatment will be used in a unique arm of the study. trial open for enrollment, august, 2015, with ten treatment arms. actually, they enrolled 500

patients in 2 months. and that was like 5 times the expected enrollment. so they had to pause for an interim analysis in november, 2 months after they opened. they reopened may, 2016, with 24 treatment arms with each arm enrolling 35 patients.

so this is just for you to look at. these are the different arms and the different target mutations. i'm not going to take a look at it, it's in your handout. and then just briefly, the nci impact trial, nci dctd launched this trial in january, 2014,

purpose to assess whether assigning treatmentbacied on specific gene mutations can provide benefit to patients with metastases solid tumors. during the screening process samples of tumor from patients with various cancers would be genetically sequenced to look

for a total of 391 different mutations in 20 genes. nci impact trial does, the design, mutations of interest, these patients are enrolled in the trial. they're randomly assigned to either arm that has targeted therapy or other arm that has

non targeted therapy. if there is disease progression on the non targeted arm they can cross over to the targeted arm. so what we don't know right now is whether this -- using this approach is really effective, providing clinical benefit. because clearly, most tumors

have multiple mutations. and it is often not clear which one to target to achieve maximal benefits. once a gene is mutated it can lead to multiple pathways resulting in disease pro investigation, requiring several interventions.

this trial is designed to determine whether people with specific mutations will benefit from a specifically chosen targeted intervention, and if these interventions lead to a better outcome. this is currently ongoing. so just a few slides to end.

so what are we going to do in the future? we will use our rapidly increasing knowledge in the fields of cancer genomics, and cell biology to develop more effective, restoxic treatments for breast cancer, continue to improve our ability to identify

cancers that are more likely to recur inch we will also continue to use and improve this knowledge to tailor breast cancer therapy to the individual patient. this is personalized therapy. gene expression analysis has led to the identification of five

subtypes of breast cancer that has distinct biological features, clinical outcomes, and response to chemotherapy. this knowledge can be exploited in the continued development of treatment strategies based on the specific characteristic of each tumor.

further, in patients response to chemotherapy is also influenced not only by the genetic characteristics but also by inherited variations in genes that effect the body's ability to absorb, metaps will, interact with and eliminate drugs. our growing knowledge should

enable prediction of tumor responses to individual chemotherapy drugs or classes of drugs as well as the likelihood of severe adverse events from them. for example, if a exhibit tro genreceptor is mutated we know patients won't respond to

hormonal therapy. if we have a genomic analysis that showed a mutated receptor, that might indicate moving to chemotherapy earlier, and not giving a patient a drug that is not going to work for them. >> so we will use our increasing knowledge of the immune system

to enhance the body's ability to recognize and destroy cancer cells, breast cancer one of the major cancer types for which new immune based cancer treatments are in development. so again, the breast cancer of tomorrow, immunotherapy for breast cancer, we have

therapeutic vaccine, ongoing for sometime. now we have check point inhibitors and immune modulators, target molecules that serve as checks and balances in the regulation of the immune responses. so they kind of wake up the

immune system. so it's activating the immune system already there, to kill the cancer cells. so they do this by blocking inhibitory molecules or activating stimulatory molecules. preexisting immune responses are

enhanced. this is a list of the immune check point inhibiters in trials, not only for breast cancer but other cancers as but these are the ones that we're looking at in breast and we do a lot of this in the malignantcy branch. so we look forward to using

advanced technologies, including jokes to improve ability to -- genomics to improve ability to detect breast cancer when it's most treatable. we will strive to understand, address and eliminate factors that contribute to higher momortality experienced by

african american females compared with females of other racial ethic groups. that's it. thank you. [applause] [inaudible] >> we've had good responses, actually. so our trials are currently --

if they're early, and so we don't have a lot of data yet, a lot of the trials just opened about a year back. but what we're seeing so far is promising. definitely. >> [inaudible] >> thank you.

we have one announcement, that being on thursday, we do the core visit to the small molecule. robotics that they use. for those who signed up i'll be sending you instructions later on this week. and it's going to occur

thursday, 2:00. our next speaker, sid rye. she got her ph.d. in the college dublin, accepted into the nci cancer fellowship prevention program. and she did her post-doc training with [indiscernible], and now she's the nci earl

stephen senior track investigator. she is going to talk about health disparities. perfect, thank you. thank you very much. so as mentioned i'm an investigator here at nci. and a good portion of my

research goes to study cancer health disparities, specifically lung cancer. for that reason, as we go through this presentation, a few of the examples that i give will lean towardlon but a lot of what i'm going to talk about today and the premise behind it are

many different cancer types. if you're wondering why lung cancer is peppered throughout this investigation, that's my laboratory's bias. so the presentation has 3 names. very -- we'll go try the background. we'll talk about disparities in

cancer incidence, also in cancer outcomes. specifically we're going to drill down into the main causes of these disparities. many know, to understand and to overcome disparities, you need to recognize what they are and they also try to figure out what

causes them. before i go forward, i want to mention throughout this presentation today i'm going to talk a bit about race. and i use that word quite a bit. i want to outline to a degree the difference between race and ethnicity.

