Emmanuel S. Antonarakis, MBBCh: Prostate cancer is an oncogene addicted tumor, and the oncogene in our case is the antigen receptor, also called the AR. So, almost 100% of prostate cancers are driven and fueled by the AR. And we know that we can treat prostate cancers, including advanced prostate cancers, by suppressing androgens with so-called androgen deprivation therapy. Unfortunately, although almost all patients initially respond to androgen deprivation therapy, they almost always invariably recur and develop progressive disease. And that disease state is known as castration-resistant prostate cancer, also called CRPC.

We now have even newer drugs, novel hormone therapies that can modulate the androgen receptor axis in different ways. So, for example, abiraterone, which is a relatively new drug, decreases the ligand, the testosterone produced by adrenal glands and by the tumor itself. That has been shown to produce wonderful responses in some patients. In addition, enzalutamide, which I consider a super antiandrogen, also inhibits the AR signaling. But, unfortunately, patients still will progress despite enzalutamide and abiraterone, the novel therapies as well.

One of the questions that arises is how do patients become resistant to androgen deprivation therapies and the novel AR-targeting therapies. And our group and others have been very interested and fascinated by that for several years now. There are 2 or 3 mechanisms that have emerged to explain androgen receptor resistance, one of which is that the CYP17 enzyme, which is responsible for producing androgens in the adrenal gland, can be upregulated or overexpressed. The second is that the androgen receptor itself can either become amplified, overexpressed, or mutated. And some of these mutations can activate the androgen receptor. So, these are activating mutations that are stimulated by other ligands other than testosterone. And then the third piece, which our group has been most interested in, are these AR-splicing variants, of which AR-V7 is the most well studied.

So, let’s talk about AR-V7 for a second. Androgen receptor-splice variant 7 is an abnormal splice variant of the androgen receptor that is created because of abnormal splicing of the introns in the AR gene. Just to remind everyone, genes have 2 major parts: the exons, which are the coding regions, and the introns, which are the non-coding regions in between the exons. Introns are normally spliced out of the RNA before the protein is synthesized. If an intron is abnormally retained, if it’s not spliced out, then that can result in an abnormal protein that is truncated; it’s shorter and smaller.

And in the case of AR-V7, the AR gene—which is composed normally of 8 exons—has a splice problem at the third intron, such that the first 3 exons of the protein are transcribed and translated but the remainder of the protein is truncated, so that part is missing completely. So, the AR-V7 protein that results as a consequence of the AR-V7 splicing variant contains the first 3 exons but is missing exons 4, 5, 6, 7, and 8. Those exons that are missing are the parts that encode the ligand-binding domain, the part where androgens bind to and also the part of the molecule where all the antiandrogens bind as well. Our hypothesis was if your prostate cancer patient has a cancer cell that makes the AR-V7 splicing variant, that patient will be resistant to both enzalutamide and abiraterone. So, that was our hypothesis for looking at AR-V7 as a resistance mechanism for hormone therapies and more specifically for abiraterone and enzalutamide.

Andrew J. Armstrong, MD, MSc: AR variants, let’s back up a little bit, are really deletions or splice variants of the androgen receptor where the ligand-binding domain is essentially missing. And so, the DNA-binding domain, meaning the business end of that receptor, is still active. It can still go into the nucleus, turn on a transcriptional program, make cells divide, produce PSA, and turn on a hormonal program.

These receptor variants became important in their discovery over 10 years ago, when it was learned that the androgen receptor is still turned on in many of these tumors, but it’s not blocked by typical antagonists. And so, the ligand-binding domain being missing is a characteristic hallmark of now over 2 dozen androgen receptor variants.

AR-V7 is the most common variant that has been detected to date, clinically. It has a protein translation product. There are other variants like AR-V567 and AR-V9, which are also expressed, and many are likely to be further characterized over time. AR-V7, to date, has the most clinical data as a predictive biomarker and a prognostic biomarker in patients.

AR-V7 was really discovered by several laboratories in finding that prostate cancer cells still remain androgen receptor—dependent in their growth, but they remain ligand independent. So, whether there’s testosterone in the media or a testosterone-free environment, these tumors are still thriving, and they do without the ligand-binding domain. And so, since then, about 10 years have passed in the development of biomarkers of the variant using circulating tumor cells or tissue or drugs that block novel parts of the androgen receptor.

The AR-V7 has come a long way as a clinical biomarker in circulating tumor cells, for example; it is able to predict whether a patient will respond to abiraterone or enzalutamide and how long that response may be. And so, that’s where it has turned in to a clinical assay.

Transcript Edited for Clarity