American Lung Cancer

>>> we're going to get started. and last week i gave the lecture on small cell lung cancer, i'm supposed to be the speaker today but i took the place of dr. nichols, who is going to talk today about radiation oncology. so she got her undergraduate degree at duke university,

subsequently consequent to the university of maryland medical center in baltimore, m.d. degree, internp, specialty is radiation oncology, now an assistant professor at the university of maryland, marlene and stuart greenebaum cancer center.

her title, introduction to radiation oncology. >> i'm going to go over basics regarding radiation oncology for everyone today. i have no disclosures. the outline is talking about the goals of cancer therapy, some

goals of radiation therapy, talk about the basics of radiation physics, biology and clinical radiation and how this impacts patients and i'll spend a couple minutes talking about future research opportunities in i would like to thank dr. camphouseen for some slides as

well. in terms of principles of cancer therapy we've listed five things that i'll spend a minute or two touching on. so basically in terms of cancer therapy we want to cure patients of their cancer, but we want to do that in a way where we're

meeting some goals here. first and foremost minimizing therapy, only giving patients therapy they need to actually treat and potentially cure cancer and make sure we're minimizing toxicities the patients may experience minimizing the amount of time

they are in the doctor's office or hospital, and being mindful of the cost of cancer care. all of us have seen in the news many articles talking about the cost of cancer care in our health society. we want to minimize the negative impact on the patient's quality

of life so , again, this gets back to toxicity a patient experiences from therapy, maximizing their function, so, again, back a long time ago when patients had sarcomas of upper extremity for example they had the arm removed but now in today's world based on studies

done at the nih we can do a lymph therapy so maintaining patient function and trying to maintain a patient's cosmesis, especially in breast cancer, from side to side. sometimes in terms of cancer therapy we're improving quality

of life so this is often seen in the met stat setting where it spread to the bone or lung and can palliate symptoms such as bone pain as a result of cancer being in the bone. we can do this by talking about organ preservation, in bladder cancer patients can have a

cystectomy or removal of bladder or they can undergo an organ preservation approach. we want to cure a patient and help them live as long as possible or maintain patient outcomes and improve outcomes. the discipline intersects with a

lot of fields, radiation biology which we have pictures here, molecular and animal level, at university of maryland we do research on small animals but also on large animals as well. physics is an important part of radiation oncology, as well as clinical aspects of delivering

radiation therapy and these intersect and are important in our field. just going over some basics of radiation oncology physics, so the physics of radiation oncology, what is radiation? this is the definition, which is one of our radiation therapy

handbooks, the complete process by which energy is emitted by one body transmitted through an intervening medium or space and absorbed by another body. there are various types of radiation therapy, and they are divided on this slide into the naturally occurring types, alpha

particles, beta and gamma particles, or gamma rays. we have kind of more man-made particles like photons and proton therapy. alpha particles are like a helium nucleus, heavy particles and can be stopped essentially by a piece of paper.

so if everyone remembers back in japan a couple years ago when we had the nuclear disasters, by wearing clothing on a person's skin they could block alpha particles from hurting them. beta particles can actually go through mediums a little farther than alpha particles but still

can be blocked essentially by a piece of aluminum, don't travel as far. we think of beta particles as traveling a couple millimeters, this is how when we get prostate implants with seeds, as beta particles. lastly there's gamma rays,

basically the way radiation therapy used to be delivered in the '60s and '70s, we had machines called cobalt machines, the cobalt would decay and emit gamma rays. gamma rays need a bigger medium in order to stop them so here you have a piece of concrete and

so just for reference sake in the vaults there's often six feet of concrete which is used to help stop any potential emissions to patients. when we look at gamma rays and talk about photons, nowadays, photons have similar properties but the difference is gamma rays

are naturally occurring. so when we talk about where are we on the electromagnetic spectrum and radiation therapy we're on the left area here, okay, so we can't see gamma rays or photon energies. how are x-rays generated? on the left is a diagram of an

electron tube, so this probably reaches back to a lot of physics from high school or college. basically an anode and cathode, apply an electric current and create x-rays. this is what the modern day radiation therapy machine looks like, and so basically this is

the head of the machine, where the radiation therapy is emitted from the machine, this is the treatment table that the patient lies on, and both this part, which is called the gantry, can rotate in a 360-degree fashion. the treatment table has six degrees of freedom, it can move

in and out, left and right, up and down. and more mod modern tables cando rotational changes as well. the modern day radiation therapy machine is called a linear accelerator, a linac. what happens is these machines can generate high energy photons

as well as electrons. what happens is there is an electron gun inside of the treatment machine, the electrons are accelerated, what we call the wave guide, seen right here. the electrons update go -- get put through a bending magnet to have the energies come out this

way. what happens is electrons strike a piece of metal, they then undergo a reaction and create photons, the photons are emitted by the head of the machine, shaped to create various shapes. when the radiation beam exits the machine, it's filtered so we

have a uniform beam characteristics. as you go from side to side the x-ray energy is exactly the same. we can do precise field shaping to half a cm now. and then we can deliver -- have precise delivery with the

rotation of gantry, couch and patient immobilization. we do radiation therapy planning, there's really two techniques that we now use for radiation therapy planning. one is called three dimensional conformal radiation therapy, 3d crt.

the other is imrt which we'll get to in a second. when we talk about 3d conventional radiation therapy, basically the patients undergo planning cat scan, we can draw where the tumor is located, also where the normal organs are located, and we create multiple

different beam angles to deliver radiation therapy to the patient. so in this example here each of these boxes represent a different beam angle from the radiation therapy, and all those converge on the central point here to deliver the radiation

therapy. the cat scan that we performed allows us to create a virtual patient in that manner. imrt stands for intensity modulated radiation therapy, giving us more of what we call a dose cloud, a complex type of radiation therapy and required

inverse planning. we talk about inverse planning, what that actually means is that before we actually go ahead and start trying to figure out, well, how do i want to deliver the radiation treatment in terms of beam angles, we tell the computer, i want the tumor to

get this amount of radiation and want surrounding organs to get different amounts. for example if i wanted to treat the bladder to 50 gray, i can tell -- sorry, the tumor at 50 gray, i want to keep the bladder which is right nearby at 20 gray, we give that information

into the commuter and different treatment planning softwares have different algorithms to create a radiation plan. what actually happens is from different beam angles you might get funny shapes and patterns, which are demonstrated here, and those all converge on the single

area and create fluent mass, right here. we're able to get a high dose of with minimal to no doses of radiation to the surrounding structures. this is commonly used in head and neck cancers with a lot of critical structures adjacent to

tumor. this is a sagittal slice of a red area represents a tumor. we have important areas nearby, spinal cord, brain stem, some swallowing structures, airway, radiation planning software can modulate the dose to get a high dose of radiation here with

minimal to no dose anywhere else. the areas here represent a different fluence map of what the beam would actually look like. another type of radiation therapy is called brachytherapy. it's when we're able to place

the radiation source inside or adjacent to the tumor allowing high doses of radiation to the tumor itself. brachytherapy is used for certain tumors such as those in body cavities, so basically a lot of gyn malignancy, cervical, endometrial, vaginal cancer,

head and neck cancers, and tumors close to surfaces. for example, prostate cancer, sarcomas, sometimes tongue and lip and even breast cancers. this is an example of a brachytherapy plan for a patient who has a uvial melanoma, we place a shield on the eye which

delivers radiation therapy for a brief period of time. here is another example of cervical cancer patient receiving brachytherapy with a tandem and ring, the circular piece is the ring. these are two hollow instrument a radiation source

travels into them and stops and delivers radiation therapy for pre-specified amounts of time based on radiation therapy plan. if you looked at a cat scan or treatment planning scan for a patient undergoing this type of implant, this is a coronal image here so each of these dots

represents a place where the treatment planning software said we want to deliver radiation, and the lines difficult to see show the isodose lines for radiation distribution. you can appreciate here close to the very high dose of radiation, but within about 1 to 2

centimeters there's rapid dose falloff, so essentially minimal to no radiation dose getting past that area. therapy that's a little bit different is stereotactic radiosurgery, this technique used to treat brain tumors, through the gamma knife.

