Dr. Joel B. Nelson Video (Text Version)
Title of Talk: New Horizons in Prostate Cancer: Challenges and Opportunities
Dr. Howard Soule
So Don [Kauffee] wrote many years ago about an endangered species called the Physician Scientist. And I would probably take that another level and say the Urologist Scientist is probably the most endangered species of them all. So you know the number dwindling, the training program doesn’t support and foster scientific investigation by urologists like it once did. But we’re fortunate to—to have a generation of urology—urologist scientists. I think it’s the prototype to my left, Dr. Joel Nelson, who is a Professor of Urology and the Chairman of the Department of Urology at the University of Pittsburgh, extraordinarily well-trained, inquisitive, constructively critical and actually thinks about the prostate cancer patient not only at—at the localized disease level but has translated his science into the clinic for a number of treatments addressing issues related to advanced prostate cancer. So we look forward to—to your survey of localized disease and some of the issues and challenges; thank you, Joel.
Joel B. Nelson, M.D.; Frederic H Schwentker Professor, Chairman, Department of Urology, University of Pittsburgh School of Medicine
Thank you very much; thank you. I—it’s really an honor to be here and I appreciate your kind words. I think I do exemplify the physician scientist in the sense that I’m almost distinct myself, so. I’m going to talk a little bit about some of the things that we face in—in this field and I’m taking this really very much for the perspective of a clinician who is in the trenches every day dealing with men with—in most cases newly diagnosed prostate cancer and unfortunately many men growing who have failed their primary therapy.
And as we sort of look to the future through a crystal ball I think that what I’m about to talk about indicates that we really need to change what we’ve been doing because this isn't really very difficult predict what’s going to be happening here in the near future. So, June Chan who is at UCSF looked at the population in the United States and men being men there’s always fewer of us and the age groups are across the bottom there. And the dark bars at the bottom represent death from breast cancer and one of the great tragedies of breast cancer is that it actually affects women at a time in their lives when they’re young and have a lot more life in front of them.
As you can see, death from prostate cancer is shown from the hatch-bars; it’s actually very much related to age, so older men unfortunately die of prostate cancer much more commonly than younger men. And so this was sort of the state of affairs in 2005.
So this event took place in the summer of 1945 when the conflict in the Pacific ended and this activity was taken off the streets of Broadway and into the bedroom and as a result we had a phenomenon known as the baby boom. I was born in 1960 which was ostensibly the last year of the baby boom phenomenon, so you can see again this population curve as we age—there’s fewer of us. But there was this big increase in—in birth rates. And this increase—this increase has moved along to 1980 and now as you can see in 2000 the baby boomers themselves have enjoyed the pleasures of bedroom activity and now we have a baby boom echo. But in 2020 we will have gone from what is really a triangular-shaped population curve to one that looks like a rectangle. And the consequences of this—is that there are going to be very many more people alive who are elderly. And that’s partly aided by the fact that with all the statins and control of high blood pressure, death from heart disease is actually plummeting, so according to the American Cancer Society if you’re under the age of 85 you have a less chance of dying of heart disease then you do of dying of cancer, which unfortunately is still flat.
So June took these data and extrapolated them into the future and in 2025 there’s going to be a lot more people alive and unfortunately that means there’s going to be a lot more men who are going to develop prostate cancer who unfortunately unless we change things will die of prostate cancer. And in fact, at 2045, the death rate from prostate cancer in men over the age of 85 is expected to eclipse the entire death rate that we see currently. So we really need to change how we’re managing this disease or we’re going to be faced with literally a tsunami of prostate cancer death.
We’re also going to see an increase in the growth of the diagnosis of prostate cancer again because this is the disease that’s linked with age and men are living long enough to be told they have prostate cancer. So we need to do a better job of identifying the disease and frankly identifying the disease in the right patients.
And we’re facing as you know in this country another problem, which is we’re running out of money and we have a significant amount of promises we’ve made to our population about how we’re going to care for them, both from Social Security and in terms of their health care and you can see that this increased rate of prostate cancer disease and death is going to be facing a decreased reimbursement and decreased funding to care for those patients. So if—if ever we needed to do a better job of managing this disease, it is now.