many of you in the room may know ha this is. i wanted to go through this. when people normally talk about race, they talk about biological differences between groups that are assumed to have a different bio graphical makeup. the degree of ethnicity might

also incorporate biology but also other things. multidimensional, a construct reflect not just biology, but geographic origins, shired customs, beliefs, and traditions that may not have a genetic apsectry at the common core. so it's been known for quite a

while, this graph goes from 1975 until 2005. has been known that there ra racial differences and life expectancy in the united states. this is one example here. black men in gem in the united states, there is 6.5 years shorter.

black women live approximately five years shorter compared with white women. that's quite a big significant difference. contributing factors to these differences, cardiovascular disease without question, one of the biggest components.

if you look at the second part of the pie chart, you can see that cancer is another major contributor to these differences in life expectancy in the united states. so the nci defines cancer health disparities as differences in the incidence, prevalence,

mortality, and burden of cancer and related adverse health outcomes that exist among specific populations within the ups. and in addition we know that african americans of the highest death rates in all cancer sites combined and for malignants of

the lung, breast, prostate and cervix of all racial groups. this is quite a significant public health problem. one example oflung cancer, a lot of the presentation will be on differences between alabama verses white but they are just two of the ethic groups between

the united states and there are many more. five of the main groups include black populations, white, hispanic, asian pacific islands or alaska native. you can see for lung cancer, going back, even further, back as far as 1972 you can still see

thee trends. even though the overall prevalence of lung cancer is declining, proposerally, african americans still have the higher incidence of lung cancer. it's not just the overall prevalence and incidence of cancer between african americans

and european african americans. we also see for many type tips of cancers, the age at which the ability to designate is earlier. not just that getting more of this -- different types of cancer, it's also occurring earlier. this is a recent paper published

by a group in dcg here at nci which outlines many of the different cancer types for which african americans have an earlier age of diagnosis have you might ask what is the driver behind that? several possibilities. one is that the cancer itself

might be diagnosed at a younger age in -- for example, examples are [indiscernible] soft tissue sarcomas and lung. why would this be. one of the possibilities, etiological heterogeneity. it could be in different populations the causesful cancer

could be different. or they could cause the cancer at different ages. another example could be the exposure associated with the cancer, let's take tobacco as an example. associated with lung cancer. perhaps the exposure, or the

timing differs between the two populations. that's definitely a possibility. also be differences in the timing and prevalence and frequency of early cancer nci has recognized the importance of the specific question so why is lung cancer,

why can breast cancer, colon cancer, why are all these cancers diagnosed earlier in african americans, so much that they have launched the nci early on set malignancy initiative. the collaboration with dcp, and they have invited 12 different [indiscernible] to participate

with the goal of really trying to understand why are these populations being diagnosed earlier? so hopefully in the coming years we'll get more information from it mentioned that not only is incidence higher in the united mortality rates are higher.

this is another example here of you can see that for those men and also for women, the incidence are these -- mortality rates are higher in this population. as we go through the talk rater on we'll try to drill down into what's the potential causes and

reasons for this could be. as i mentioned a moment ago, african americans have some of the highest rates of cancer specific mortality. and just going to show you, this is a nice paper from carp about a year or two ago. you can see here, for african

american verses european americans, the prostate, cancer, lung and colin have a much higher likelihood to die from the disease compared to european you can see the opposite trends for asian populations, much less significant, less likely to die from their disease.

for hispanic patients it's more mixed. it's interesting to know, the recent paper which tried to track these differences over time to ask the question are things better, worse, or staying the same? and you can see from this graph

here, this is the likelihood of dying from prostate cancer across different racial groups over and over time. and you can see that overall, things are getting better. so populations are being treated better, diagnosed better. we have more drugs.

so in general survival improving. disparities are not. even though there have been advances in cancer care and treatment, disparities and outcome exist. so this is a construct, put to try to drill down into multi

layered pact factors that -- trying to understand in terms of incidence and outcomes what can be done to try to understand themself, to try to overcome so i'm going to break it down into two pieces. first of all, we'll talk without the differences that exist

within incidence, potential, causes, what do we know, not know. also moving to think about mortality and survival, and various factors there. the first thing is to say we don't know all the answers. second thing, within an hour it

really isn't possible to go through every study. so what i tried to do is pull out some common themes about the things that we know and try and work within that context. so in terms of incidence, the first thing we'll talk about is geography.