and what you see here on the upper image is what would be a gamma knife plan, and basically what happens with the gamma knife is that there's 201 different sources that converge on a single point. i tell patients if you drew 201 lines intersecting at the same

point you get a high dose of radiation in the middle of all those intersects lines but a minimal to no dose of radiation through any one of the beam paths. stereo tactic radiosurgery, the concept, we're providing the effect of radiation with surgery

if you were to take the tumor out. it's that precise. precise. we can now treat tumors in other body sites using the same type of technique and call this stereotactic radiation therapy, sbrt, treating lung, liver, bone and other gi organs.

what's important is the ability to immobilize the patient, making sure what we're aiming at stays still. this is an example here of a lung plan, and stereo tactic body roadation for lung tumors is so successful it's challenging the paradigm of

giving surgery to patients for early stage lung cancer. i always like to point out because oftentimes people have heard of this, that cyberknife is the brand of a machine that delivers stereotactic radio surgery but there's other machines that deliver that same

type of therapy but oftentimes now the word cyber knife is synonymous with this as well. so a little bit on the basics of radiobiology, switching gears, this is what a radiation cell survival curve look likes, on the y axis you see a plot and on the x doses of radiation.

from the figure on the left we have two different types of radiation particles, so alpha rays an x rays, somewhat synonymous with gamma rays. alpha rays or alpha particles are what we call densely ionizing, big particles, they stop everything in the track

they hit. what you can see is that with small doses of radiation therapy you get a very high degree of cell kill. conversely when you look at photons or x-rays, and you give different doses of radiation therapy, what you can see is you

have to get to a pretty high dose of a single fraction of radiation therapy to get an equivalent logarithmic value of you can see the alpha beta ratio, tumor tissue and normal tissues have something, a unique inherent property, the alpha beta ratio telling us how

sensitive that tissue is to radiation therapy. so generally speaking tumors have alpha beta ratio between 8 and 10, normal tissues have alpha beta ratio between 2 and 5. on the right here what this figure is also showing us is

that there's what we call a shoulder to the radiation if you look at the curve here for the x-rays and you see kind of where if you were to draw straight line up, from this area, this is what we call the shoulder of radiation, so basically what that indicates is

you have to get over a certain threshold of radiation therapy before you get a significant amount of cell kill. now, we give radiation therapy, we're not typically giving one large dose of radiation. we give multiple fractions over several weeks.

the reason that we do that is what you see here. so by fractionating the radiation therapy, it changes what our cell survival curve looks like and allows it to look more linear like the alpha particles i just showed you. so this is actually the reason

why we fractionate radiation the other reason is this allows us to take advantage of the fact that our normal tissues can actually pair the minimal amounts of radiation therapy that happen on a daily basis. so our tumor will look like this, whereas our normal tissues

will have a different type of curve, that's what's supposed to be demonstrated here. so when we talk about radiation biology, there's what we call the four rs, repair, reassortment, known as distribution, reoxygenation and repopulation.

repair, healthy cells or normal tissue repair dna damage during each treatment of radiation so do the tumor cells. because tumor cells are tumor cells they have a defect in the dna pathway, this allows healthy tissues or normal tissues to repair the dna damage where our

tumor cells are not able to do so or not able to do so as effectively. in terms of reassortment, radiation therapy causes cells to accumulate in certain phases of the cell cycle, we'll talk about that more in a second. reoxygenation, tumors

reoxygennate after being supposed. and repopulation, tumor and normal cells can repopulate between doses, so this is why when we give radiation therapy we want to give it over consecutive days and not give you one fraction one week,

another fraction the next week because basically the cancer cells can regrow in that time frame. we talk about repair, so dna is really the primary target of there's two different actions of how dna is the primary target. one is the indirect action, the

next is the direct action. we talk about the indirect action, that's seen on the top of this figure here. if you have a photon energy coming in, it knocks an electron off. that electron creates some sort of free radical.

it's that free radical that causes some dna damage, okay? so that's what the indirect action is. this is probably -- this happens most when we give radiation in terms of direct action, what can happen sometimes is the photon energy can come in, it

can knock the selectron off and damage the dna. similarly, if we give particle therapy or proton radiation, protons are much bigger particles than electrons, and the proton itself can actually directly damage the dna. we can have two different types

of dna damage so what we call single strand breaks, or double strand breaks. double strand breaks are key in what we're looking for in unfortunately, double strand breaks are actually the least common effect, it happens when we give radiation.

as i mention, this oxidative damage is the thing that happens the most. secondary are these single strand breaks, so those happen on the order of a thousand for every gray of radiation that we give, and when we give one gray of radiation therapy, double

strand breaks, there's typically only 40 per gray of radiation therapy delivered, okay? single strand breaks are much more easily repaired by two different processes which we call non-homologous and homologous and joining repair. we talk about redistribution or

reassortment, another thing that to patients is that radiation because the radiation therapy induces dna damage, the cellllactually pause in it cycle to repair that dna damage because if it continues to go on through the cell cycle it will die. this is one example of pathways

that happen when we give radiation therapy, you can see here and here that these can both cause cell cycle arrest in different parts of the cell cycle. if we look at kind of the cell cycle here, there's two checkpoints.

one the g-1, and the g-2 checkpoint. so radiation therapy can cause cells to accumulate in both of these areas while it's repairing the dna damage that's occurred. this has become a therapeutic target now so there's drugs in development currently being

tested in multiple phase 1 and 2 trials, checkpoint inhibitors, to exploit this part of the radiation therapy process as well as chemotherapy process. to make that more complicated, we also know that radiation therapy cells are much more sensitive to the damage that

radiation therapy can cause in so in particular, the m phase, mitosis, and g2 are the most sensitive to radiation therapy damage. that's because in those parts of the cell cycle the dna is much more condensed and foiled, if it you can imagine as you have a

photon energy passing through with the dna being foiled and close together there's a higher likelihood it can cause direct look at the other phases, so this is what the cell survival curve here is showing as well, m and g2 are the most sensitive followed by g1 and f phases.

reoxygenation, if you have tumor, the farther from theblood vessel, it will be more hypoxic, these areas are more resistant to radiation therapy, the more oxygen it has the more damage it can cause in terms of that oxidative damage as i mentioned. there's different ways when we

take all that information into account, how can we make radiation therapy work best for us. there's different ways to modify the first is through radiosensitization. we have two graphs, the top is the tumor curve, and the tumor

control probability, over dose of radiation delivered, and right is normal tissue. the example here we gave 60 gray of radiation therapy, this might result in 65% local control and 15% normal tissue damage. if we give some sort of radiosensitizer, we shift this

curve -- we give the same 60 gray of radiation therapy, we can improve our local control by well maintaining the same degree of normal tissue damage. the most common way to radiosensitize tumors is giving patients chemotherapy. what happens when we give

chemotherapy, you look at this cell cycle here, chemotherapy will preferentially place cells into the d-2 and m phase depending on type of chemotherapy, that allows us to have more damage when we're giving radiation therapy. another idea is giving

radioprotectors. again, the same curves here, but instead here when we give a radioprotector we can escalate the dose of radiation therap potentially, that's the way we can improve the local control while again maintaining the same degree of normal tissue damage.

in reality, the different areas of research that have looked at different radioprotectors hasn't been as fruitful in providing different drugs to patients as giving chemotherapy in terms of radiosensitization but this continues to be an area of active researh for cancer

therapy but also for protection of all of us should any disasters occur. so when you look at what are some radiation therapy targets, there's lots of ways we can target cells to help them be more sensitive to radiation therapy damage.

so there's single agent targets, such as growth factor receptors, egfr and vegf, inhibitors to both of these drugs that have been developed and can be use in combination with radiation there's been areas looking at dna repair proteins, transcription factors and signal

transsection proteins. the pathways are complicated with no single direct area we can target the pathway to inhibit it. there's also what we can do is multi-target inhibition, looking at chaperone proteins, looking at the microenvironment with

angiogenesis and blood vessels, this has been done somewhat successful with introduction of bevacizumab and epigenetic modification. there's radiation inducible targets as well. this is just one example of egfr receptor and vegf receptor, we

can see how complicated this is. you can see different drugs used to target these different areas, bortezomib and so forth. developing drugs with radiation therapy some issues for targeting agent developments are the mechanism of how the drug is

going to act, so what types of cells are we going to act on? are we going to act on the surface cells, try and act on cells more in the center of the tumor? or are we going to try and have these targets and agents work on specific cancers, just kidney

cells, with the renal cell carcinomas, antibodies that can attack the receptor, successful with egfr and vegf. are we going to use small molecules like gleevec, and consider the therapeutic ratios to kill more tumor cells than normal cells.

