Dr. Howard I. Scher Video (Text Version)
Basic Science to Phase III Trials: Abiraterone: Translating Biology Into Effective Therapies
Dr. Howard Soule
It’s now my pleasure to introduce our last speaker of the session, Dr. Howard Scher who is the Chief of the Genitourinary Oncology Service at Memorial Sloan-Kettering Cancer Center. Just—just I know we’re running a little bit over but I have to tell at least one story. When—when Howard took this job as running the—the consortium through Sloan-Kettering’s coordinating center activity many years ago, even before the DOD came along, he was told by the Physician-in-Chief Bob Wittes that he needed his head examined. Well I was at Sloan-Kettering last week reviewing a [November] project with—with Dr. Scher and some of his colleagues. Dr. Wittes stopped by to say this was the best thing that—that ever happened to Sloan Kettering. So it’s amazing what a few years of difference makes and if I were a patient I would be very thankful to many of the doctors in this room, but if not for the tenacity, the organization, and the skills of—of Dr. Scher. A lot of the drugs that are now coming to patient availability would be lagging behind by months and some cases years and others may never have seen the light of day at all; so Howard, it’s a pleasure.
Howard I. Scher, M.D.; Chief, Genitourinary Oncology Service, Memorial Sloan Kettering Cancer Center
Well it is indeed a privilege to present what is truly a collaborative effort and you can say I pay tribute to my wife, but looking around I’ve been surrounded by some phenomenal women to work with, Dr. Kaime and Dr. Best, Dr. Hussain, and Dr. Higano, so they keep me in line and a special thanks to Jake Vincent who has really helped in terms of the organization of the consortium and has a skill set that we simply are not taught. So I’ll address disclosures; rest assured this is not the way to make a living. But it has given us an opportunity to see a number of very exciting products. And if we look at the last year alone we’ve seen three trials hit their survival end points which was previously unheard of to the point of fact where sponsors are now looking at prostate cancer first, recognizing that this is a very fruitful area and an important area for therapeutic investigation.
So just to place things in context, we’ve looked through 2010, the approval of programs in the cohort of patients who have progressed that have castration-resistant disease, but are not ready or not interested in starting chemotherapy. We had a—a second cytotoxic agent approved and we do anticipate with the survival benefit—I’ll discuss with Abiraterone—that this too will be approved shortly.
If we think about what we do with hormonal therapy, we are blocking or depleting the level of androgens. This will cause—cause PSA levels to decrease and we could ask why this patient did the PSA not go to zero? We can see that in the past we would find castrate disease is 50 nanograms or less and if you heard Dr. Mohler’s talk we now know that this can go down to less than 1 nanogram. As we see in most patients when they begin to progress, the PSA will rise, and one of the questions we could ask immediately is just as the PSA continues to rise are these tumors in fact hormone refractory and I think there will be ample evidence to suggest that this is not.
So how do we start to dissect and understand some of the mechanisms of progression? The late William Gerald conducted a study of patients who were treated either just with surgery by course of neoadjuvant hormonal therapy or patients who had progressive disease, post-hormone treatment—that is the metastatic. And what he was one of the first who observed that there was an increase in the androgen biosynthetic machinery which we—as we now know is—can cause an increase in the levels of androgen within the tumors.
There’s the second very common alteration that occurs that is the level of the receptor will increase and as shown by Charles Sawyers, this too can contribute to progression. But this was the first observation of the androgen biosynthetic increase. This is a non-castrate; this is one of the rate limiting enzymes. Here’s a metastatic tumor showing expression. Pete Nelson looked at the series and his group looked at a series of metastatic samples showing that the level of SIP-17 which is the target of Abiraterone is increased along with other enzymes in the pathway and also showing in metastatic samples that the level of androgens is increased relative to the normal prostate in the setting of normal androgens.
Now if we start looking at the pathways and I won't walk through it but one of the contributors to progression is androgens initially thought primarily from the adrenal gland; we now know this is from the tumor as well—Abiraterone will target this pathway and there’s a second agent called MDV-3100 which was developed by Charles Sawyers and his group and development started actually when he moved to Sloan-Kettering which was specifically targeted for cells with over-expressed androgen receptor. So Abiraterone will hit—decrease androgens in three locations; the testes originally developed as a so-called oral Lupron, the adrenal glands as well as within the tumor, and if you look at the levels of androgens that are achieved, again you need a mass-spectrometer in order to even measure it at this level and a major effort being led by Dr. [Inaudible] in Dr. Mohler’s group is to standardized these assays so then in fact we can reliably measure them in the clinic. So the question is do these changes in tumor biology which are more common in late-stage tumors impact on—can they affect clinical practice and how do we develop drugs?