so does it matter? this is a map of the united and it shows the incidence of all types of cancer combined. this is data there the cdc. the first thing that you can see immediately is that the -- there is heterogeneity in across the united states.

the highest rates of cancer are found in states like kentucky, rhode island, delaware, louisiana, new jersey. they're shown here in the dark blue. some of the lowest rates are found in new mexico, arizona, wyoming, alaska around virginia.

we're not talking about small differences either. if you look at the lighter color we're talking without between 371 to 431 cases per 100,000. compared to highest, as far as 515. these are significant differences in the prevalence of

cancer across these different so geography does matter. again, taking lung cancer as an this is again state by state, and taking one example of this one state. you can see that for white populations, there is quite a low prevalence of lung cancer.

very high prevalence for african the reason is not known. and the point is to make that -- whether there are differences across the country, the you look at every cancer and every if you look in a more refined way, you can also see differences by race.

so what could contribute to these differences? in terms of geography, it could be differences in smoking prevalence. tobacco is an example. i'm sure lung cancer. many are thinking, well, maybe the smoking rate is different in

each of these states. of course that's very true, they do. and you might have noticed on the last virus one of the highest -- kentucky is one of the highest rates of lung cancer in -- or cancer across the u.s. that state has one of the

highest smoking rates also. there are links that you can kernel make. certainly make. it's possible this are other things that contribute. such a residential migration, racism, advertising, cultural differences, community structure

and social stress. these could all contribute. in addition to -- it shows an example of lung cancer. there are other examples, examples of prostate cancer. you can see the same heterogeneity across the country in terms of incidence.

more about why geography could be important. on a global scale when we think about it, it's very possible that a low socioeconomic status, that these neighborhoods that have a low sds could have additional incidence or sometimes we have studies done,

there is an uequal burden of pollution within certain neighborhoods. we know areas with the highest percentage of african americans and residents have the highest exposure to cancer associated pollutants. secondhand smoke is also an

additional cause of cancer. disparities and exposure to secondhand smoke could be a contributing factor. the question has been asked as to whether or not a neighborhood effect early in life could effect health outcomes. this is similar to the barker

hypothesis, the idea that in neutral exposures -- utero expors exeffect health later in life in terms of heart disease and diabetes. this isn't entirely known m but there is the idea that things like epigenetics could play a role.

there is plenty of evidence to show that geography is important, but we really don't know yet exactly the very fine details of why. these are just suggestions and outlines of things that could make a difference. a lot of people think about

geography, one of the things things that come to mind, the idea of the rural verses urban some studies have been done to figure out if there is something about rural resident which is more likely to provide to disparities in incidences and in a study we just did with the

cdc, where you looked at disparities in lung cancer disparities, we asked this question. we know there are disparities. are they confined to a specific location? we sheathy weren't. we saw whether it's urban,

rural, suburban, it didn't matter. disparities existed everywhere. what we did see is that as you move from a urban location to a more rural location, those disparities increased. what could it be about a rural population?

drives these disparities? there are studies that have tried to ask this question. some of the cases they find that populations that live in rural populations, more likely to forego medical care, report poor health and health related unemployment, and also to

experience psycho social distress. and, of course, there were other unknown factors that we don't but in summary, does seem geography in terms of incidence and mortality as we'll discuss later, is important. other possibility that could

contribute to disparities and cancer incidence within the united states includes kinetic accessibility. what do i mean? one of the best examples we have for this is prostate cancer. prostate cancer is one of the diseases greater, very strong

difference in incidence when you compare european and african americans, so much so that african americans have over two-fold greater incidence than european americans. what drives this? well, one of the things that seems to be drive it, region of

the genome called 8q24. there seems to be variance that -- only found in populations of african decent. in addition to that, there are certain regions only found or certain losigh are only found in the differences in these regions between european americans and

african americans actually explains about 50% of the increased incidence we see in also another region on 17q21, about ten percent of prostate cancer disparities. there is other evidence, breast cancer is amplified, p15. certainly doesn't seem too

extreme, most of the differences as we see in prostate cancer. and then there is also a paper from a couple years ago which looked at recombination hot spots across the genome, asked whether or not there was any inherited loci that might contribute to increased the

combin nation. the combination is a specific event in cancer. the question they were trying to ask is whether or not there was a genetic driver of this that could be associated with risk. while the paper found the gene, this prdm9 gene associated with

high risk as a combination, it's unclear whether or not this specific region and the events that take place, the consequence of it, associated with an increased incidence of cancer. because, of course, this as i mentioned, this region is specifically enriched in african

americans but that next step of linking these specific regions with cancer health disparities has yet to be done. so i wanted i want get intothis. tobacco is one of the main etiological agents associated. not just lung cancer by many cancers.