moving to clinical practice of radiation therapy, again some goals of radiation therapy are obviously to cure cancer, so if the cancer is localized to an organ or region we can aim the radiation. palliation, patients with metastatic cancer, we can

palliate symptoms they may have such as bone pain from bone metastases, if a tumor is growing on someone's airway and causes the lung to collapse we can give radiation therapy to open that up or if patients are bleeding from a cervical cancer we can give radiation therapy to

stop the bleeding as well. this list is by no means comprehensive but here are some lists of cancers we can cure with radiation therapy. prostate cancers, bladder cancers, breast cancer, lung cancer, head and neck cancers, gynecologic, malignancy

and palliation bone pain, pain from other issues. radiation therapy doesn't operate in a vacuum, it's an integral part of the team. this gives a sense of the other disciplines that up had us take care of our patients. medical oncology surgery, social

specialties, radiology, pathology, even primary care physicians. we have to interface with one another to take the best care of our patients. we're giving radiation therapy, the first thing we have to do is develop what we call

multi-modality plan. there are pretty few cancers treated by one modality alone. oftentimes we're combining radiation therapy with surgery or surgery with chemo and radiation therapy, and so we often all have to meet together to develop a plan for the

this is often done best in the multi-disciplinary setting, so a lot of hospitals, including here, when we see new patients the whole team of all the doctors will actually meet together either before or after the patient is seen, and talk with one another before

presenting the patient with their final treatment plan. sometimes there's other localized therapies that are included for the delivery of the treatment plan, including what we call focal ablation techniques through microwave treatments or cryotherapy, as

well as ways we call focal drug delivery with particles to attach radiation modifiers to and deliver that, radiation therapy is delivered to where the molecule is delivered. what happens in the radiation therapy process? first the patient is seen in

consultation, whether it's just us or multi-disciplinary setting as i mentioned. once we determine the patient should undergo radiation therapy, they undergo ct simulation or planning session. at the time of the ct simulation, we simulate the

patient in the exact way we want to deliver their treatment. so depending on the type of cancer that might vary. so for example in anal they might have the patient on belly or prone. breast cancer i would have them laying on the back with the arms

above the head like you see in this picture. sometimes they may be laying in the this position. we'll make immobilization devices so the patient can lie in the same position each day and we place marks on the skin to line them up using lasers in

the room. i mention this because it's fun, but radiation, we used to line them up each day. we transfer im images, and wecan fuse other images to delineate the targets. for example, a pet scan and mri or other cat scans.

once that's done we do what we call contouring. we draw where we want to deliver the radiation, so the tumor, the lymph node region, depending on type cancer, contouring out the normal structures we're worried about in that area. a plan is created by a group of

people dosimetrists, and i evaluate the plan, if i like it we move forward, and that plan is also evaluated by medical physicists to look at the different physics properties as once everything looks good we would transfer that plan from the treatment planning software

to the software that delivers the radiation therapy treatments and the treatment is actually delivered. as i mentioned a little bit ago, oftentimes for treatment we give the treatment over one fraction per day typically over many weeks.

for each type of cancer a different dose is determined to be efficient in terms of curing cancer without causing too many side effects. cancer that might range from three weeks to up to seven to eight weeks for example in prostate cancer.

treatments delivered one fraction per day, five days a week. patients typically get a weekend or two days, two consecutive days a week off. that allows for normal tissue repair in between fractions as this is an example of one of the

patients i receas planning. this is head and neck cancer patient with an oral cavity cancer. he underwent surgery, and now is coming in for radiation therapy. what you can see is typically now in the treatment planning

software we can contour actually in all planes and this is an axial, sagittal and coronal, we can draw where we want to deliver the radiation, this is where the patient had the original tumor and surgery and for example the carotid gland here is an area i want to spare

from radiation therapy. moving to a couple patient presentations as well, again we'll walk through each of the steps i just mentioned. this is a 55-year-old female who developed a new lump in her left breast, because of the lump on the breast the patient underwent

a mammogram which showed a suspicious abnormality, biopsies and consistent with the most common type of breast cancer, infiltrating ductal carcinoma, no family history of breast the patient is assessed, there's various treatment options this patient could undergo.

breast cancer we are often giving all three modalities, so surgery, chemotherapy and so the patient has a couple different options in terms of surgery, she can undergo a mastectomy or removal of the entire breast, okay, and this is what -- not anymore but kind of

back 30, 40 years ago a mastectomy would look like where they removed the breast and removed the pectoralis, or breast conserving therapy with lumpectomy and radiation in a lumpectomy we see where the tumor is and remove the tumor plus normal tissue around that

area. the patient in this scenario selects breast conservation, allowing cosmetic outcome which is beneficial for patients. they undergo a lumpectomy and sentinel lymph node biopsy, with tracer, allow that to sit for a couple hours, draining to the

sentinel lymph node in the area and so in the operating room we can detect where the sentinel node is located using a small probe or injecting blue dye, so they can -- the surgeon can remove the first draining lymph nodes of the breast cancer. so at the time of pathology,

this example, the pathology reveals a 3-centimeter tumor, t2, and four axillary lymph nodes that are positive. because this patient has some positive axillary lymph nodes they would receive chemotherapy, following their surgery, and after that the patient would

need radiation therapy in order to give them the best outcomes. in terms of radiation therapy, again we have a couple things to think about as radiation oncologist, how am i going to deliver the radiation, is it going to be external beam radiation therapy, or

brachytherapy, again, where we can put this in the area. radiation therapy we have different types of radiation so photons and electrons are the more standard type of radiation or proton radiation therapy. brachytherapy, we talk about sealed sources, sources that are

contained in something, so that if you spilled it, it wouldn't spread everywhere, or unsealed sources as well. this is an example here of what brachytherapy would look like for a breast cancer. it looks probably not too comfortable so we don't do this

very often for our patients for breast cancer anymore. the patient would undergo -- let's say external beam radiation therapy, the patient would then undergo ct simulation, you can see this patient is laying on their back and their arms to the sides,

this is where the breast tissue is located. we put these little markers outside on the skin to help us line up the patient at the time of the planning scan. we remove those. we don't need them for everything.

and then here is where we would deliver, figure out what beam angles, what size of beam, what energy of radiation therapy do we want to use as well. so we create the plan, again the dosimetrists help us do this. one beam coming from this way, the medial beam, one beam coming

from this way, the lateral beam, and what we're balancing in breast cancer is trying to avoid the heart here in the middle and minimizing amount of lung tissue in this field. so in this plan this little thin red line is the 100% idodose line, we're looking for that to

cover the whole breast area. this is how a patient might lay for breast treatment. we take x-ray images on a weekly basis to confirm the patient is in the same condition, we don't just take skin marks to be only source of alignment.

here what this shows is it's also confirming what the beam looks like on a weekly basis as the next example is a patient, 54-year-old male, elevated psa on routine screening, the most common way to find prostate otherwise healthy, no other medical problems.

we find gleason 6 disease, or the lowest risk prostate cancer. so, again, talk about developing a treatment plan, for low risk prostate cancer there's a lot of options. i should have included here as well observation or active surveillance, the patient could

undergo surgery, they could potentially need surgery and radiation if at the time of surgery you find high risk features. radiation therapy, there's a lot of options as well. again, brachytherapy, external beam radiation therapy, or

combination of therapy, or sometimes what we call intermediatate and high risk prostate cancer might give radiation and hormone therapy. brachytherapy, this is a patient i did last week. so what is drawn out here but hard to see is that this kind of

darker gray area is the prostate. and then what we do in the computer is basically place needles directly into the prostate through the perineal skin, each of these green dots represents where we placed the seed in the planning software,

and then here you see the isodose line. the blue line here is the 50% line, so you can see between here and here, which is only about -- less than a centimeter, there's only 50% of the radiation therapy that's getting to the other tissues in the

and then the red is the -- the yellow is the 100% isodose line covering the whole prostate. this is external beam patient getting prostate radiation thorough, red prostate, blue is the ptv margin. what we do for patients getting external beam radiation therapy

is we put two margins on where we see the tumor, so one is the ctv, clinical target volume, the idea behind that is account for micro scopic extension of the disease that we can't see a cat scan, and there's pathologic data. for certain lung cancers, we

know squamous cell carcinoma, there can be microscopic extension 6 millimeters outside what you would see on a scat scan. day-to-day even though we have all these great tools to set the patient up we know we're going to be off within a couple

millimeters on any given direction, that accounts for internal motion of organs. for example prostate cancer if a bladder is full it might push the prostate a different way than if empty.is is the clo, theprostate and the lines carving nicely around the prostate.

