BERKELEY, CA (UroToday) -
Understanding and Implementing Immunotherapy for Prostate Cancer
Prior to the beginning of the interview, Dr. Charles Drake prefaced the conversation by explaining that the idea behind immunotherapy is that the body’s own immune system can be enlisted with the goal of putting cancer into remission, to bring about actual shrinkage of tumor cells, or in the absence of that, have cancer stop growing, develop stable disease, or even just slow down the growth of disease. “That is the general concept, and there are a multitude of really good reasons why this makes sense. We have to remember that one of the reasons why cancer defeats us is that tumor cells evolve over time, and they retain the capability of adapting to pressures of the environment, particularly to agents such as chemotherapy. They develop new mutations, and, thus, ways to escape from most of our established treatments. The difference with immunotherapy is the immune system ‘understands’ that; the immune system was designed to respond to pathogens, which behave similarly. So, as the pathogens change their structure, or their molecular structure, the immune system can follow and hopefully get ‘ahead of the game.’ That is why we think immunotherapy is an attractive approach to an evolving disease like cancer, and that is the rationale for recruiting the immune system, or immunotherapy, in the fight against cancer.”
Q: In immunotherapy we talk about re-programming the T-cells, is that applicable to prostate cancer?
A: Yes, I think the idea about re-programming T-cells is applicable in essentially any cancer. I think that we have to be a little bit cautious at this time in our scientific understanding of exactly how we describe these sorts of manipulations. In some cancers we have very good evidence -- in melanoma, in kidney cancer -- that there are T-cells within the tumor and they generally show some specificity for the tumor, but they are just not working. In that kind of a setting, what we think is blocking molecules that are turning T-cells off is a good strategy. This is born out of a series of clinical phase I data using anti-bodies that block these molecules, the check-points that turn the immune cells off, actually. So in cancers that already have tumor-specific lymphocytes, I think what we can do is we can re-program those cells, at least in certain patients and with certain manipulations, to then change their minds and then do what we think they were born to do, and that is to kill tumor cells. In prostate cancer, to be fair, the data for this are less robust. When these agents were tried as single agents in prostate cancer, the rate of objective immune responses was smaller, so re-programming prostate cancer might be a little more challenging than in other cancers, like kidney cancer or melanoma, or perhaps even lung cancer. The other thing we need to keep in mind is that there is another way of doing this instead of just re-programming T-cells -- we might actually just need to program them in the first place. When we talk about things that program T-cells, we are talking about the idea of raising T-cells that are specific for the tumor, so we are talking about inducing an anti-tumor immune response. Manipulations that rise, or initially program a T-cell, are different than the ones that re-program, and usually when we are talking about programming a T-cell response, we are talking about things like vaccines. In some cases we might actually need to do both. So we might need to use a vaccine to program, or start, or induce an anti-tumor immune response. Those cells have a chance of working. In clinical trials these are manipulations that can sometimes induce a stable disease or, often, in the aggregate, slow down the growth of tumors. But to really get these cells to fully function, we need to change this so the program will be started, the tumor will tend to turn that program off, and then we will need to later on re-program those cells by blocking immune check points. I think all these concepts are applicable to basically all cancers in general, frankly, but prostate cancer might require multiple steps, both the programming, but also the re-program, to see the maximum benefit of immunotherapy in that setting.
Q: What are the optimum conditions for delivering immunotherapy in a patient with prostate cancer?
A: It is a great question. We have to think about this in two ways: we have to think about this in terms of our current clinical reality but also in terms of the future. So in terms of current clinical reality, there is one FDA approved immunotherapy for prostate cancer, which is Sipuleucel-T. This is a personalized cellular vaccine product that is administered to patients with metastatic castrate-resistant disease, and the reason it is administered to these patients is because that is where it was FDA approved. So basically Sipuleucel-T was shown in multiple randomized controlled clinical trials to extend the life of men with metastatic castrate resistant disease by around 4 months on the average in the patients tested. Among those patients, if we do sub-group analyses, which is always a little bit dangerous but it is something that we often do to see where these kinds of treatments will work best, it turns out that if you look carefully, the patients who were earlier in their disease course are the ones who seem to have most benefit. By that I mean men who were asymptomatic, men who have less of a disease burden, men who have a lower PSA and so on. That is the current reality but if you take that to its logical extension, maybe we do not want to wait until the men are metastatic and castrate resistant to give immunotherapy, it might have a chance to work better in earlier disease stages. Clinical trials have been initiated in that space and we were fortunate to participate in a trial of men with biochemical recurrent disease which is a much earlier stage. I need to point out that this is not the FDA approved indication for the current immunotherapy but certainly that is where we are thinking about going. In summary, in the current time immunotherapy is approved and we search for men who have early stage, metastatic castrate resistant disease. We honestly need to catch them as early as we can, at the first sign of metastasis in the metastatic castrate resistant patient, that is when we start to talk about whether they are interested in immunotherapy or not. But in the future, I am not sure that is going to be the optimal way and we will see how these things will play out over the next several years.