i wanted to address it a little bit, if i may. this is a data that's adapted from a 2006 paper by chris hymen, and it was done in the multi ethic cohort study. they were trying to look at the relationship between tobacco and lung cancer incidence.

because there exists something of a smoking paradox in relation to this. this shows the lung cancer incidence rate per 100,000 of the population for men and women in the united states. what you can see clearly is that the incidence in both men and

women varies across each of the various ethic groups. in both men and women. you know and it's undisputed that tobacco is a very significant cause of lung so when you look at the relationship between smoking and lung cancer in each of these

groups, and this presence of smoking in these groups you might expect to see a near relationship. you don't. quite striking, looking at high intensity smoking presence, these were the highest rates. african americans have the

highest incidences of lung cancer, they have amongst the lowest prevalence of high intensity smoking. so that doesn't seem to make sense. what else is going on? this is one of the questions my lab tries to study and

understand in terms of lung cancer health disparities. so what we do know and this is a little bit too earlier, when we talked about the timing of the polks much. this is one of those examples. we know african americans tend to start smoking later in life.

so the average on set of smoking is 17, almost and a half years for black, age 14.7 for whites. that is quite different. we know that african americans tend to smoke fewer cigarettes per day. the prevalence seems to be similar, for when they're

younger it's lower. and even for other smokers, disparities persist even among [indiscernible] smokers. tobacco is not explain all the disparities that we see in lung many might say, well, maybe the type of cigarette is different. now, it is.

because we know that about 70% of african americans tend to favor smoking menthol, compared to 20-30% of european americans. you may ask it's not how much, maybe it's the type of cigarette. and when you look at that, even though there is this difference,

it also doesn't seem to account for the difference in the incidence. so there is yet an unknown perhaps smoke related factor that has yesterday to be discovered that can explain the smoking paradox. and this graphic is also -- this

image is taken from that paper. and again quite nicely illustrates the paradox in terms of smoking and lung cancer in these populations. the top panel here is looking just at men. and it's showing that if you take the reference of the light

blue line here for white populations and dark blue for african americans, at the very row intake of ten cigarettes a day is a very significant disparity in men and women. so for some reason, what this suggests is at a lower consumption of tobacco, about

ten or less a day, african americans after higher incidence of lung cancer for reasons we just don't understand. as i mentioned this is some of the research my lab is trying to figure out. one of the other things before we go on to mortality, there are

many different -- think of aspirin. this may be something you've come across already in your presentations. use of aspirin is associated with cancer prevention. now, another similar type of -- not similar, but another type of

medication that is somehow being [indiscernible] in terms of prevention is metformin. and that's one of the knows prescribed medications in the united states today. and it was interesting because in recent studies, of veterinary-- veteran's association records, especially

you see me, finds that metaphorman is associated with reduced risk of prostate cancer. the association was only in hypothesis men. i want to highlight that here. that's what's highlighted here. you can see within african americans this is a non

significant reduction. and non hispanic whites, same thing. a non significant reduction. and when use of metformin was combined with an [indiscernible], it was even a greater reduction in incidence of prostate cancer.

so what this says is even for these other medications, that may be repurposed for reducing cancer incidence, there seem to be differences by racial and ethic group. and that may be related to biology and how differences in biology might be manifest in

cancer is from different and again, this idea of different biology that we're going to touch on in a few slides. going back to this framework again, those [indiscernible] more in terms of in mortality, but a lot of the things in a we

think about potential causes, they are most [indiscernible]. so we talked a little bit about tobacco. that's an exposure. we talked about differences in geography, in genetic susceptibility. there is other nutrition,

activity, infections, things we didn't have time at a talk about today. so you have to think about multi different factorsers multi different contributing factors, how they may work together. [indiscernible] and that will become clearer when we talk

about mortality and survival, because their access to care is very important and we use the examples of that. one of the -- probably one of the main reasons that gives rise to many differences in outcomes in populations, even in the united states, the lack of early

so every year many of you will be familiar with the paper that comes out from the american cancer society where they talk about different rates of incidences are much higher across states in the united states and they give all the they publish an updated version

pretty much every year. every year it pretty much shows the same thing, for most of the major lethal cancers, the state at which that cancer is diagnosed in that population is later than early populations. and as many of you know, and this, again, the earlier a

disease is diagnosed, the better the outcome. so very strong correlation between diagnosis and prolonged survival. what we see in many, many different types of cancer is that these tumors are normally diagnosed at a later stage.

so if you're trying to think about what the drivers are for differences in mortality, this is certainly one of the so when we think about why cancers are diagnosed at later stage, what are the factors that contribute to that. one of them is access to

screening. because one of the best ways to diagnose a cancer early to avail screening services. that is to say not every cancer type has a proven efficacious screening and modality in place. there are some such as breast cancer, cervix cancer.