the other thing is image-guided for prostate cancer in particular the things we look at are tools called calypso or gold fiducial markers. cone beam cat scan, we can do a mini cat scan on the treatment machine to confirm where the organs are located, make sure,

say, the bladder is as full as i want it to be, no gaps in the rectum, monitor the tumor to see if there's drastic shrinkage or growth too, to modify the radiation plan while we're treating the patient. so what calypso is, three beaconses can be placed,

intramonitoring where the patient is having radiation therapy delivered. if the patient passes a lot of gas on the table and that causes the prostate to move we can stop the radiation beam while that movement is occurring. gold markers, which you can kind

of see here, can be placed in the prostate and this can allow triangulate the location of the prostate before treatment, and account for interfraction motion. what happens here is these are x-ray or kb images each day prior to treatment.

so on the left here is what happened at the time we did our planning scan, on the right is what happened that day on the treatment table so we're looking for those three little gold markers to be in exactly the same type of position. if they are not we can move the

table to make that happen for the patient. this is an example of what i was mentioning when we called a cone beam cat scan, easier to appreciate here on this sagittal image. the yellow line here is what the bladder looks like at the time

i have a little window box that i'm outlining here with the cursor that shows what's going on right there while the patient is on the treatment table and what you can see while the patient is on the treatment table is this is their bladder now, not as full as on the day

was not lining up, we would have the patient get off the treatment table, drink more water, re-seat the scan and see fit matches up. we would deliver the treatment for this prostate cancer this is an example of a patient under treatment.

so is it all that easy? so this is getting to the point that radiation therapy isn't without its own side effects. the things to remember, they are all local. we only affect areas we're there are cumulative, meaning at the beginning the patient

doesn't have any side effects, as the therapy goes on they develop. and once the therapy is done they taper off again. i tell patients it's a bell-shaped curve. patients can have acute side effect due to normal tissue

damage which the body repairs over time as well as potential late effects as well. short-term effects don't start for the first couple weeks, it's really that second half of therapy and couple weeks thereafter there are the hard part, where the body is healing.

for the majority of the time most of the patients' acute side effects are resolved within a month of radiation therapy. late effects on the other hand don't occur during treatment but an happen months or years down the line. so for example for late effect

for prostate cancer, sometimes down the line patients can develop what we call urethral stricture, they might feel like r that they are not emptyingthey have to force the urine out their bladder. we have something we can do about late effect to get rid of them.

oftentimes they have to do with the formation of scar tissue, there are times where we can't reverse or help that. classify them into different groups have to do with stem cell depletion, chronic oxidative damage, destruction orfibrosis. radiation therapy is dosed to

normal tissue and not tumor. it's about how much dose can we give to the tumor without hurting the normal tissues. generally speaking what we look at is the risk of 5% of patients developing and injury at 5 years is what we specifically look at. if we talk about lung fibrosis,

this is something that can happen down the line for who got treatment in the early 1990s which is really not the technique we deliver radiation therapy a lot anymore but what you can see is this is what the field looks like. it was the rectangular box.

you can see on the cat scan, all this white stuff here and here is fibrotic damage from radiation therapy, in the same shape as radiation field. fibrosis is thought to be mediated by tgf-beta, there's a lot of work being done on how to mitigate effects tgf-beta has to

prevent pulmonary fibrosis. once that occurs there's nothing we can do for that patient. the good news is nowadays we don't see this very often, but it occasionally does happen. another example, lymphedema. here is a breast cancer patient, the right arm is significantly

bigger than the left arm. oftentimes for breast cancer this isn't purely due to radiation therapy, also to a combination of having lymph nodes removed, but radiation therapy typically increases that risk in breast cancer patients by 10% or so.

you get some scar tissue that happens around the lymphatic channels, so the lymphatic fluid can't drain as well. they are treated with physical therapy and massage or compressive sleeves and often do quite well down the line. a short-term reaction as to do

with cell cell depletion, when we give radiat therapy to mucosal cells, you can see the whitish areas are a little bit of bleeding as well, and this is one of the side effects we e a lot during radiation therapy in terms of pain control and so forth typically resolving

within three to four weeks of from here, radiation therapy evolved a lot in the last 60, 70 years, from the development of the first linac down to imrt. and we're continuing to see more advances as well fore the future, biologically, how can we use

radiation to induce targets for other agents? how do we get better radiosynthesizers, and protect monoclonal antibodies and is that safe? there's research in all these areas. in terms of the physics of

radiation therapy, how can we improve our treatment delivery, make our radiation beam even more precise than it was before, how can we account for better organ motion, so with the lung moving as we breathing how can we immobilize the lung tumor and improve our targeting maybe by

fusing additional studies or using cone beam therapy or cone beam ct scans to adapt radiation therapy planning. and then in terms of clinically how can we translate some exciting lab findings that many of you here are working on to help improve the outcomes for

our patients? so in terms of some newer modalities, there's radiopharmaceutical therapy where is we attach radiation therapy particles to different chemotherapy agents, or sometimes we can deliver radiation therapy through

unsealed sources, into the bloodstream, that can congregate into areas of tumor, we do that sometimes in metastatic prostate cancer with irradium 226. this is an example here of cyberknife which i mentioned earlier, so it's on a radiation therapy machine that operates on

a robotic arm, so again can deliver radiation therapy through any beam angle, and proton therapy as well. proton therapy, i want to mention that because we'll be having a major center open in the region in the next month. so proton therapy has a

different physical property than photon radiation therapy. so what the graph looks like here is so the blue area here is what we call regular photon this is the skin surface, this is as we're going into the patient's body. at the skin surface there's a

lower dose of radiation, that dose builds up in the tissues, and depending on the energy of the x-rays used at a different depth for different energies. into the patient's body the x-rays then -- the dose decreases. x-rays do not stop.

they pass through the patient. proton therapy is very different in that it has a lower entrance dose but we can stop the proton beam wherever we want it. if we have tumor here, by choosing a different energy of proton therapy we can stop it. you see there's not this tail

you see here with proton what this allows us to do is give radiation therapy to our target tissues, and even further decrease the dose of radiation therapy to normal tissues. this is the therapy that's thought to be helpful in pediatric patients or areas

where we have a lot of critical normal structures like head and neck cancers. so this is an example here of is the breast cancer patient, so it compares proton radiation plan and a photon radiation plan. so we might say why are we

worried so much about breast cancer? we see the heart is close to the breast in certain women as well as lung tissue, an important paper two years ago showed for every gray of radiation we can minimize to the heart there's a 7% less risk of having some sort

of cardiac event which they define as heart attack, need for a stint or death from heart disease. potentially be a very important long-term effect for women. what you can see from the proton radiation plan here, the yellow contour is the breast tissue,

red area is the high dose of radiation, the prescription dose of radiation, and the blue here is much lower doses of radiation probably 20% of the dose, we can see the red area here gets a high dose of radiation therapy, but importantly the heart here is basically getting no

radiation therapy, and the gray circles are the coronary blood vessels as well. with the photon radiation therapy plan you can see a lot of the part is getting the low dose bath of radiation therapy and all the coronary vessels are in the area as well.

you can see this is subtraction image where we put this plan on top of this plan and subtract it out and you can see all that difference is in the lung tissue as well as the heart as well. so in this example the proton plan could deliver much less dose to the heart which based on

the more recent data would have a clinically meaningful significant impact on the patient's life in the future. as i mentioned, the university of maryland in partnership with a private company, apt, is opening the maryland proton treatment center down in

baltimore. we'll be starting to see patients this next month, and turning the radiation therapy beam on in early january. so this is thought to hopefully provide additional resources for our patients, in particular the pediatric population, where

again the long-term effects, we worry about much more than say an elderly person. our vision is to become a center of excellence in terms of education and research and be accessible to patients and partner with regional health care systems and the oncology

providers. it's a $200 million facility with 110,000 square feet, five treatment rooms, at the end of the day we'll treat up to almost 200 patients per day there. so the takehome messages, radiation therapy is an important tool used in cancer

radiation the way it works is by causing dna damage which leads to cell death ultimately. effects can be altered by modifying a lot of things including fiscal factors, physiological factors, fractionation, drugs and other variables.

radiation is interesting, especially in our era now with some new advances coming to the region. so thank you. [applause] >> questions? is the proton therapy good for patients with glioblastoma?