Q: When you look at patient selection, is there any aspect of patient selection that you would think about in terms of the sequencing?
A: This is another interesting question. The one thing we should really be clear about, and one thing that is tough, is that it takes the immune system a while to mount a response. We are not talking about years but it is in the range of months so if a patient has a really rapidly progressing disease it is probably not a good idea to give them the immunotherapy first because they are going to wind up needing something else sooner. That could be hormonal therapy with abiraterone acetate that comes with steroids as we mentioned earlier, or even chemotherapy with docetaxel that also comes with steroids. This is going to tend to turn the immune system off so we do not want to find ourselves in a situation where we are treating somebody with a very rapidly progressing disease. We can get the immunotherapy in, it really takes about 6 to 8 weeks total, and then we immediately have to give them something else, either chemotherapy or hormonal therapy. Instead, we want to have a patient with more slowly progressing disease so we have time to allow the immune response to consolidate before we got to do something else. In addition to the standard things, asymptomatic, minimally symptomatic, lower disease burden, or lower PSA, we are also taking a look at the pace of the disease to see if that is overwhelming. This is based on logic, it is not based on prospective clinical data and we do not know whether this would be the case until we carefully tested it prospectively in a clinical trial. But we think, once again, it is better to have some time to allow the immune system to consolidate before we are forced, either by the patient or by the patient’s disease, to move to the next step. These are all clinical decision making considerations we are thinking about, but it could turn out that immunologically there could be some biomarker, for example levels of certain cytokines or something else about the patient’s immune response, that is a lot more definitive. This would give us more data to see if this person is a good candidate or not a good candidate. Some of these correlative studies are currently underway at this time and we expect to see results over the next couple of years. For now we have the clinical sense of the patient. In addition to the standard things, I do not prefer, and I think my colleagues would agree, to use immunotherapy in patients with really rapidly progressing disease when your hand would be forced to do something else soon after that.
Q: How do you know when it is working, are there any indications that tell you whether the Sipuleucel-T or the immunotherapy you are using in these patients is actually working?
A: That is a great question and that is one of the challenges that our current immunotherapy faces. Sipuleucel-T is an autologous cellular product, it is very similar to a vaccine. In the 3 randomized phase III clinical trials of this agent, objective responses, i.e., tumor shrinkage were very rare. It was certainly less than 5 %, not unheard of but very, very rare. What we think this means is that sort of immunotherapy sets up a new state of equilibrium where the tumor progresses but perhaps progresses less quickly than it would have in the absence of immunotherapy and what you wind up with is a difference in median overall survival that is quite statistically significant and also meaningful for patients -- especially in light of the fact that this kind of immunotherapy has a low rate of side effects. The more simple answer is that if they do not have an objective response, which is very rare, it is hard to tell whether it is working. One thing that I should point out is that when the pivotal trials of Sipuleucel-T were performed, the field did not have quite as good a sense that PSA kinetics, or PSA doubling time, was something we were using to predict outcome or use to make clinical decisions, so during those trials, PSA data were not carefully gathered. It could be that if we look at these a little bit more carefully, what we might see is a blunting in the rate of rise of PSA in patients, corresponding to a slowing down of disease progression. That is one thing we look for in our clinical patients and it is obviously easiest to assess when it is your own patient and you have a lot of PSA data over time. It is a lot harder when you wind up with a patient coming from the outside. This is one thing we can look for in these patients moving forward. I should also mention that in a clinical trial with a DNA vaccine against prostate cancer, conducted by my colleague Dr. Doug McNeil from the University of Wisconsin, he in fact demonstrated this quite nicely -- so patients who did get this DNA vaccine did show a slowdown of the rate of rise in their PSA that did seem to correspond a little bit with outcome. In standard clinical practice you give the patient sipuleucel-T and it is not like when you give them hormonal therapy, you do not see large effects of tumor shrinkage in most patients, so it is a little harder to know that in the real life situations that we encounter now.