So I’m going to—I didn't realize you were going to use so many analogies from war and that’s actually populating my thing. So this is a view from the trenches; these are actually some Canadians in World War I in the trenches and you can see that guy looking out of box trying to keep his head down to see what the enemy was doing. And so I’m going to give you the view of somebody who spends a lot of time in the operating room and seeing patients with prostate cancer.
And we’re constantly faced with what Willard Whitmore who was really the father of urologic oncology said about this disease and that is if cure is necessary, is it possible? And if cure is possible, is it necessary? And I’m going to focus a little more on the second part of that conundrum which is a fancy word for a pun that’s wrapped up in a riddle.
And the problem we face as shown here very nicely by Gabriel Hass’s work where he went and took prostate glands out of men who had died of another cause and carefully examined them and the number was about 164 and he found in men who did not know they had prostate cancer that very commonly you would find prostate cancer in their gland which would only be histologically evident and to meet the criteria of having a clinical significant prostate cancer always lagged behind the histologic prostate cancer and only until men were near the age of 90 at the time of their death did it appear that these two curves came together.
And unfortunately in our field, we believe that many men who we find with prostate cancer actually has the histologic version because it’s significantly more common than the clinical version.
The other thing that’s happened is there’s been a real change in how we diagnose this disease. So when I was a Resident, PSA didn't really exist yet. And so we would regularly have men who came to the hospital with symptomatic metastatic prostate cancer who had not—didn't know they had it; they just had pain in their bones. And in fact that accounted for roughly a quarter of the men who walked in the door with prostate cancer. And less than half the men had what we would consider to be a curable disease by today’s definition.
Well in the current era, and people would say this has been a great advance for our field and, at one level I would agree, that 85% or more of the men who we see with prostate cancer had clinically localized disease and only a fraction—about 2% of them actually present the novo with metastatic prostate cancer.
Well, what are some of the lessons that we have learned from this PSA era? The first is that today less than 15% of newly diagnosed cancers are deadly, so we are actually providing a significant amount of care for a disease of which the natural history is not one where it will progress to the point of death. And I’ll just use the rather crude analogy; if you look at the number of cases estimated this year it’s going to be about 221,000 and if you look at the number of deaths it’s about 33,000 and that turns out to be about a 14% of everybody with prostate cancer actually will die, little less than 15%.
Second thing is that if we take men and we just put up a—a clinic and start biopsying their prostates independent of any clinical features, we’re going to find roughly a quarter of them have prostate cancer. So in this room, 25% of us that don’t know we have prostate cancer in fact have it—if we were to just do a biopsy. And this is based on the findings of a very large clinical trial known as the PCPT or Prostate Cancer Prevention Trial; if you gave men a sugar pill and biopsied their prostate after 7 years you found about 25% of them had prostate cancer and even in men who had PSAs that were in the normal range. So it appears that PSA as a marker of disease doesn’t do a very good job.
And the other thing we know and this is now coming from growing centers where we have had the courage to actually not treat prostate cancer in men who have indolent disease is that most men with untreated prostate cancer will actually die of another cause. So Laurie Klotz has recently published a follow-up of a cohort of over 400 men who he’s observed in Canada who have had the prostate cancer with no definitive local treatment. If you were over the age of 70 in that cohort you had a 33-times greater chance of dying of another cause than dying of prostate cancer. And even in men under the age of 70, the hazard ratio was nearly nine, nine-times greater chance of dying of another cause than dying of prostate cancer.
I’m going to give you a quick 30-second history lesson. So this is a tower; this is known as a radar tower and this actually is sitting or had sat on the northeast—northeast edge of the Island of Oahu. And this is called the Opana Point. And from that tower a Private back in—one June—excuse me—one December Sunday picked something up and very carefully wrote down what he was observing on this actual piece of paper at the Opana Point. And he notified his superiors in the southern part of that island about what he was observing and they said oh, you’re crazy; they’re—you know it’s birds or there might be something coming from the Mainland. And in fact, what this young guy was picking up were the—the Japanese planes that were coming down and taking advantage of the topography of the Island of Oahu where there’s a mountain range both on the north and—well mostly on the north and he very clearly detected this invasion. And yet nobody picked up the fact from the information he was detecting that this in fact was a true positive.