>> so the idea behind proton therapy since it can be stopped at any place we want, in particular for brain cancers like jbm or astrocytoma it can allow potential to doses cancer therapy late, give higher doses of radiation therapy to the tumor than what we can currently

achieve in photon because of the brain stem and optic nerve and so forth. the idea is that if we gave the same amount of radiation with proton therapy it might not be better than photon. proton will allow us to doses cancer therapy late which in g

bm data says that might be beneficial in terms of curing (inaudible). >> the question is what percentage of patients get radiation therapy? it depends on the type of when you take all in all with

the exception really of some of the leukemias and lymphomas, almost all cancers have some role for radiation therapy at a certain stage. about 50% of patients we're treating for what we call curative purposes, as initial treatment to cure cancer, the

other half have metastatic disease and we're palliating but a large majority of cancer patients receive radiation therapy as part of their >> (inaudible) >> great question, yeah. the question, with radiation

therapy, malignancy, it would be help if the if the immune system was boosted. radiation causes inflammatory reaction, cytokines are released into the body's system. for certain cancers in particular we have actually found that that immune response

actually helps to -- helps the cancer respond in all other areas of the body. we call that abscopal effect, most recently shown in melanoma treatments. we know that disease in and of itself is modulated by the immune system.

and there's a particular antibody therapy, ipiluminab, that we've given to patients. when you take that therapy alone, it does kind of help the cancer respond in all areas of the body. if we give radiation to a single site while getting that drug

we've seen responses in other for example, someone has a spot on their arm, we're giving radiation therapy and combination with a drug, a tumor in another part of their body will shrink even though we're not giving raid radiationtherapy to it, looking at how we can

harnett effect -- harness the defectiveness of the immune system. ipiluminab and other drugs recently out are looking at exactly that. thank you all. >> okay. our next speaker is dr. ryan,

she got her undergraduate degree at the university of college cork in ireland, subsequently ph.d. in cancer biology at university of college dublin, and came to the nci -- cancer prevention fellowship program, she currently works with kurt harris at nci and is an nci earl

statin tenure track investigator. the title is the causes and consequences of cancer health disparities. >> thanks very much. can everybody hear me okay? all right. i work here at nci, i did my

postdoc here, before that i came from ireland and didn't pay all that much difference or attention to health disparities because i came from quite a homogenous population to be totally honest. i came here and was doing research on lung cancer during

my postdoc, translational epistudies, populations included european americans and african-americans. started to become clear to me there are disparities in incidence and outcomes across many cancers. of course for any scientist who

observes that, the key question you're going to ask is why do the disparities occur. in the lecture today what i hope to do is take you through the backgrounds of the field in general and discuss some disparities in cancer incidence and outcomes and present you

with some evidence from many different fields of science which are trying to address and overcome each of these okay. i want to mention something specifically at the beginning. this is just to give quite boring definitions in some terms

of what race and ethnicity mean. i'll refer to race today, i want to make sure we're all on the same page. so you know what it is i'm referring to. by some definitions race refers to biological differences between groups, assumed to have

different biogeographical ancestry or biogenetic makeup. ethnicity goes further, a multi-dimensional construct, reflecting biological factors but including origins, customs, beliefs, sometimes they may not have a common genetic origin. for the most part todayly speak

to race but it's important when we're talking about the field to understand that both race and ethnicity can play a factor and role in health disparities. the other point i want to make from the outset is in my laboratory study of lung cancer, we'll talk about that later on,

when i study these populations i refer to populations of european descent is european-americans and of african descent is for the most part with as asterisk they mean the same thing. i don't mean to confuse you in any particular way, if i become

confusing let me know. and white, some use the term european-american and african-american so we tried to stay faithful to that. observe if you read the literature, they are indeed amongst our lives on a daily basis is race difference in life

expectancy in the united states, there are significant differences. on average if you're a black man you will live 6.5 years shorter, a black woman can expect to live 5 years shorter. what are the key contributors to these disparities in terms of

disease? the first thing is cardiovascular disease. there's a whole field of literature that addresses the rule and contribution of cardiovascular diseases to health disparities. this is not something that we're

going primarily to discuss today even though it's a very important topic. the second biggest contributor to the differences in life expectancy between men and women in the united states is cancer. that's what we're going to talk about today.

so the nci defines cancer health disparities as differences in the incidence, prevalence, mortality, and burden of cancer and related health outcomes among specific populations in the united states. numerous, numerous, numerous studies have shown

african-americans have the highest death rates from all cancer sites combined, and espely from malignancy of the lung, colon, rectum, breast prostate and cervix. it needs to be researched and addressed. if you look at the incidence

rates of cancer, all sites combined in the united states, this is looking at men first of all, what you can see clearly is that african-american men have a much higher incidence of all cancers combined compared to every other race. look at women, you see white

women have slightly higher rates of all cancers compared to all other groups. perhaps one of the main driving factors for this is because cancer, breast cancer in women, is actually higher in white people compared to black women. this is probably one of the

reasons why overall you see a slightly less incidence of lung cancer -- i breast cancer through the groups. one of the other interesting things to note is that in general, even though african-americans are more likely to have -- to be

diagnosed with cancer, they have also noticed for quite a few cancers african-americans are more likely to be diagnosed at a younger age. on this graph you can see we have gray bars and black bars. to highlight those that are black indicate those that are

statistically significant. in previous years studies have been done that addressed this issue and found quite wide ranges and earlier age of diagnosis for african-americans. this study published this year looked at the problem through a different way and controls their

analysis in a different way than others. they still find in general african-americans are diagnosed at a younger age, but the difference in diagnosis was somewhat smaller. there are several cancers where these differences remain.

these are shown here at the bottom, including non-hodgkin's lymphoma, anal, soft tissue, karposi sarcoma. quite a few are related to hiv infection. most of the differences that are observed in the diagnosis seem to be driven by hiv-related

cancers but there are still differences overall. excuse me. i have a small cough. there are four main cancers where the cancers are diagnosed at a younger age in black men and women. so what could the potential

causes of the early diagnoses be? as i mentioned one of them could be an etiologic heterogeneity. perhaps the cause of the cancer is different across the groups so there could be a different etiological agent in different populationses or cause cancer at

different agents. different subtypes could be causedded by different cancers that could contribute to across different populations there could be differences in the timing or intensity of the exposure. for example, exposure to tobacco

could occur at one earlier in one population versus another and on the graph previously that i showed you lung cancer was one of those cancers which is detected or diagnosed earlier amongst african-americans. and in addition the timing, the prevalence and frequency of

early cancer detection can vary. through screening cancer might be detected at an earlier age. i will come back to this later and talk about why the age at which cancer is diagnosed in important in combating those disparities in cancer. switching from incidence rates

to mortality rates, shown here you can see graphs for both men you can see again for every single cancer type the mortality rates are higher for both black men and black women than every other group in the united states. as i mentioned on the previous

slide african-americans have the highest rates of cancer-specific mortality, and we know that the racial differences are not reducing over time. in general, the survival from cancer almost from every cancer site is getting better. but those differences in

outcomes are not getting smaller. that's something that has to be for cancers like breast cancer, an indication they are getting worse. if you look at this graph they show four major lethal cancers, prostate, breast, lung and

colorectal. for all combined african-american men and women have a much worse survival than europeans. so what are the key determinants? his a figure that was taken from a recent paper that's

complicated in organization but i liked it because it touches on each of the factors that contribute to the disparities. most of these we're going to go through to different degrees in the subsequent slides. so if we think about incidence for a moment, what are the main

causes of disparities in incidence? well, one possibility is geography, so location,hiclocation. another possibility is perhaps genetic. could it be due to different susceptibility perhaps? others are tobacco use,

nutrition, exposure to viruses such as hepatitis b and hpv. lack of early detection, lack of timely and aggressive treatment, access to care is a huge problem in this area, but also towards the end we'll discuss potential role for genetics and biology. going back to the incidents and