so is access to screening a barrier in terms of contributing to health disparities? the evidence would suggest, this is possible for some cancers, breast cancer, mammography screening, use is similar to equal access care setting. cololecting cancer screening is

roar among african americans, and also, a lot of studies showing that uptake of screening for cancers, such as hpv, may also be lore in minority and this is a field that i don't work in, so i'm not going to -- just giving this one slide. there is a whole very important

bullet of work that studies and evaluates the screening, even in equal access to care. outside of that, what type of interventions can be put in place to encourage the uptake of access to screening modalities. that is very important. what i mentioned before, while

it is possible that access to cancer screening does contribute to some of these disparities and outcome comes, it's important to mention there are other cancers where no valid dative screening exists. so while i said to some degree a contributing factor that is to

say it is thought the only factor that contributes to disparities. so is access to care important? this slide and the next one that comes after is a study from about 3 years ago now, which to answer the question, would suggest that yes, access to care

is very important. so in this teal color shows populations that have private medical insurance. the blue line has mid caid and the brown orange line are those uninsured. this picture here is showing is the outcomes of patients with

various different types of cancers shown here, non hodgkin's lymphoma, prostate cancer, europeanry bladder cancer, according to the type of insurance they had, and theism publication is the more insurance that you have or the better insurance that you have,

a surrogate for your access to care. what you're seeing regardless of cancer type onth this slide and this slide, those that have private healthcare insurance have better outcomes. this is a study of 2012, the example is lung cancer.

it's important to make the point, what i'm going to show you here, many types of cancer. this was done with a military and they all had the same access to care. they all had the same doctors, pretty much the same access to that healthcare.

when people studies the survival of those patients, asked the question are there differences in racial outcomes, they found there wasn't. this was regardless of histology type. this suggests that access to care is very important.

that if you develop an environment where populations have equal access to care, you can get rid of a lot of health disparities and outcomes. this is an example, as i said, from lung cancer. there are many other examples of this in the literature.

this is an example of areason alcell carcinoma. it's not a carcinoma that i'm particular with, but the point is to make that this is from a military health system, and the graph here would show there may be -- african americans may have a slightly better outcome

compared to european americans. making the point that in the paper, once other things were -- race -- prestige, gender and stage of disease, those differences went away. and they say in the paper, the last of racial difference and survival in this particular

healthcare system, equal access important message was improving access to care is probably one of the single most things we can do in terms of reducing cancer health disparities in terms of a lot of what i've talked about has focused on how african americans of a higher incidence

of cancer verses why or how this particular population has a worse outcome. but it's important to mention that there are some examples where african american patients have a better outcome. and this is actually one of multiple myeloma, there is

increased incidence among adverse disparities and outcome are not observed. this is an example where you can see the black population in blue here, you can see that they have-- this population has a and the reasons for this are not understood. it has been suggested that

perhaps the type of disease or the foreign of multi myelomas that developed in african americans is more [indiscernible] but that's just a hypothetical. we don't know the answer to so i told you about access to what about the uptake of care.

is that different across populations? could that make an difference? this is an example from a slide that [indiscernible] made in presentation last week. i saw it and was quite struck by this may drill down to why geography is important.

so what this graph shows the number of oncologists per 100,000 residents. so 0 is shown here in the light yellow color. and more than 7 or 8 is shown in this brown color. so you can see that without question, across the united

states, depending on where you live your access to care in terms of an oncologist can vary dracula. why is that -- dramatically. why is that important? one of the things to -- we're talking to this slide about uptake of care.

and again, just like in terms of outcome in general, there are examples where access to care shows that it is very important. so in this example here, colon cancer, the access to care and the uptake of care amongst the population, there was no so if you look here, whether

you're african american, european american, there was no significant difference. you were likely to have surgery or as to whether or not you would have [indiscernible]. but it's important to say there are examples where there are differences.

and lung cancer is one of those. so even if you look at medicare patients, this is an example from the american lung association's report a number of years ago. they show that african americans even in equal access to care, so every patient had medicaid,

sorry, medicare. the patients who were african americans were less likely to reef treatment for lunch, in terms of surgery, also in terms of immune therapy. other studies have shown that african americans are renal cancer are less likely to

receive a surgical treatment and more often to die of competing causes. this is a paper or data in -- from 2013, which shows in a regular medical setting, this means one that you or i might visit on a daily or regular basis.

it's not a military hospital, -- it has shown that regular medical setting, that there are racial disparities in specialist consultation as well as subsequent treatment. and this is an example of metastatic colon cancer. if you walk into a regular

hospital in this example, and you're african american, you're less likely to receive surgery or chemotherapy. and you're also less likely to use radio therapy. in terms of treatment and access to medical professionals this is also different.

so access to end of care seems to be important in terms of cancer health disparities than outcomes and addressing them. so what are the factors that influence, potentially influence uptake of care? this is a list of several examples.