And as you know these planes went around the Island and attacked Pearl Harbor and here’s a photograph of one of the moments of explosion. If you read the accounts of this event in 1940 and you look at the accounts of 09/11 they’re super-imposable. So for that generation this was their 09/11, and as you know big consequences from that. Well, I belabor this point only to point out to you that this was the beginning of a science that we in medicine have hijacked and what this is—the science is when you get signals how do you detect whether they’re true or false? And this is known as signal detection theory and you come up with what we call a receiver operating characteristic. Is this noise or is this real?
And we now have applied that to this idea of a receiver operator curve. And so you’ll see lots of us put up data where we’re trying to talk about how good a test is in terms of the detecting the presence or absence of disease.
So what would be the ideal biomarker? The ideal biomarker would be 100% sensitive. That means it would detect every single case of the disease that exists. And it would be 100% specific. Specificity means that you will detect every single case that’s free of disease and if you put these two together you basically have a perfect test.
Now I would argue that actually this ideal biomarker would be pretty bad for prostate cancer because we do not in fact want to diagnose every single man walking around with prostate cancer, in particular, those men who don’t know they have the disease and never need to know they have the disease. And the idea of 100% specificity at least in the current technology is nearly impossible with prostate cancer. Let me remind you; 25% of men who don’t know they have prostate cancer in fact have it, so you don’t have normal controls on what you really can test your marker.
So what would be the ideal biomarker for prostate cancer? Well I think the most important one is that it needs to detect those who require treatment—not those who have the disease, those who require treatment. And here’s some data from Dan Lin of the Pacific Northwest who looked at a large group of men in a national database and we would all argue that men who have Gleason 8 to 10-disease have a disease that we want to detect and we want to try to intervene. And in particular, this really ominous group of 35 to 44 year old men who present with this and we all as clinicians have one or two men we remember who show up at the age of 40 with widely metastatic prostate cancer. Had we only known that when they were 20 and could have saved them from the inevitable death that would occur?
Second ideal biomarker is that it will allow us to safely monitor observed patients. So I would argue that the majority of men told today that they have prostate cancer, at least for a period of time, probably should be observed. They should not immediately jump to treatment. Unfortunately our current tool which is PSA is really short to the task, so these are data from Al Carter’s group looking men who were on active surveillance, who they did biopsies every year and when they went from an indolent cancer to one that was potentially clinically significant they asked the question—did PSA predict this? Well this is basically a coin toss; PSA does not predict really at all or PSA kinetics who is progressing or not on active surveillance. So we definitely need a better marker than PSA.
And we need a marker that will allow us to adequately reflect response to therapy and this is particularly important in advanced disease. As you know PSA quickly gets disconnected from prostate cancer in—when take away androgen. I think one of the more exciting things that’s come out—I don’t think it’s really ready exactly yet for primetime is looking at circulating tumor cells. And in this study men were—had their circulating tumor cells assessed and then received therapy and if you converted from having a lot of cells to having no cells as shown there by the blue line, you actually enjoyed a survival rate that was very close to those men who had no cells at the beginning. And so we were—we would like to see more markers like this that allow a clinician to say to a patient, yes, in fact you should stay on this therapy because the biomarker is telling us you are responding—in other ways that we couldn’t delineate.
Well how as scientists should you develop meaningful biomarkers? I think the first is you should not underestimate the strength of known clinical features in predicting disease course. So we regularly will see people who will come up with a gene of interest and then show that it correlates with Gleason score. Well let me remind you; Gleason scoring is actually probably one of the best prognosticators we still have going, so if you’re going to do better than Gleason score you have to show how your marker is going to separate the Gleason scores that aren't lethal from the Gleason scores that are lethal.