looking at geography, this is a graph of all cancers combined across the united states. even if you combine all cancers together, irrespective of race, incidence of lung cancer differs by state, kentucky and delaware are some of the highest, you can see there are geographical

differences in the incidence of if you look at lung cancer specifically, lung itself, there are states where there are high rates of lung cancer, others where it's quite low. if you break this down by race, you can see some states have similar incidences, there are

others where the differences are quite different, this is one here. it's quite low and you can see in african-americans it's quite high. these are observations that we can see. we do not know what the causes

of these geographical differences are. some people speculated it could be due to residential migration. it could be related to racism, advertising, cultural influences, community structure and also social stress. it has been shown low

socioeconomic status can confer risk beyond an individ's ses. it could be because neighborhoods could have an unequal burden of pollution. for example areas that have the highest african-americans living in them have the highest exposure to cancer-associated

pollutants and increased exposure to secondhand smoke, a risk factor for lung and perhaps pancreatic cancer raisings the question whether a deleterious neighborhood could affect outcomes through physiologic adaptation to the neighborhood. the mechanisms of adaptation is

likely to include epigenetic modification of gene expression, this is something we'll refer to later on when we talk about biology. what about genetic susceptibility, is there role for genetics in susceptibility for increased incidence of

cancer in different populations? some of the best evidence comes from prostate cancer, many of you will be aware of the important link between the 8q24 looking at 8q24, identified an area of increased risk, but importantly the risk more common

among african-american men compared to european-american men, excess of african ancestry at 8q24. up to 50% of disparities between african-americans and european-americans in terms of prostate cancer incidence can be explained by this specific

locus. recent work identified another location, 17q21, associated with prostate cancer risk, alleles more common in patients with african descent. this locus it is believed accounts for another 10%. 60% of prostate cancer

disparities could be explained by genetics. in terms of breast cancer, there are racial differences in the presence of 5p15 location. the risk alleles at this locustwice as common in populations of african descent compared to european-americans, and african-american women more

hikely to be diagnosed with er 3 negative breast cancer. the evidence for genetic contribution to disparities for breast cancer are not as strong as prostate but genetics can play a role. if we look at genetic variation in recombinase gene it's found

it drives hot spots of combination, specifically in recombinase as we whether or not this is associate the with disparities in lung cancer or others remains to be studied. let's look at lung cancer in men

and women, smoking is the mean etiological factor. look across different populations, you see that there's very striking difference in designs of lung cancer across groups in men and women. african-american and native hawaiians have the highes%

latinos seem to have the lowest, a trend in men and women. it's different in tobacco use. the mean driver of this disparity, when you look at tobacco use among this population you might expect to see quite a similar pattern, especially given tobacco use is

the major etiological cause of over 90% of lung cancers. seeing a trend, high intensity smoking prevalence, 5.8% of african-americans are classified as heavy smokers compared to 19.2% of european-americans. what this suggests is that while we're not excluding tobacco

might play a role at least initially it does not look like it is a major driver of the disparity. we also know african-americans tend to initiate smoking later in life. they smoke fewer cigarettes, as adults, the prevalence of

smoking seems to be the same but of course the intensity seems to be less. individuals and researchers have looked at never smokers, if you look at never smokers you can see disparity in incidence. if you look at patterns of smoking it's known

likely to use menthol cigarettes, studies looked at whether or not actually smoking menthol cigarettes in itself can carry a higher burden of the literature is conclusive to say to date that smoking menthol cigarettes is not associated with an increased risk of lung

cancer relative to non-menthollated cigarettes. a recent study did a nice study where they looked across each of the ethnic groups, looking at those who smoke less than ten cigarettes a day and those high dose smokers, more than 30. they looked in men and women.

what you can see in the graph is that again without question similar to before cancer compared to every other group. when you look at higher doses of smoking you can see this difference is attenuated. data from this paper suggests

for reasons we don't yet understand, it is possible susceptible to lower doses of tobacco. we do not know the reason for this, it's never been tested in a laboratory but the data for disparities does exist at higher doses of smoking.

there seems to be an increase the susceptibility at lower dose of tobacco, something we'll discuss later as part of the consequences of this. so as we mentioned we went through a few disparities, geography, genetic susceptibility and things we

don't know yet. we move to potential causes for mortality. the first thing is lack of early detection. for example, if you look across at these five cancer types, you can see on parallel, for all of these different cancer types

african-americans are less likely to be diagnosed with an early stage disease. you can see that here for colorectal, esophageal and medical melanoma. kidney and liver no difference. we know being diagnosed with

cancer at earlier stage leads to better prognosis. by detecting cancer in early stage you're more likely to survive. this is something we have known for a long time. could the disparities be due to lack of early detection? well, early detection is

inherently linked to screening, when they are more likely to be treated and cured. there could be an access to screening for some but breast cancer mammography use is similar among different populations in a setting of equal access to care.

we know regardless of equal access to care colorectal is lower among african-americans, and for hpv there seems to be a lower screening in these populations. it's important to mention that differences in survival exists for cancers even where there is

no validated screening program including cancer of the lung and esophagus, so access to screening does not account for differences in lack of early detection for cancers such as lung and esophageal because there isn't mechanism of screening in place for

populations, i should say there's a screening mechanism for that this year. so could the differences in cancer mortality and disparities be related to access to care? this is a nice study published two years ago. it shows that if you take three

groups of populations, this is for several cancers, those appear in green insured people, the -- yeah, the blue is medicare patien, and the dotted red line are patients who are uninsured. you can see for almost every single cancer that's shown here

those patients that are -- oops, sorry. those patients with insurance have a much better outcome. that suggests access to care is an important factor. several studies have been done in setting equal access to care. this includes v.a. hospitals,

dod health systems, regardless of your age or gender or race you have an equal access to care. this is one example from lung cancer which was published but there are more examples in the field. when there's a non-equal access

to care, if you take those patients and look at them in equal access to care settings regardless of histological types there's absolutely no disparity in outcome. so what this suggests is that access to care is a very important contributor to the

disparities in mortality across different populations. this is another example here for renal cell carcinoma where there isn't equal access to care the mortality is higher in populations, but if you look at equal access to care setting there's a suggestion they do

better, similar trends seen for pancreatic cancer. multiple myeloma is a cancer where there is a significant difference incidence. likely to be diagnosed early and more often than european-americans. if you look at equal access to

care setting, disparities in outcomes tend to go away, and in some cases they do better, suggesting access to care is an important factor in looking at the determinants in cancer outcomes. in addition to having access to care an important factor is

uptake of care, and we're going to discuss that. even amongst those on medicare which is considered in some ways to be equal access to care system there are differences in uptake of care when you compare european-americans and here is one example from lung

if you look at surgery, you can see that european-americans are surgery than african-americans. the same is -- similar trend for chemotherapy. if you look at african-american renal cell carcinoma patients they are less likely to receive surgical treatment, and also

more likely to die of competing causes compared to european-americans, it's not always about access to care. another important is uptake to african-americans were less likely to receive surgery and chemotherapy compared to this is a regular medical

setting, not equal access to racial disparities, this would consultant, disparities in treatment for metastatic cancer exist. these are important things, all factors that contribute to what are potential reasons

related to uptake of care? some studies try to address this. some potential factors that influence the uptake of care include in some cases personal beliefs, fear, specific cultures, ethnicity, the patient-doctor relationship or

provider relationship, patient bias, provider bias, communication can be very important. indeed in some cases certain comorbid conditions can increase or decrease likelihood a patient will go for surgery or another important potential

contributor to disparities is smoking cessation. we talked about tobacco early talking about reasons for disparities in cancer incidence. it's been known for a while now patients who continue to smoke after cancer diagnosis have a significantly worse outcome than

those who actually quit smoking. this is the scale, the bottom is messed up, i apologize, but this second example is lung cancer. those who continue to smoke heavily have an eight-fold or nine-fold increase relative risk of death which is quite high. that is why smoking cessation

messages really should be a very important part of your cancer treatment and there is a push, an additional push, to make sure that message gets across to you can see not just lung, it's for colon, pancreatic, and liver cancer, increased risk of death among those patients who

continue to smoke. why is that important in terms of disparities? we know that racial disparities and continued smoking exist. especially amongst we know that what are the possible reasons? socioeconomic vulnerabilities,

poverty and secondhand exposure can contribute. we know although the majority of african-american smokers express a desire to quit smoking, they are less likely to receive and use evidence-based treatment. what are the evidence-based treatment?