it is not exhaustive. as i mentioned before, there is a whole another very important set of work, some of which is being conducted guy nci investigators. try to understand, in addition to that, trying to develop interventions to overcome them.

sometimes understanding the values that exist to medications, to avail of those various treatments can be very important in contributing to reduceing health disparities. we talked a little bit about tobacco, in terms of cancer what many may not know is

tobacco is also adversely associated with patient's this is true for lung cancer you might expect but also several other cancer types. so that is why smoking cessation remains one of the most important public health messages that is out there, because it's

so important in terms of reducing cancer incidence, but also in terms of reducing -- or improving cancer outcomes. we know, many studies have been conducted for quite a few years now, but african americans remain at high risk for continued smoking.

thus, it is very possible this is something that contributes to poor outcomes. why would this be? why are african americans at a ryer risk for continued smoking. >> we know that racial differences exist. so what could they be?

we know some of the factors that contribute to continued smoking, poverty, version secondhand smoke exposure. we also know that although the majority of black smokers express a desire to quit, they're less likely to receive and use evidence wayed

treatments. black smokes are hess likely to enroll in smoking cessation trials. and we talked earlier about this-- factor of menthol cigarette smoking. that is different between european iamerican, and african american smokers.

and there is a hypothesis, and some evidence for and against the fact that smoking menthol cigarettes makes it harder to quit smoking moch these are all factors that could contribute to african americans being at higher risk for continued smoking.

that message of smoking cessation applies to every population, of course. but again, it may be a potential to targeted specific populations based on evidence. so we talked about factors that contribute to racial differences and outcomes.

we talked about access to screening, to care, uptake of care and talked about smoking as but there are some cancers where even in an equal access to care setting, disparities in outcomes of survival persist. one of the best examples is this is a study shown, an old

graph, but the point is the same. that is, even with an equal access to care setting, african americans with breast cancer have a poorer outcome. they have more aggressive disease. interesting data from a clinical

trial, from clinical trials published in 2009, again, this is an example of equal access to care setting. every patient in the clinical trial by design of the trial gets the same treatment. and they looked at the outcomes of these patients across the

study. what they found, as i mentioned, for breast cancer, african americans have the worst and perhaps maybe for ovarian cancer as well. for a lot of other cancers there was no difference. breast cancer without question

was one of the more -- biologically, more aggressive prostate cancer, at the end of this study, stood out. it was important to mention some of the evidence is a little bit dichotomous. there is some evidence that says offs are the same.

references in these papers here. but for prostate cancer there is also evidence suggesting that prostate cancer diagnosed in african americans is likely to be more aggressive. this is a very, very intense area of research at the moment. a lot of investigators trying to

figure out is it really more aggressive and if so, what are the biological processes? and the same has been done in breast cancer with the reasoning being, if you can define this, it might in itself lend to the design of strategies to try to specifically target that

aggressiveness. so i mentioned while data suggests and we have gone through a degree in the talk earlier, access to quality care is very important in trying to address cancer health there is other evidence to suggest that biology, genetics,

might also play a role in terms of cancer health disparities. that's very important. if there are differences in the biology, it makes perfect sense logically, may also be differences in response to therapy. i -- i didn't see all of it, one

thing that i'm hoping will come from these trials that have been conducted on oncotype, and -- phenomenal advances in terms off precision medicine, do they predict outcome? equally in the hispanic populations, european, african americans populations?

hopefully they do and maybe do the job even better. we don't know the answer to that question yet. so sorry if i stole someone's question for later. something we think about a lot when we're designing these predictive and prognostic tools.

you mimust also think about the usefulness in different and that's something in our life as well that we certainly take very severals lee and think about and ingreat any into our biology is important. why is it important? what is the evidence that it's

important? when we think about biology, we can define it a few different ways. we think about genetics, sow math mutations, molecular subtype, systems biology, inflammation, cell biology. so is there evidence that any of

these aspects of biology contribute to cancer health disparities? and if you read the slides you know the answer, already, the answer is yes. i'm going to take you to some of the evidence for that. can we -- 8q24.

we mentioned this in terms of disparities in prostate cancer -- more common in african it seems that this particular region is associated with higher grade and more aggressive prostate cancers. so we're inking this region not just to higher incidences of

prostate cancer in african americans but higher incidences of a more aggressive type of prostate cancer. also evidence of a faster disease progression among african americans in this particular link, this particular region as well.