Well if one anticipates giving a drug that will lower androgen levels, we have to be prepared for how to study it. So a number of—a number of us got together and said how are we actually conducting the trials and what end points should we use in order to assess treatment effects? So we were anticipating that there would be unique issues in the development of these drugs. The first was to be very careful how you interpret PSA levels. Sometimes they can go down when a treatment is not working; other times they go up and it can be. We know that there are issues with bone scans, so make sure you confirm what you see on an abnormal scan. Be a doctor and treat your patient; that is continue the treatment until you are sure it is not providing benefit to that patient—not simply because somebody proposes a response criteria. And when you have Phase 3 trials that are of the size that are now being conducted, you have unique opportunities to learn and test biomarkers which in turn can inform future trials.
Here’s an example of a patient who was enrolled under Randomized Cougar 301 Study. He initially started treatment after his PSA was rising. Eight weeks into treatment he had an increase in pain and a marked elevation of his PSA. We said okay; let’s check the scans. The scans were stable, relative to his baseline, so we said let’s wait and he was starting to feel better and had a transient drop. To make a long story short, he never showed progression of disease until 2 years later with no changes in scans and no changes in his PSA. Arguably this patient benefited from his treatment. But in the old days, a protocol mandated criteria would have said stop. The same issue holds for the biologic agents because we know now that the effect can be delayed for months.
So the trials with—with hormonal agents are typically conducted in the area in blue—that is in patients who have not had chemotherapy; but based on the biology of the—of what we know about these tumors that in fact their androgen axis is overactive, we proposed why not try in the atypical space, that is in patients who already had chemotherapy. The importance of this—one, there is no standard for this group or at least there was not at the time it was started, but if you’re looking for a regulatory approval it’s a much shorter track to the goal line.
Again, a series of trials were begun at the Royal Marsden by Johann De Bono and his group, Chuck Ryan at UCSF focusing primarily on the pre-chemotherapy population. We no longer argue about percentages; we just present the PSA declines as what we call waterfalls. Dr. Danila looking at patients who have had—or not had Ketoconazole. And Dr. Reid at the Marsden again looking at patients who have not had prior chemo in the post-chemotherapy space, and again it’s pretty obvious that there is very distinct patterns, significant declines in a significant proportion of patients, and we still have a group who are primarily resistant.
It was based on the post-chemotherapy data that a Phase 3 trial was designed for post-chemotherapy patients. Twelve hundred men participated; they were randomized two to one. That means you had a two out of three chance of getting the drug, Abiraterone plus Prednisone or Abiraterone plus placebo. The primary end point was an improvement in survival by 25% and in regulatory terms everything else is irrelevant if you miss this primary end point.
There was a planned interim analysis; that is, if the trial was working better than expected, one would want to stop it early and make it available sooner. If the difference was not seen we would go to what’s called the final analysis. This was all based on events—that is unfortunately patients had to succumb to their illness for us to see the difference. Again, very gratifying indeed was to see this. I can tell you when it was presented at ASCO in October I had chills seeing it as you will on the big screen. But the importance of it—this was the observation that a drug which is a hormone prolonged life in a point in the illness where hormones were not supposed to work. But in addition to demonstrating a significant survival benefit, this set a new metric for how we conduct trials by what else was studied in the context of the evaluation of the drug.
So if you start looking at—did all patients benefit; yes. You can see that the only group in whom a clear survival benefit was not demonstrated was patients who were sicker, but this was a relatively small group. You can see it—it did not matter if you had one or two chemotherapy regimens. Again to the left is in favor of the Abiraterone. It didn't matter if your tumor progressed radiographically or—or by PSA. Age did not make a difference and even so a benefit for patients who had liver or lung metastases, again areas or points—or a type of disease spread where we do not anticipate hormonal therapies will work.
This was in fact very encouraging when you—if you actually compare the adverse events that is all things that happen versus the severity, this is again a placebo plus Prednisone; you can see they are extremely well matched. There was a slightly higher incidence of fluid retention and cardiac events on the order of 1 to 2%, all of which were consistent with the known action of this hormonal type agent.