Q: Finally, as we look forward, what are the challenges in translating into clinical practice the research discoveries that we spoke about earlier in terms of what we see in the animal?
A: There are a number of challenges that inhibit the translation of both the animal work and even the clinical work that we have now. The first one is that the training we have as either urologists or medical oncologists is not very heavy on immunology. Most people get a week or two week experience in medical school and that is all. Therefore, understanding how these agents work is a little bit of a barrier to using them because usually people feel more comfortable giving drugs, about which they understand how they work. For example, in medical oncology training we spend a great deal of time understanding how various types of chemotherapy work, and we do not spend anywhere near that time on immunotherapy, at least not yet. One big challenge is a global unmet need in the field to have a better understanding of how these therapies work. The second major challenge is one we were just speaking about, the fact that vaccine therapies generally do not wind up with objective responses and there is some resistance on the part of medical oncologists, and perhaps among urologists as well, based on that. In prostate cancer specifically we have a fairly high bar in that hormonal therapies usually do produce PSA responses and objective responses, so compared to that, an extension of survival with minimal side effects but without any objective tumor response is sometimes a little bit harder to understand by some physicians and urologists. I think that the third challenge is that in animal models, and even in retrospective data in people, it is fairly clear that using immunotherapy earlier in the disease course is better, and to be quite frank, metastatic castrate-resistant disease is certainly not early stage prostate cancer. This is one of the later stages of disease so if we could move towards a paradigm where we were able to give the drug to men in earlier disease stages, for example castrate resistant non-metastatic, or maybe even to take a step earlier to men who just have a rising PSA in the absence of metastatic disease and who have not been treated with hormones. These are men who have so called biochemically recurrent disease, who would probably be associated with a greater clinical benefit. The challenge in that, though, is in moving these therapies earlier, the men who do well tend to live a fairly long time, so trials in that space with an overall survival endpoint are quite simply impractical. What we would need is trials in earlier disease stage patients with meaningful clinical endpoints, like time to metastasis or time to obvious progression, or even time to PSA progression. Frankly, time to PSA progression is a fair endpoint because that usually means you are going to give another therapy, either hormonal therapy, usually, or perhaps another therapy. I do think that those are clinically meaningful endpoints, but working immunotherapy earlier into the disease course for prostate cancer is going to be a challenge in translating the clinical data and optimizing the efficacy of these sorts of treatments for men with the disease.
____________________________________________________
Charles G. Drake, MD, PhD
Associate Professor, Oncology, Immunology and Urology
Johns Hopkins University School of Medicine
Baltimore, MD USA
Charles Drake received his PhD in Immunology from the National Jewish Center for Immunology and his MD from the University of Colorado Health Sciences Center. After completing an internal medicine residency on the Osler medicine service at Johns Hopkins, he entered the field of medical oncology. During Dr. Drake’s oncology fellowship, he began to investigate the immune response to prostate cancer, developing a unique murine model to perform studies on antigen-specific T-cell tolerance to tumors. Experiments in this model supported the concept that androgen-ablation could mitigate tolerance to prostate cancer, creating a window during which tumor vaccination could prove successful (Cancer Cell 2005).
Currently, Dr. Drake is the co-director of the Multidisciplinary Prostate Cancer Clinic at Hopkins, where the combination of androgen-ablation and vaccination is being examined in a translational, pre-surgical trial. Using additional in vivo models, the Drake laboratory showed that LAG-3 is relatively over-expressed on non-functional T cells, and that LAG-3 blockade affects T-cell function in both Treg-dependent and Treg-independent mechanisms (JCI 2007). More recently, the laboratory was able to show that blocking the immune checkpoints PD-1 and LAG-3 could prove synergistic, in several in vivo cancer models. Ongoing work in the laboratory is focused on understanding the role of LAG-3 in regulatory T-cell induction, as well as the regulation of immune checkpoint molecules and ligands in prostate and kidney cancer.