And secondly PSA although I just vilified it, actually does do a pretty good job of predicting who is going to die of prostate cancer. So this is work from Steve Freedland where he looked at the Hopkins data and looked at PSA kinetics and if you have a rapidly rising—rising PSA your chances of dying of prostate cancer are certainly greater than if your PSA kinetics is slow.
The second thing is that I think as scientists we really need to apply our tests to the population that it needs to be applied to. So if you go out and HIV test lots of people you’re going to get lots of false positives and raise anxiety. That’s why you only HIV test people that are at risk to have that disease. Likewise, if you enrich the population you’re looking at your test will be more meaningful. One way we can do this is genotyping, so for example this work came out of Wake Forest and Hopkins where they looked at this specific genotype on chromosome 17 and they found that if you had this genotype your chances of developing aggressive disease or high-grade disease was certainly greater than if you didn't have this genotype. And although it’s quite small, the number of men in those larger bars is actually little. And that’s really what we want to do; we want to focus our efforts on those who really need to be diagnosed and treated, not the population at large.
And I think we need to try to trump clinical disease features. We need to predetermine indolence or aggressiveness. And this is some—these are some unpublished data from our institution where we looked at copy number variants at the DNA level in the primary tumor in men who didn't recur of their disease; men who recurred slowly; and men who recurred rapidly. And you’re looking at the entire genome there and you can see there’s a large increase in amplification and deletions in men who have a more extensive disease. And this clearly does much better than Gleason scoring in these patients.
We’re facing I think unfortunately a backlash against our indiscriminate treatment of prostate cancer as evidenced for example in this book, Invasion of the Prostate Snatchers, which says no more unnecessary biopsies, radical treatment, or loss of sexual potency. And I think it’s driven in part by the attack of the robot, so if you look for example at robotic prostatectomy in this—in this country it has gone up increasingly and now unfortunately it’s being increasingly applied to populations of men who historically we didn't really think were good candidates for surgery. So you can see the age of men 65 to 84 getting robotic—getting prostatectomies in general in this country has increased significantly largely because the robot is being billed as this procedure with no side effects, which is simply hogwash.
And the challenge I think for those of you who really want to make an impact on primary therapy is to recognize that prostate cancer is multifocal as shown by Bill Nelson in this paper in The New England Journal of Medicine where you can see add-mixed prostate cancer, precursor lesions and benign glands, and unfortunately the prostate cancer imaging we currently have is simply way, way too crude to diagnose these small lesions. And if we could find their location then we could apply any one of these ablative therapies sparing glands, sparing a lot of the side effects of whole gland ablation and I think that we’re just one imaging technique away. We have lots of tools to kills cells; that’s easy. What we need to know is where to aim.
So what are our challenges and opportunities? And I’m sorry I’ve gone over a little bit. I think for the scientists you need to keep your eyes on the prize. Our job here is to not be coming down to Orlando to sit in a ballroom when it’s beautiful outside. There’s no meeting like this for testes cancer, right because we basically cure testes cancer. That—we should be out of business by the end of my generation and I really think we should hold ourselves to that. Secondly and this gets back to the comments earlier; you need to know the enemy. So as—as a basic scientist in the laboratory you need to ask yourself every single day is what I’m doing actually going to make a difference, not is it actually going to get funded, not is it actually going to get into PMAS; is it going to actually going to make a difference in the lives of the people that we count on?
You know I used to say that if poetry would solve the problem of cancer then we should all become really poets, right. We just happen to do science because our job is to help people. Now if you like science for its own sake—wonderful but I would say get out of the way. Help us get people, like this man, cured of his disease.
And then finally I think as scientists you need to bear our burden, which is our patients and we are really very limited today in what we can offer our patients. We need new therapies. We need new ways to diagnose this disease and manage them and we’re counting on people in the laboratory who are going to lead the way. So that’s the last of my comments and I am supposed to introduce Dr. Sartor or are you going to do that? Okay; thank you for your attention.