Emmanuel S. Antonarakis, MBBCh: At the ASCO 2017 meeting, there are several interesting data being presented in relation to AR-V7. I think the most important one will be presented by my colleague, Dr. Mary-Ellen Taplin from Dana-Farber Cancer Institute. She will be presenting the results of the ARMOR3 clinical trial. The ARMOR3 trial was a study where patients were selected based on AR-V7 positivity at baseline, so they had to be tested for AR-V7. If they were positive for AR-V7, they were then randomized in a 1:1 ratio to receive either galeterone or enzalutamide. Galeterone is a novel hormone therapy agent that supposedly has 3 functions, one of which is a CYP17 inhibition, just like abiraterone. The second function is it’s an AR antagonist, just like enzalutamide. But the third function is that it was supposedly able to degrade the AR and the AR-V7 proteins, so it was an AR-V7 protein degrader.

And so, the goal of the ARMOR3 study was to enroll patients who had castration-resistant prostate cancer who had not previously received enzalutamide, abiraterone, or chemotherapy, and then to randomize the AR-V7—positive patients to enzalutamide versus galeterone. Unfortunately, as will be presented later at this meeting, the study was closed early. It was prematurely terminated because of a futility analysis. But the analysis thing about that trial is that we screened 953 patients for AR-V7 in a very short time course, and that really will be the biggest patient population, I would say, ever that will be screened in such a short time fashion for AR-V7.

The top line result is that in the first-line CRPC population—pre-enzalutamide and pre-abiraterone—of the 953 patients who were prescreened for that study, 8% of them were AR-V7—positive using the modification of the Johns Hopkins AR-V7 mRNA assay. So, 8% of the 953 patients were positive. Now, we also looked into the correlations of clinical characteristics with the presence of AR-V7, and 3 interesting correlations emerged, which we are learning more about. The first is, patients who had higher baseline PSA levels were more likely to have AR-V7–positive circulating tumor cells—perhaps not surprising—but patients who had a PSA level above the median had a higher prevalence of AR-V7 than patients with a baseline PSA level below the median.

The second correlation was with the number of bone metastases. Patients who had 20 or more bone metastases at baseline before enrolling in the trial had a higher AR-V7 prevalence than those who had less than 20. And, in fact, patients who had 10 or fewer metastases in the bone had even lower prevalence, and then patients who had 5 or less metastases had the lowest prevalence of AR-V7. So, a very direct correlation between higher level of disease burden in the bone and a higher prevalence of AR-V7.

And the third correlation, which to me was a bit surprising and perhaps the most interesting, was that a subset of these patients had received chemohormonal therapy for their initially diagnosed metastatic hormone-sensitive prostate cancer. And this was a result from the data from the CHAARTED study and the STAMPEDE study, suggesting that 6 cycles of docetaxel when given early at the onset of metastatic hormone-sensitive disease improves survival.

Many of the patients in the ARMOR3 study, by the time they developed first-line CRPC, had previously received docetaxel chemotherapy for hormone-sensitive disease. Again, contrary to some expectations, the prevalence of AR-V7 in patients who had previously received chemotherapy was increased rather than decreased. So, prior use of chemohormonal therapy for the metastatic hormone-sensitive setting increased the chance of finding AR-V7 in circulating tumor cells. On the one hand, that might seem a little bit paradoxical. On the other hand, it perhaps suggests or indirectly implies that docetaxel may be an AR-modulating therapy. In other words, it can also inhibit AR in some way. And, therefore, by inhibiting the wild-type normal canonical AR, perhaps the AR splice variants emerge as a way to overcome or escape the therapy. Again, this is just a hypothesis. It’s quite intriguing. It has not been proven at this point.

Howard I. Scher, MD: An important presentation will be presented by Dr. Taplin. There was a drug that was shown to have activity in AR-V7-positive patients, and it was tested in a first-line registrational trial, looking to see if the drug were active in that specific population. Unfortunately, that trial had to be closed because the frequency that the AR-V7 was identified was less than 10%, which meant that a large number of patients had to be screened, and that it wasn’t clear that the drug had the activity that it was hoped to have based on some of the earlier studies.

Transcript Edited for Clarity