Expert Discusses Promise of Immunotherapy in Genitourinary Cancers

Charles Drake, MD, PhD

In the treatment of genitourinary (GU) cancers, immunotherapy continues to move toward the frontline setting. Charles Drake, MD, PhD, says he is optimistic about the future of immunotherapy, as trials are showing promising results in various GU cancers, echoing trials in lung cancer.

Immunotherapy is showing progress primarily in bladder and kidney cancer, but researchers are beginning to turn to prostate cancer, as well.

“Now that immunotherapies are approved in later settings, there is a lot of enthusiasm to move these earlier. In prostate cancer, we are very excited to try some trials for men who have rising PSA after surgery and to give immunotherapy either alone or with hormonal therapy to try to reset their whole course of disease. We're seeing this in kidney cancer and bladder cancer where immunotherapy is being used in the first line, and we'd like to do the same in prostate cancer,” says Drake.

OncLive: What are some key recent developments with immunotherapy in GU cancer?

In an interview with OncLive, Drake, director of Genitourinary Oncology at NewYork-Presbyterian/Columbia University Medical Center and associate director for Clinical Research at the Herbert Irving Comprehensive Cancer Center, discussed pairing immunotherapy with chemotherapy, the efficacy of current immunotherapy agents, and the future use of immunotherapy in the treatment of GU cancers.Drake: The big developments are mostly in bladder cancer. In bladder cancer, the anti—PD-1 antibody pembrolizumab (Keytruda) showed level 1 evidence for activity. At the 2017 Genitourinary Cancers Symposium, we saw randomized data from a large trial comparing pembrolizumab to chemotherapy in second-line bladder cancer. This will certainly lead to this drug being used widely. Its joining a drug that is already out there—the PD-L1 inhibitor atezolizumab (Tecentriq)—and the other PD-1 antibodies can be used in this setting as well. By the end of the year we might have even another PD-L1 antibody, durvalumab. It’s fascinating, by the end of the year bladder cancer will have multiple treatment options with immunotherapy—I think it is moving into the mainstream.

In prostate cancer, I think the significant development is that we are starting to see signals. Pembrolizumab has shown activity in prostate cancer, in the context of patients who responded to hormone therapy, and now they are progressing on enzalutamide (Xtandi). Also, if you add anti—PD-1, some of them respond.

Will immunotherapy pair well with traditional therapies?

It is interesting, though, because these are very different settings. The bladder cancer setting is like phase III randomized data, "Here we go!" While prostate cancer has this early indication of activity. To me, these were the most significant developments in immunotherapy.In my lab, we spent a lot of time modeling the combinations of immunotherapy with chemotherapy in mice. And you can do it, and you can get an additive signal—usually not synergistically though. It was tricky, though, you had to give lower doses of chemo, and the timing was absolutely critical. For example, low-dose cyclophosphamide given 1 day before immunotherapy—when you add that to immunotherapy it is actually really nice, it’s additive and the low dose of cyclophosphamide kills these suppressive cells, these regulatory T cells. But if you gave full-dose chemotherapy on the same day as the immunotherapy it was really bad, actually—neither of them worked particularly well. You don't really lose much of the chemo efficacy but it was not a beautiful thing.

In humans, this is not what happens, actually. It is really interesting—in lung cancer, clinical trials were done that I originally thought were, frankly, kind of stupid experiments. The idea was that they gave full-dose chemo with immunotherapy and kind of hoped for the best. And honestly, the best sort of happened—the response rate was much, much higher. It looked like the immunotherapy was adding to the response rate of chemotherapy, so progression-free survival was really improved. The problem with this is that we do not know whether that is going to translate into overall survival. So, the question is, do you use up all of your bullets in the beginning of the game, get beautiful responses and a long progression-free life? If so, will that translate to the patient living longer in the end or not? Nevertheless, I was completely wrong, it turns out that chemotherapy and immunotherapy—particularly checkpoint blockade—work reasonably well together.

Because of that, experiments are being done in bladder cancer now in which traditional chemotherapy regimens are given upfront with immunotherapy. And based on the lung cancer data, I'm actually sort of optimistic. I think we are going to see higher response rates and increased progression free survival. The question remains whether we'll see better overall survival or not. It is very interesting, I never thought this would happen.

What immunotherapy agents are looking the most promising?

What aspects of immunotherapy should community oncologists be paying particular attention to?