So the importance of trial design again can't be overemphasized. We’ve focused on a population that had critical unmet need and were able to show that it benefited all sub-sets. We show that the treatment was safe. Now the question became what else can you learn in the context of a Phase 3 trial that will help you for the next trial. And specifically we know we are facing the problem that—a good problem that with more treatments, it becomes much harder to prove that the next treatment is in fact superior. And if we have to wait until patients die, to me that’s completely unacceptable. So the question is—is there another approach?
So there are a number of biomarkers that you have heard about; we have a particular interest in circulating tumor cells. These are cells that are detectable, can be captured, counted, characterized at the molecular level in the circulation. We were particularly interested in these because this is non-PSA based. That is it’s going to be independent of your PSA, and we’ve already shown that whether a PSA drops or rises can do so independently of a change in circulating tumor cells.
There has been a lot in the press about this of late and I get between 15 to 20 papers per week that are discussing this topic, but in order to use a test such as this in a regulatory fashion it has to be clear—that is the performance of the test has to be shown to be consistent across laboratories and provide reproducible results. So here, there’s a surface marker which is essentially an antibody that captures the cells and it’s attached to—in this case, a magnetic fluid. Once it’s captured it’s pulled to the surface and then we look for the tumor cells based on specific characteristics—that is, it has a nucleus in pink. It’s surrounded by cytoplasm in green; it’s intact morphologically and it’s not the cells that are of particular interest to Dr. Drake, that it is not a mononuclear cell. So we get a numerical count.
We included the study of circulating tumor cells in the Phase 2 trials of Abiraterone and again of the 89 patients that were treated, 60% had what are called unfavorable counts. That is they were actively shedding cells; 40% of them stopped shedding cells afterwards. I won't go through a lot of background data; but based on that Phase 2 observation, we had two very courageous industry collaborators, Dr. Molina at Cougar Bio-Technology now with J&J, Bob McCormack at Veridex who again makes the cell search device. And again they—we were able to bring everybody together in order to—to embed this question in a 1,200 patient Phase 3 study. In this, I worked very close with Johann De Bono who is as the Royal Marsden as I mentioned where the development of the compound started. But again if you have a trial in which you show a survival benefit then you can start looking at what any biomarker is—PSA, radiographs, or in this case circulating tumor cells—how does it relate to survival.
So once we completed this trial, we’ve been in continuous discussions with FDA who have been extremely collaborative on this effort. We embedded this in subsequently the trial of Medivation; it’s also embedded in the anti-CTLA4 study, in the TAK700 study. They said okay; you seem serious. Write a briefing document and again this is a collaborative process in which data are shared over a period of time to study whether the circulating tumor cells can be used as part of a biomarker to predict for changes in survival early.
You can hear the waterfall class of Medivation 3100 very similar to Abiraterone, although it targets a different point in the androgen signaling axis; very sensitive patients, patients in whom it does not appear to be a response. What we’re hoping to learn with the circulating tumor cells is to identify the molecular reasons for these changes.
So if we summarize what we’ve learned from Abiraterone, again these outcomes clearly validate what we’ve learned from the molecular profiling studies and shows that there are many pre- and post-chemotherapy progressing tumors that are still hormone sensitive. It established the survival benefit. Just to bring you up to date, the trial of the Medivation 3100 is fully accrued and just waiting to mature, and we’re doing everything we can to take advantage of these opportunities to study other biomarkers. I think this illustrates it clearly. If you start looking at the time lines—I’ll add Dr. Mohler’s—this—his second study, first study here, but you can see we first learned about the up-regulation of the androgen biosynthetic machinery and then we learned about the increased androgen levels. This was confirmed and again up in Boston; the second study—another study in Washington showing the higher androgen levels in metastatic tumors. But if you look at this—the trials that the consortium conducted, first the trial beginning with the Royal Marsden, pre-chemotherapy at UCSF, post-chemo in Marsden’s PCCTC, PCCTC Marsden—again look at these time lines. This is almost unheard of.
And then the Phase 3(s) were accrued and folded right into the Phase 3s. This one accrued; here is the ASCO presentation with the survival benefit. And again I should put a happy face when the drug is approved, but here is the NDA filing. That was a—that was a merry Christmas indeed. But I think this is one example where the PCCTC can clearly bring needed therapies to patients faster.
I just want to thank all the members of the group and I’m undoubtedly leaving some people out. But again it was the foresight of the—of the Prostate Cancer Foundation to put—provide the glue money to allow us to do what we need to do rather than wait 2 years to try to appeal for it for areas that most funding agencies will say is unattractive. And it wasn’t until the oversight of the DoD that we really started behaving a little bit better; thank you.