Challenges in mCRPC Landscape Surround Sequencing, Biomarkers, and PARP Inhibition

Tanya B. Dorff, MD, discusses several important components regarding metastatic castration-resistant prostate cancer treatment, including the addition of novel drugs, optimal therapeutic sequences, and the identification of predictive biomarkers.

Tanya Dorff, MD

While progress has been made in the management of metastatic castration-resistant prostate cancer (mCRPC) with the addition of targeted therapies to the armamentarium, there are still several unresolved issues, according to Tanya B. Dorff, MD.

“[There are] 3 big questions that remain for how to manage mCRPC,” said Dorff. “One is whether we should continue agents beyond progression. Another is whether there are any biomarkers that can be used to help us decide between subsequent choice of therapy. Then, the third is whether there is any sort of optimal sequence that can be discerned.”

OncLive: What agents do you currently look at when a patient’s disease progresses?

In an interview during the 2017 OncLive® State of the Science SummitTM on Genitourinary Cancers, Dorff, an associate professor of clinical medicine at University of Southern California, discussed several important components regarding mCRPC treatment, including the addition of novel drugs, optimal therapeutic sequences, and the identification of predictive biomarkers.Dorff: The questions are how to best use the tools we have, which include abiraterone acetate (Zytiga), enzalutamide (Xtandi), docetaxel, cabazitaxel (Jevtana), radium-223 dichloride (Xofigo), and sipuleucel-T (Provenge). It has obviously gotten a bit more complicated as Dr Guru P. Sonpavde said in his talk [during the meeting] about upfront intensification…We would approach the patients with single sequential therapy, and, typically, we might want to use immunotherapy earlier rather than later.

Are there biomarkers that are currently being explored for this population?

Some of the newer agents, including immunotherapy, such as pembrolizumab (Keytruda), or PARP inhibitors are in clinical trials. They are in clinical trials that are early, so they are often combined with abiraterone acetate, enzalutamide, or one of our existing drugs—or they're being compared against one of the existing agents. If patients are going to access some of the more novel therapies beyond what we already have, they need to be referred earlier for clinical trials.There are 2 major biomarkers that I discussed during the meeting. One is the PARP inhibitor story with the DNA-repair deficiency. There is increasing recognition that this is common. We should probably be testing for germline mutations in our patients with metastatic prostate cancer, since between 10% to 20% of them will harbor germline DNA-repair deficiency. Obviously, that’s a lot of patients to be testing, so some of us are selecting patients on the basis of family history.

Then, there are somatic mutations in those pathways, as well. Those seem to be strong biomarkers that predict for potential benefit from PARP inhibition. Although, in an interesting plot twist, it looks like they will predict for response to androgen receptor (AR)-targeted therapy, as well. Whether it is just prognostic or whether it is the crosstalk between AR and PARP is an evolving story, but it’s becoming increasingly clear that we need to be screening patients for things like BRCA and other DNA-repair deficiencies.

The other major biomarker, of course, is AR-V7, which has been a really interesting rollercoaster story. It looked initially like AR-V7 was really “it.” It could predict whether someone was going to benefit from abiraterone acetate or enzalutamide, or whether we ought to go to chemotherapy. Then, these studies started coming out that showed that patients with AR-V7 expression were responding to enzalutamide.

What factors do you look at to consider the optimal sequence of therapies?

I highlighted a clinical laboratory improvement amendment-certified assay in California. There are probably other places you can get the test, and we might use it if you want to choose what's next for the patient. Maybe you would use chemotherapy next, but I wouldn’t use it to say, “This patient should not ever get enzalutamide or abiraterone acetate,” because then we are potentially denying those patients what could be effective therapy.It's a tough question. There are often 2 good choices for an individual patient, but there are some patients for whom I might prefer abiraterone acetate that might be an elderly patient who is frail, maybe has had some falls, neurologic issues, or has had weight loss, because those are all side effects that we can see more with enzalutamide. I might start with abiraterone acetate for them.

If apalutamide gets approved, how could that agent fit into the mCRPC paradigm?

On the other hand, for someone who has fragile diabetes or has a history of a bad reaction to prednisone, that might be someone for whom I would prefer to start with enzalutamide even though that prednisone level is fairly low. It's not supposed to be super physiologic; you do see some of the prednisone side effects. That’s a conversation that definitely some patients have strong opinions about when you bring that into the discussion.Well, it will depend a bit on how apalutamide’s approval comes through and in what context. Assuming it comes through for biochemical recurrence—for example, the SPARTAN trial— then the whole paradigm is just going to shift earlier and earlier. We’ve gone from using these life-extending drugs in the castration-resistant setting to now upfront for metastatic castration-sensitive setting. Moving it into the M0 setting is a continuation of that.

Are there any next steps that we're looking to take with radium-223?

It will be interesting to see the side effect profile and what the magnitude of benefit might be. Not every patient can be an optimal candidate or choose to start on therapy that early. We also won’t have a sense, from the early data, of what the long-term consequences might be from early and intense androgen-deprivation therapy with the novel agents. Of course, we are all excited to welcome a new tool but, just like with the other drugs, as there have been new data and new ways to use these tools, there are also still going to be some new questions.Radium-223 is interesting because we don't actually know so much about what it does in the marrow. We know it’s an alpha-emitting, calcium-mimetic [agent] similar to strontium-89 and samarium-153 of the past. We know it acts in bone-only [metastatic disease] and that is where prostate cancer affects most of our patients. We use it as a life-extending therapy just like we would use abiraterone acetate, enzalutamide, or docetaxel. It is a nice alternative for some patients who aren't really a good candidate for docetaxel or who are resistant to that treatment because the side effect profile might be a little bit more acceptable. But, we have a lot to learn.

Looking to the future, what potential do you think immunotherapy and PARP inhibitors are going to have in the field?

We are going to be looking at the bone marrow. What is happening is that there is immune activation that might give us some clues as to whether there is synergy with immunotherapy. There are also clinical studies asking whether the combination of radium-223 together with abiraterone acetate or enzalutamide may be advantageous compared with abiraterone acetate or enzalutamide alone—so a combination versus single sequential therapy, which is the current paradigm. Hopefully, from those studies, we will learn a little bit more about how to use the agent best.It looks like immunotherapy and PARP inhibitors will definitely benefit a subgroup of patients with prostate cancer. That is really the key—identifying who those patients are. We saw with ipilimumab (Yervoy) that there were definitely responses with mCRPC. Then, the phase III trial was negative because it was an unselected population. The pembrolizumab story seems to show that it's effective in patients who are progressing on enzalutamide, in which you have PD-L1 upregulation. That's a smarter course for development—to look at, hopefully, an enriched subgroup.

Similarly, if we tested the PARP inhibitors in an unselected population, they would undoubtedly fail. We know that the presence of a DNA-repair deficiency mutation is a strong biomarker predicting response. The questions with the PARP inhibitors will be, in that selected population, “Which agent? In what sequence? Early or later? Should it be combined with androgen-targeted therapy?” Those are some of the questions that are being asked in numerous ongoing clinical trials with PARP inhibitors.