Another example that we thought would work well together is hormonal therapy. Hormonal therapy in prostate cancer leads to the glands sort of involuting, and there is an influx of activated lymphocytes that beautifully accompany that involuting. So, it does look like immunotherapy pairs well with hormonal therapy in prostate cancer. We did a trial where we did these things before surgery and it sort of worked. From the hormonal therapy, we see a nice T cell influx, a nice immune-cell influx, and the immunotherapy seems to add to that. It is a great question—which drugs are better? It is very interesting actually—if you look across the bladder data that we have so far from 4 different drugs: 2 anti—PD-L1 antibodies, atezolizumab and durvalumab, and 2 anti–PD-1 antibodies, nivolumab (Opdivo) and pembrolizumab. They all look very similar—response rates are 15% to 25%, grade 3/4 adverse events are similar at about 15%. There is a suggestion that the anti–PD-L1s might be a little bit better tolerated, but these are cross-trial comparisons that are inherently flawed. It is actually stunning—none of them are really distinguishing themselves, particularly in bladder cancer where we have the most data on multiple agents. When we give immunotherapy, the rate of adverse events is lower than chemotherapy, which is a good thing. There are severe grade 3/4 events but they are different. When you give chemotherapy you're looking for neutropenia, infections, and nausea and vomiting. The side effects of immunotherapy can be very, very subtle and they can come on very slowly. The ones that can be dangerous are colitis, which can come on with just a little bit of diarrhea, increased bowel movement—but it can proceed to something that is life-threatening, and there have even been fatal colitis cases in some of these trials. Other things such as inflammation of the lungs and pneumonitis pose questions of whether this just a cough, the flu, or, it could indicate the cancer progressing.

The take-home message I want to provide is if you suspect that a patient is having an autoimmune adverse event you need to do 3 things. The first thing is that you have to work it up with imaging studies and appropriate labs—you can't just ignore it and hope it is going to get better.

The second thing is you must manage it. Keep in mind, though, that oncologists are a little afraid of steroids because we think steroids are going to turn off the antitumor immune response. And you know what? That is completely untrue. In a large case series in melanoma, patients with autoimmune adverse events, if they were managed with steroids and the tumor was shrinking, it always kept shrinking. So, I guess the other message is, do not be afraid of steroids.

Is there anything else that you are working on, or anything developing in the space that you would like to comment on?

The third part of that message is that there are really nice algorithms that talk about this, and if you follow the algorithms and do what they say, you usually don't get into trouble. The times we see people get into trouble is when they are slow to use immuno-suppressive drugs when they are needed. And the other times, it is not the doctor's fault; it is actually the patient not reporting symptoms. You need tell them over and over, "Listen if you have multiple bowel movements, you need to call, or if you develop a new cough or shortness of breath, you need to call." If a patient develops crushing fatigue, it can be an indication of inflammation in the brain and they need to call their doctor. We try to educate them the best we can, but it doesn't always work out. Two things. First, there was a really seminal paper that was published this year in breast cancer that I think is going to guide some of our clinical trials in the future. In models of metastatic cancer, the question is always whether you do surgery for the primary tumor or not. So, in the study, there were models of mice that were given metastatic cancer, and if you took out the primary tumor and gave chemotherapy after the surgery, it really extended life only a tiny bit. But, what was beautiful and fascinating about this paper was that they also tried the same experiments with immunotherapy. The idea was that if you take out the primary and give some immunotherapy, maybe you could help these animals live longer—so, they were testing adjuvant immunotherapy. And this is something that is happening. There are at least 3 or 4 adjuvant immunotherapy trials in kidney cancer, there are some in bladder cancer, and there have been ones in melanoma.

It turns out that adjuvant immunotherapy in this setting worked a little bit better—it helped the animals live a little longer, but they were not cured. Conversely, if you take mice with metastatic breast cancer, you give them some immunotherapy and then take out the primary tumor, the still have cancer. But for some reason, that combination of neoadjuvant immunotherapy plus surgery led to about half of the animals being cured with multiple models and multiple kinds of immunotherapy.

I think this paper suggests to us that when we see patients with metastatic cancer, the kinds of trials we should think about doing is giving some immunotherapy first, resecting the primary tumor and seeing if we can make a difference in terms of overall survival. The other part about those trials that is very advantageous is that you get to see the primary tumor, and by looking at the cells and doing molecular analysis, you can determine if you activated the immune system and if you did not, why not. I think that is one of the biggest take-home messages of the entire year—that we should really think a little more strongly about neoadjuvant trials of immunotherapy in patients with metastatic cancer. And that is a new thing, that is not really being done that commonly.

The other thing is now that immunotherapies are approved in later settings, there is a lot of enthusiasm to move these earlier. In prostate cancer, we are very excited to try some trials for men who have rising PSA after surgery and to give immunotherapy either alone or with hormonal therapy to try to reset their whole course of disease. And we're seeing this in kidney cancer and bladder cancer where immunotherapy is being used first line, we'd like to do the same in prostate cancer.