screening for tobacco use, advice for quitting, smoking cessation pharmacotherapy and counseling. likely to enroll in smoking cessation trials and menthol cigarettes may be harder to it request and contribute to poor so we discussed several factors

that seem to contribute quite largely to disparities in outcome among african-americans. it includes access to screening, access to care, utility of this care, of course smoking. but there seem to be some cancers even in equal access to care settings disparities exist.

i'll take you through a few. the first of these, the most prevalent and reproduced in literature is example of breast cancer, studies have shown even in the equal access to care setting, african-american women have a much worse outcome than european-american women.

this is shown in many studies. this is another example showing a clinical trial setting, where again the patients have equal access to care and in this study what you see is that for several cancers including pre-menopausal breast cancer, postmenopausal breast cancer, advanced

non-hodgkin lymphoma, advanced ovarian and prostate persistent disparities in outcome exist. biology can be a contributing factor. i mentioned prostate cancer a moment ago but it is important i mention that the literature regarding disparities in

prostate cancer is more breast cancer. there's some studies that show if you have an equal access to care setting, african-americans and european-americans have a similar outcome. we have similar outcomesthere's many studies showing where african-americans have a

worse -- a poorer survival. it seems to be that perhaps some of these disparities in literature could be explained by the fact that some of the prostate cancers diagnosed are lethal, some are indolent. there's faster growth rate in european-american versus

african-americans, compared to european-american tumors, clinically localized tumors in african-american most closely resemble metastatic than white men, actual cancers diagnosed might be more aggressive than in african-americans than they are in european-americans so it's

important that while data suggests access to care is a factor in many cancer disparities, we went through a few examples for multiple myeloma, renal carcinoma, other factors specifically for breast cancer and prostate cancer suggest that tumor biology an

genetics may also play an important role. so when we referred to differences in biology and genetics what kind of things should we be talking about? we include genetics in cells, susceptibility of loci, somatic expression, methylationtype,

profiles, indeed inflammation and cell biology. we're going to go through a few examples of these in relation to breast cancer and prostate this locus is one we mentioned earlier, the 8q24 locus, a higher incidence of prostate cancer in african-americans we

know that same locus is associated with higher grade and more aggressive prostate cancers. this is an example shown here for that region. can you see this is linear increase in the aggressiveness f the prostate cancer as

diagnosed for individuals who carry -- who don't carry the risk alleles for those who have one copy, and those who have two copies. these patients are also more likely to be diagnosed acally with the lymph node metastasis. there's a genetic contribution

to aggressiveness in african-american men, seems to be driven by 8q24 locus. there's also a duplication event at 14q 3.33 gene, discovered i should say in familial studies suggesting it could lead to inert haded predisposition and

is possible but not proven it contributes to higher prevalence and mortality of prostate cancer in african-american men. aside from these particular familial and inherited studies, there's also evidence to suggest that if you look at the tumors themselves and compare them

between african-americans and european-americans the copy number at different loci across the genome is different between the two populations. here is one example showing african-americans, you can see some loci, yellow are those gained, blue are lost.

you can see looking at two different populations, you can see there are several differences suggesting genetic level there is a contribution, difference of the biology of those tumors. looking at somatic mutations, there's a very important role

for driver mutations, not just in the biology of a tumor but also in the outcome. in lung cancers, prostate cancer, excuse me, some of somatic changes are p and p-10, studies compare the different somatic profiles between african-americans, there are

some differences what they find is some of these key somatic drivers mutations are also differences in them. so while the higher frequency of fusion gene in european-americans is lower in african-americans and even lower again in asian populations.

p 10 loss is common in populations of european descent, it's uncommon in asian and african-americans showing from a geographic population perspective there is global heterogeneity. breast cancer is one of those cancers where there is specific

subtypes at a molecular level that have been ascertained. we've known for quite a while now, quite a few years, that there are population differences in these prevalence of these specific subtypes. one example would be there are in the prevalence of

triple-negative breast tumors. if you look here on this graph, you can see the basal like phenotype is found in 39% of african-american patients but 16% of pre-menopausal. how could this contribute to poor outcomes? we know that this particular

subtype is extremely aggressive form of breast cancer for which there is very few therapeutic that could be one of the reasons why it drives the worst we also know that in populations who live in africa and breast cancer patients from africa there are even more likely to

have this type of breast cancer. again, suggesting that there is some kind of genetic contribution to this particular phenotype in amongst breast cancer patients which today hasn't been ascertained. one important caveat to note, and a little bit to counter what

i just said but not to confuse you, that the association between a difference in survival between african-american and european-american breast cancer patients is certainly seen across several different cancer types. even though i've mentioned that

this triple-negative breast cancer is more common in that's not to say that it is the only or the sole or the key driver of those disparities that is important to mention but likely to be a contributor. what about the cell biology itself?

some studies that we do in our lab on lung cancer look at this. and there's been several studies done in recent years as well where looking at breast cancer and prostate cancer you've taken apart and tried to dissect through the cell biology of when we do this across every

cancer type we seem to see a certain set of genes that seem to be race specific. we see this in lung, breast, prostate, colorectal. these specific genes are not typical cancer genes. there are racial differences in gene expression they don't to be

specific cancer genes or what we've come to know as cancer genes. take a step back and look at what the pathways are that are represented by these genes. in prostate cancer we can see that the majority of these genes that are different by race in

prostate tumors seems to be related to the immune system. an example here of some genes upregulated in african tumors compared to european-americans tumors, many genes are related to viral infection response genes and interferon signature. in similar work in breast cancer

interferon signature predicts survival and poor outcomes, also associated perhaps to resistance to chemotherapy suggesting that the biology of the tumor itself could be a driver of the as i've mentioned in breast cancer a similar thing has been found, the same gene signature

was not observed there was an increase in macrophage infiltration, differences are important drivers of differences it's important to know that information because once you identify a difference it could be related to outcomes, to

leverage that difference to improve outcomes. so we also know that even though differences in methylation are not as prominent as differences at gene expression there are methylation differences in tumors. you can see there's a middle

section here which is much different than the other two subgroups on the left and right. you can see that most of european-americans cluster here. globally they are not extremely prominent, there are changes in methylation between the two and to go back to something we

mentioned earlier with regard to socioeconomic status and neighborhood effect, a study was done where they looked at population differences in gene specific dna methylation at birth and saw -- this is looking at 100 european-americans, african-american newborns.

they did see differences in the different populations but the differences between the two populations clustered in cancer-related genes. this is an observation, we don't know if it's causal or not but certainly something some people are following up on.

i've mentioned inflammation that seems to be important in terms of prostate cancer and breast cancer disparities. it's interesting because there's quite a long literature that describes different inflammation between african-americans, european-americans, in general

outside of cancer context. we know there are some autoimmune and infectious diseases more common, there's increased circulation levels of il-6 and other cytokines and we know cytokine profiles in the serum are also different and

interestingly inflammation and non-cancer prostate biopsies found to be more preval amongst african-americans suggesting a heightened role for information, and it's also known likely to have some comorbidity. the idea that inflammation could be a contributing factor from a

cell biology perspective is backed up by strong literature that supports a role for inflammation in differences between biology of these two so if there are different cell biology signatures, we talked about germline snps, increased il-6, driven by a different

haplotype in african-americans. mbl 2 is associated with colon cancer, only in studies that we've done we've found that that particular genotype is associated with increased circulating levels of this particular gene. this becomes a risk factor for

it could be quite due to different mutations required in the tumor. it could be because of different etiology. we know smoking is associated with p53 mutations. there can be they tight link between exposure and somatic

mutations i mutation which can provide clues to the etiologic source of tumor to differences in etiologies. of course it could be due to toxins, air pollution, carcinogens, so many things. what we don't know is if and how exposures contribute to biology

to determine cancer disparities, that's one of the important questions we're trying to address. so i want to go through a few additional perspectives to finish up. one of the things that is a good -- well, one of the

positive things in the advance in cancer research in the last number of years is there are many more cancer survivors. this is something we're trying -- it's a good thing. the problem is cancer survivors are also at increased risk of some studies have started to be

conducted to compare the incidence and prevalence of these second cancers across perhaps it shouldn't be surprising to note while there were differences, we also now see there are differences in the prevalence of second cancers. this is important because from a

prevention perspective, from a treatment perspective, from an access to care perspective these are all very important observations that should be kept in mind from a medical care perspective. in addition one of the other factors that

has been more and more prominent in health disparity research in the last number of years and driven by advances in technology is something called ancestry informative markers. for example shown here is the frequency histogram which shows percentage of the african

ancestry in a population in cleveland. we know african and european ancestry vary greatly. if we study and administer questionnaires, i'll ask you what is your race, self identified race, some people say white, some people say black,

some people say hispanic, some people say native american and that's okay but there's an increased granularity we lose by asking that question. if you ask somebody what they are race is, and they say i'm white, if you actually do a genetic test on their dna and

say, well, okay, how much dna has a european ancestry, in most cases you can see it's pretty homogenous. most people self identify as white have a fairly strong european ancestry. there's only a small, like 3% of african ancestry in their dna.

