Transcript:Noel Clarke, MBBS, FRCS, ChM: If we take patients who have received homologous recombinant repair or HRR-mutation status—in other words, if your tumor is mutated—then we know that a PARP [poly (ADP ribose) polymerase] inhibitor will be administered. Olaparib was the first to be tested in these studies from the group of Johann de Bono, MD, PhD, which has culminated in the TOPARP trial; it has shown that there was a response. What has become interesting is that, in combination studies with olaparib and abiraterone, there seemed to be benefit for those with HRR [homologous recombination repair]-mutation status. There is certainly a signal.

I was a part of that multicenter team that looked at this combination of patients with castrate-resistant disease, all of whom were heavily pretreated. The majority had docetaxel [Docefrez]. They were late in their cancer journey and were given the combination therapy from a large phase II study with 142 patients randomized to abiraterone with olaparib versus abiraterone alone.

The unique design of that trial was that all the patients were given it irrespective of HRR-mutational status, but they were all tested either with tissue or with plasma and germline mutation status. Overall, about 15% of the patients had some form of alteration of HRR mutation. But what was found in the results—and this was published this year in The Lancet Oncology—is that the combination seemed to work irrespective of HRR-mutation status, which was quite a surprise. And I think that opens up the field because, as Nick has mentioned, getting tissue from patients is difficult; doing HRR-mutation status for somatic mutations on tissue actually fails in about half the patients. It does raise the question of whether this will create a new kind of niche, whereby we can use combination therapies in patients who are castrate resistant.

Nicholas James, MD: I think the other thing here is tumor heterogeneity, isn’t it? If you look at patients with germline status and tumor status, they’re not always matching one another. We know that obviously a tumor is metastasized to liver and bone, and that these tumors are going to be different from one another. One of the better ways potentially to look for these mutations is not to biopsy people, because biopsying bone is a general anesthetic job. You’ve got this challenging processing issue with decalcification and getting tumor cells out. We need to circulate tumor DNA, because that way you get a sort of aggregate snapshot of the entire tumor burden, and then you’ve got a potential monitoring tool as well.

We know later on, as these things are inevitably going to be trialed, that you’ve got substantial circulating tumor DNA levels, which could be used both for diagnosis and monitoring. The thing about biopsies is that they’re just one sample. The reason you had a response was because the classical drivers were elsewhere—with the mutations, you simply do not know. But you might get a view of it if you have the circulating tumor DNA.

Noel Clarke, MBBS, FRCS, ChM: I couldn’t agree more, Nick, and I think there’s a huge amount of naïveté in the cancer community about what is achievable in a single biopsy. We just published a paper in European Urology, whereby we took radical prostatectomies from 10 patients. We scanned them, sliced the fresh prostate, sampled it, and did intensive genomic testing on each tumor. What we found was that first, the genomic classifiers didn’t stand up. We found that we could take a tumor core biopsy from the same tumor in the prostate and do differential genomic analysis on those 2 adjacent cores to get different results. I am skeptical that this approach of biopsying a metastasis and then treating will not yield much; it will help us understand the biology. I think it will not help us in the clinic.

Nicholas James, MD: If you’re picking up something that’s on the stem of the branching tree that leads down to all the different lesions you’ve got, then maybe it might. It doesn’t matter where you biopsy. But for a lot of these targeted therapeutics where you have to have quite a specific thing, I’m very skeptical that that is much of an approach at all.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: It sounds like precision, but it’s not really precision because you may be missing a totally different part of the tumor.

Transcript Edited for Clarity.