so it seems like genetics in terms of inherited susceptibility does seem to be also evidence for events between different populations. just one example in african american prostate tumors. it's the ighg3 region. the question is currently asked

whether or not this offense to higher prevalence of prostate cancer in african american men. this is not a somatic event. this is a germ line event. there is other studies done where they're comparing somatic, as an example on the right-hand side, in prostate cancer tumors

and lung. other people have done it in breast cancer ibetween even just by an eyeball look at the particular graph you can see that across the whole genome, the 207 line here is -- top line is african americans. the bottom here is caucasian,

you can see that there are overlapping commonalities between the two. they're also very focused specific differences and these differences most likely define different biology and if we're lucky, to specific targeted so there are differences in

genetics. many of those are very familiar with somatic genetics. mutations. are there differences in mutations across different types of tumors is this this is one example of prostate cancer. it's actually one of the best

examples that are there. so anybody that works on prostate cancer has probably heard of the [indiscernible], very important fusion gene. what's interesting, it's one of those genetic events, somatic events that really does seem to differ by genetic ancestry and

joe graphical ancestry as well. we really don't know why. probably related to etiology as it's quite prominent. if you look here, this is the gene, 43% in european american mutations, only 25 in african it's actually more rare among asian populations.

quite a striking difference. but such a very prominent driver fusion gene that we see very strong geographic and racial don't really know why. there are other examples. last is common among european and even european american quite rare also among asian

populations and african again, there are things we don't important in terms of etiology. this is on a very strong relationship between the mutation that you find in a tumor and etiological exposures. of course in the world of precision medicine it's

important to know what the mutations within a tumor are, regardless of race. depending on what mutation you may have, there could be targeted therapies available to you. you should never lump cancers together and assume evidence

suggests that the frequency of those mutations would be the so some may be very familiar with this. again, we discussed previously, there are different types of breast cancer. are they the same prevalence across populations? the answer is clearly no.

what's quite troubling is that that most aggressive form, the basal type, with much more prevalent in populations of africans, african descent, and also african americans here in the united states. reasons are not known. because it's the most aggressive

form of breast cancer for which we really don't have any therapies, this is ambiguous a significant dis-- also a significant disparity that contributes to cancer health disparities in terms of outcome when we talk about molecular subtype, what i take on this

slide is true, it's important to point out that the disparities in survival tend to exist regardless of the molecular subtype. so whether it's er positive, there does seem to be differences in outcome. that's also important to bear in

mind. in terms of cell biology, differences, this is work, only a section of the work that [indiscernible] has done, an investigator in the same lab that i work in. he works in prostate cancer. and he has shown for -- this is

one example, one of the key differences in biology between european american prostate cancer tumor and african american prostate cancer tumor, is information. this is very prompt month information significant that you see in tumors, african americans

now, the reason for that is not entirely clear. and his group and other groups are trying to figure out what it is. some can be genetically different. some could be driven by the environment.

but there definitely seems to be a very strong information profile that defines a lot of tumors and aggressiveness of those tumors in african speaking of inflammation, this is one of the things we study in terms lung cancer, and how this might drive differences in

incidence and outcome. it is one of those hypothesis in the body where there is a lot of historical information describing differences in information between european americans and african americans. and even populations of other ancestries around the world.

again, if you think about evolution itself it's not entirely surprising this would be the case. we do know that there is an increase of some a u diseases, higher in african americans. but not all. it's important -- copd, again,

very much associated with lung cancer, the problems with copd is, if anything, lower amongst the emphasis is on some autoimmune diseases are higher. levels of i will 6, non specific inflammatory markers are higher among african americans. some is different allele

frequencies in those genes. we know if you take normal populations, populations without cancer and compare their cytokine profiles, some are similar, so i are different. if you then take the cancer populations, one of the things we have found is if you take

european african american and african american lung cancer patients, there level of il1 beta are the same. but il1 beta is only associated with increased risk in african so that suggests that the biology of that tumor is different and that il1 beta

itself is pleaing an important role, doesn't seem to [indiscernible] in european i mentioned a few moments ago as well, in terms of prostate cancer, inflammation in non cancer, a lot more prevalence among african american men than european american men.

several investigators trying to understand how the differences contribute to how disparityies for many different types of there is also the possibility that given the environmental exposures could drive different biological signatures. this is the idea that exposure

such as smoking into lead to mutation c53, a hot spot mutation. there is a link between exposures with biology that people are trying to track. it's difficult but there is evidence that it can occur. and by studying the biology, in

so i instances you're able to track that to figure out how exposure might have been. we don't know yesterday how or the exposure contributes but it's in this era of next generation sequencing in terms of dna, rap rna, that we're able to ask those questions.

the way we ask those questions is because we take this sequencing, match them and pair them with detailed patient demographic, life style data. important to merge together to cover those type of insights. i'm going to share one example from kevin collins, who shared

the example at the symposium last week. his area of focus is head and neck cancer. it's known that patients that have that actually have a worst outcome, african americans have a worst outcome compared to european american.