an extremely wide range. if you ask an african-american or black person same question, you see there's an extremely wide range of african ancestry in their dna. for questions, when we're looking at genetic susceptibility to disease it

becomes important to address these questions. this is just -- yes? >> gnawed (inaudible). >> yes. yeah. in fact, only -- you mentioned something i had forgotten to mention, which is only 3% of

individuals who identify as black had greater than 80% african ancestry. there's a huge variation at a genetic and quantitative level in the amount of african ancestry. for things like access to care, that may not necessarily be

particularly important, but when we're looking at biology of tumors and genetics it becomes quite important. i'll give you another example in a moment, so very well thought this is just another example looking at various african-american communities

across the united states, and the percentage of european contribution to those. it varies ten-fold. this population here, self identified as african-american, 35% of their dna is actually european-american. in some cases lower, example

over here, sorry, over here, 3.5%. to give you another example of the rain in percentage of african and european ancestry in these populations. moving into perspective of screening and also of importance from ancestry perspective, many

of you will have heard about psa, prostate specific antigen, a marker used to try to predict risk of prostate cancer development. several studies have shown psa level can vary by age and ethnicity. if you take this population and

say based on self reported race, measure psa, the cutoff is 4.0. if you ask european-american population who have a psa value of 4, you can be about 38% positive within three years those are going to get cancer, if you look at an african-american population and

take the same cutoff, this population 62%, that population is 62% more likely to get -- not more likely, let repeat that, for african-americans with a psa level of 4 there's a 62% likelihood within three years that population will develop the same markers in two

different races by self report gets different information. this is important for screening guidelines because if you are to take somebody who comes into the clinic, a doctor is going to take somebody, based on looking at everybody the same way, you're not going to get that

level of important personalized care that is needed. if we take this a step further and introduce ancestry informative markers how does that help us giving a more accurate diagnosis? remember, i told you that within those african-american

populations there's some people who are primarily european and some people who are with a higher percentage of african how is that dealt with? measure the percentage of african ancestry in each patient and divide into three groups, so low, medium, and high, what you

see is that as african ancestry increases, the likelihood of getting the right diagnosis also increases. by introducing this type of marker into your analysis you get a greater refinement and granularity and that can be very as another example, i'm talking

about screening here, but from the same perspective there's an increased proportion of native american ancestry associated with increased risk of childhood accuse lymphoplastic leukemia. again, children with more than 10% native american ancestry need additional round of

chemotherapy in order to respond. by looking at someone can you say i think it's 10%, 20%? it's difficult. introduce ancestry informative marrs brings a greater granularity to studying race. studying lung cancer we're

completing an ancestry informative marker of every sample. not to belabor the point but to give you an example of how it's so important. smoking, we always do because it's so important, we'll ask if you're a never smoker, former

smoker, current smoker, important information. if you tell me you're a current smoker that's correct information. it's not that it's incorrect but there's a lot of granularity. i could be a current smoker and have smoked 30 pack years of my

lifetime. that would be a pack a day for 30 years. you could be a former smoker and since they are not as high risk, also have smoked 30 pack years, you give up a year ago or two years ago, so if you don't also collect information on pack

years and amount of cigarettes people smoke you lose a lot of very important information. so that's why i'm glad to see there is a move in the field as a whole to incorporate more of this type of research into the questions we ask. i want to mention one other

point with regards to screening and lung cancer. the main reason for this, something we talked about earlier, this year low dose ct screening was approved as a screening tool to reduce mortality of lung cancer. it's been shown to reduce

mortality by 20%. a huge potential public health and very, very -- it has a huge potential to save a lot of lives from lung cancer. it's only at the moment approved for high risk smokers, high risk individuals. one of the criteria is you have

to have smoked more than -- oops, excuse me -- more than 30 pack years, okay? which is a high risk smoker. they smoked one pack a day for these are the highest risk people. what you remember earlier, we showed most african-americans

diagnosed with lung cancer don't have a 30 pack year smoking history. they have a 10 pack year smoking what does that mean in terms of screening? well, if you were to look at it, there's the potential you could miss 40% of lung cancers in

african-american if you use screening guidelines compared to 25% in european-americans. why is this important? right now i showed you an equal access to care session, there are no disparities in lung cancer outcomes. that's a good thing.

if we don't pay attention to factors like this and new screening guidelines and don't try to tailor them based on biology and epidemiology we see, we could actually have an unintended consequence of actually widening the that's just something to keep in

mind and not just for lung cancer but also for other types of cancer as well. the other thing i wanted to come back to is we mentioned earlier about this observation that for some cancers there is an increase -- or an earlier diagnosis i should say among

certain populations, again important from the screening the reason is that when screening guidelines are given they define age range, risk factor exposure range, and if age is one of those exposures for a particular population, it raises the question, do the

guidelines need to be population specific. one other thing to mention as we move forward, many of you will be aware of tremendous advances in biological targeted therapies, they have increased the survival for a lot of i mentioned to you earlier we're

aware of biological differences between the tumors, even at the moment for cancers like lung where it doesn't seem to contribute to difference in outcome it does for breast and there could be another unintended consequence. some studies beginning to

suggest for example carcinoma and most myeloma, there's been targeted therapies, and specific because there are differences in biology, it's highly expected there would be differences in the responses of those particular tumors. so again many of these studies

are more likely to be done in populations of european descent because 72% of population is of european descent but it's important in these studies to also consider will this drug be affective in populations of d to ask the question based onother races and ethnic groups cell biology do we expect to see

similar responses. if we don't, consider that in the therapeutic and clinical setting. also importantly, in the research setting. just to sum up before i take questions, we know there are several drivers for differences

in disparities, including geography, genetics, tobacco. this is by means not an exhaustive list, not possible to go through all of them, also it's important to say there are factors we don't know, that's why research is actually so important in this area.

in relation to the disparities in mortality, we look at things like lack of early detection, access to care, lack of timely and aggressive treatment, genetics, as playing a role, as i hope you get the merge the determinants is a

multi-displenary problem that requires a multi-disciplinary research and that's why this research is important, because we have access to experts in all these different areas. it requires a lens that looks at prevention, early detection, diagnosis, treatment, and

and there is no one factor which will explain the disparities in all cancers, no one solution that will be able to address for this particular field, it's important to work and look at these things through a multi-disciplinary approach. in our study, in our lab i

should say, we use the maryland case control study to achieve this, examine contribution of environmental and inherited factors to the excess cancer burden on african-americans, i focus on lung cancer primarily as i've mentioned, and stephan on breast and prostate.

we're always available if anybody has interest or questions or follow-up after this, if anybody is interested. with that i thank my lab and those involved in the maryland case control study and happy to take questions you might have.

yeah? >> the question was, do the mean risk loci we're aware of, for breast an prostate, come from gwas? they definitely do. they were initially discovered in populations but because

breast and prostate affect prevalent disease it was possible to build power to look within african-americans, and indeed the population (indiscernible). >> (inaudible). >> there have. so many of you would with be

familiar with the cancer genome atlas project, within those there are some samples that are from african-americans, but one of the biggest problems there is power. that's why studies like what i'm doing at lung and stephan in breast, we're doing sequencing

on those ourselves because we have a -- using this case control studies, because it's in the baltimore region we have an overrepresentation of african-americans, we're able to ask the questions. there isn't enough data within the (indiscernible).

great, thank you very much for your attention.