Insights From: Emmanuel S. Antonarakis, MBBCh, Johns Hopkins Medicine; Andrew J. Armstrong, MD, MSc, Duke Cancer Center; Howard I. Scher, MD, Memorial Sloan Kettering Cancer Center
Emmanuel S. Antonarakis, MBBCh: The last piece that clinicians like myself need to consider is that we’ve documented that these patients do not respond well to abiraterone and enzalutamide. They may respond to some degree to taxane chemotherapy. But what other novel therapies are out there? What other novel therapies are being tested in patients who have AR-V7 expression? And I have to tell you this is an unmet medical need because these patients have very poor prognosis. On average, they develop progression, radiographically or clinically, in 3 to 4 months, and most of them have an overall survival of anywhere between 9 to 12 months. So, truly, the AR-V7-positive subset of these patients represents a very lethal clinical phenotype.
So, clinicians like myself are very interested in exploring new avenues for therapeutic strategies that could be effective in AR-V7-positive patients. One that we had hoped for was the story with galeterone, the ARMOR3 study. Unfortunately, galeterone will not make it as a therapy for CRPC or specifically, AR-V7-positive CRPC, and that drug development is being discontinued by its sponsor. However, there are a few other interesting approaches that I would like to mention. The first is targeting the N-terminal part of the androgen receptor. The AR splice variants are missing the C-terminal part, which includes the ligand-binding domain. But all of the AR splicing variants, including AR-V7, retain the N-terminal, which is at the other end of the molecule. And the important thing about the N-terminal is that that is the part of the AR molecule that drives and activates transcription. So, that is the part that is responsible for the transcriptional function of the AR.
And so, if you have an AR-V7 splice variant and you’re able to inhibit the N-terminal part of the androgen receptor, you should, in theory, be able to inhibit not just the full-length AR but also the AR splice variants, including AR-V7. So, right now, there is a phase I study that is currently ongoing using a drug called EPI-506, and EPI-506 is an oral compound that is manufactured by a company called ESSA Pharmaceuticals. ESSA Pharmaceuticals has a phase I study using their AR N-terminal inhibitor, again, called EPI-506, and the phase I data from that experience are being presented at this meeting. It’s a partial phase I trial that is being presented. And the safety at this point seems to be acceptable, and there are some patients who have had minor PSA responses. Of course, the jury is still out because the dose escalation on that phase I trial is still ongoing. So, that is one potentially exciting molecule for targeting the AR-V7-positive patients.
A second strategy that we have explored at Johns Hopkins is using immune checkpoint blockade therapy in these patients. So, at this meeting, we will present the first data of a phase II trial of ipilimumab, the CTLA-4 antibody, plus nivolumab, a PD-1 antibody, in patients with AR-V7-positive castration-resistant prostate cancer. In that prospective trial, we took 15 patients who tested positive for AR-V7 and we treated all patients with the combination of ipilimumab/nivolumab. And although this sample size is quite small—as I mentioned, it’s only 15 patients—we saw clinical responses as well as objective responses in approximately 4 of the 15 patients. Interestingly, some of these patients had very dramatic soft tissue responses with large reductions in their soft tissue disease.
Now, why should this be? One of the interesting things that we discovered, and this is being presented in a few days at this meeting, is that patients who have AR-V7-positive prostate cancer might be enriched for DNA repair gene mutations. And there’s a lot of buzz about DNA repair mutations at this meeting and elsewhere because these mutations appear to sensitize to certain therapies. For example, PARP inhibitors and some types of DNA repair mutations, specifically the mismatch repair mutations, may sensitize also to immune checkpoint therapy.
So, going back to our trial of 15 patients with AR-V7-positive prostate cancer, what we found was that of the 4 patients who responded to ipilimumab/nivolumab, all of them had a DNA repair gene mutation, including BRCA2, which is a well-known homologous recombination repair gene; ATN; as well as 2 other DNA repair mutations. This is exciting to us because we think that we’ve potentially uncovered 2 things: first, presence of AR-V7 may enrich for the presence of DNA repair gene mutations in those patients and secondly, that these patients, or at least some of them, may respond favorably to immune checkpoint blockade with ipilimumab/nivolumab. So, that is a study that we will currently expand to treat more patients, but I think the initial data are exciting enough to warrant some enthusiasm.
Howard I. Scher, MD: One of the strategies that’s being potentially tested is the AR-V7. The AR-V7 still has a functional and terminal domain. Again, think of that as the accelerated portion. And there are a number of strategies that are looking to see if you can degrade the receptor by tagging it with certain drugs that will lead to its degradation. So, again, that’s one approach that’s being evaluated.