through the research they did by combining this exposure data with their clinical data, they also noticed that first of all, head and neck cancers that are hpv positive have a much better very, very different, very different outcome. but what they noticed

subsequently in their studies, the prevalence of hpv is higher in european american patients. what they found, that presence is driving the differences and that's a very interesting point. so again it's very important to link various aspects of the etiology and text poshers with

outcomes, and by doing that we can have insight in terms of the drivers of biological health i'm going to close, a couple of very quick bullet points. these -- i'm not going to go through particular detail. i want to raise emerging points in owe too terms of cancer health

everything i spoke about in the last 15 minutes or so, has been about primary cancers. but the incidence of secondary cancers is also higher in and again, nci is the leader in terms of cancer survivorship cities, and there is a lot of people trying to look at this in

detail to gather more data, also to try to understand the factors that contribute to this. it's not something as i mentioned that i work in, something that's important to be aware of. as more thankfully, more people survive cancer and live with

cancer, cancer survivers, this will be an important area of we should not forget it. the other thing that i want to mention, going back to the beginning of the talk where i defined ethnicity and race. that was self-reported perspective.

but because of genomic and next generation tools, we're able to define ancestry gently, how diverse or homogenous your dna might be. this is an example from a paper a couple years ago in the new england journal of medicine. it makes the point that race is

important, in terms of trying to understand health disparities in terms of exposures, and abscess to care, all those -- access to genetics are also very important in terms of disparities. so for that reason we need to understand from a genetic perspective, who is the person

that we're looking at. and for this example here, this is just -- i'll height one person on this graph. so here, you have someone who self-reported, less than 10% african american ancestry. and in every study, we could say this ourselves as well.

the same thing, we'll see a small portion, a portion of people that identify as one population, but ancestry perspective, are different. that is not to say you can just look at genetic ancestry, or just look at race. the point is you should look at

both. both are going to be important. i wanted to in same vain, highlight the importance of ancestry in terms of screening move it's been noted that an increasing portion of native american ancestry is associated with increased risk of childhood

aal. this is screening implicationsic of course of course. a very important study found that children who had more than 10% african americansen saystry, either an additional round of chemotherapy chemotherapy in order to respond to treatment.

unless you measured ancestry in the way i just described you would never we able to understand that. and that is one of the reasons, an example highlighting to you why looking at genic ancestry, is so important. i want to mention screening and

one other point. i mentioned earlier, we'll take african americans are more likely to smoke less. this is an example shown here. very recently, a couple years ago, nci did a study where they showed screening with low dose screening was efficacious,

reducing the lung cancer by about 20%. this it's wonderful. it reimburses the screening, but the guidelines are based on age and smoking history. 2 things mentioned earlier, a, african americans get cancer earlier, and b, they're less

likely to be heavy smokers, for that reason there is a strong likelihood that more african americans would be excluded from screening programs that european so what is the consequence of that? it could be untended consequence.

in the widening of dissipaters. this is an example of lung. many cancers where there are known efficacious screening modalities, they say what are the criteria we're using and should we lump everyone together. in this era of precision

medicine, should we sit down and think about group specific guidelines? it's an open question. the answer certainly isn't there. but it is an emerging topic at the moment. because we're making many

strides, really want to reduce the potential to widen disparities unintentionally. and the other point is to again put a finish on, we're aware of the differences between tumors, biological differences. i highlighted some of those. but the question is will these

differences -- similar to the point about the different oncotypes. mamma print modalities. we really need to in the clinical trials, enroll more people into these targeted therapy trials. really ask the question, if we

know there are differences can we expect using the same drug on a group of different people have the same outcome? the hypothesis is there will be different outcomes. you really can't [indiscernible] without the clinical trials. without clinical trials you have

to have people. so there is a rot of work to try to increase the presentation of minorities -- representation of minorities in clinical trials. more needs to be done to include minorities in the trials so that every population can leverage the exciting benefits coming

down stream. it's made within the broader context understanding there are differences in biology across these different tombs. that we need to remain vigilant of that concept as we move forward in this era of targeted this is what we talked about

incidence, some of the reasons for disparities in mortality. it is the multi modal problem. it isn't just access to care. it isn't just screening, genetics, it's all these things combined, in terms of incidence and mortality. and the [indiscernible] is one

of those questions that is a multidisciplinary problem. why i love working at the nih. we have so many wonderful people here you can work with. you don't have to be an expert, there are people who are much better experts that i have -- than i am on these areas. so

the goal today was a lot of the areas i've spoken about i'm not an expert in. i wanted to give you a flavor of all the different concepts in terms of health disparities, and to make the point, a lot of research we do tries to understand what causes them, the

idea that knowing what causes them will help us to understand and overcome them. thank you for staying with us, thank you for your attention. if you have questions, either now or after, please feel free to get in touch. [applause] thank you very much.