Best Use of PARP Inhibitors in mCRPC "Presentation" - Joaquin Mateo

November 15, 2024

At the 2024 Advanced Prostate Cancer Consensus Conference (APCCC), Joaquin Mateo discusses PARP inhibitors in BRCA-mutated prostate cancer treatment. He explores their effectiveness against tumors with impaired homologous recombination repair while highlighting key findings from PROfound and TRITON3 trials. 

Biographies:

Joaquin Mateo, MD, PhD, Medical Oncologist, Prostate Cancer Translational Research Group at the Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain


Read the Full Video Transcript

Joaquin Mateo: These are my disclosures. So we are starting now a series of three talks covering basics of PARP inhibitors. I'm going to focus on the population with BRCA1 and BRCA mutations.

But before I start, one concept that I wanted to discuss is that we need to acknowledge certain imprecision in our precision medicine use of PARP inhibitors. We know that PARP inhibitors, as a drug class, are active against those tumor cells that are incapable of repairing double-strand breaks by homologous recombination repair. But we actually don't have a test that tells us if the tumor is proficient for homologous recombination.

We and others are working on functional tests, but they are not ready yet for clinical practice. So we are using, among the indirect biomarker options that we have, the one that we think associates better with these functional defects, which is the identification of mutations in the genes that are relevant for this pathway. And among them, clearly, BRCA1 and BRCA2 mutations are the canonical example of gene events that lead to homologous recombination deficiency. But I think that we need to acknowledge this indirect effect of the biomarker that explains some of the results in clinical practice.

So BRCA-deficient metastatic prostate cancer is a distinct disease subset with poor prognosis on hormonal therapies—less clear for docetaxel. This has been shown in several studies that then merits specific therapeutic interventions that may complement what we already have for other patients.

These tumors are highly sensitive to PARP inhibition. Clinical proof of concept in prostate cancer was shown initially in the TOPARP trials. You can see here, the red bars are those patients with BRCA1 or BRCA2 mutations that are the ones that respond more often and also for longer.

And this data was then validated with different PARP inhibitors across different trials and, critically, also in two large randomized phase III trials—PROfound with olaparib and TRITON3 with rucaparib—different drugs, different control arms. And I'm here showing the subgroup analysis for the BRCA1 and BRCA2 population, as the trials included patients with other genomic backgrounds, where clearly PARP inhibitors were superior to the control arms. And since then, they are an approved therapy for mCRPC with this genomic background.

However, there are areas of uncertainty where their use is still not clear. The first one that I want to address is the use of PARP inhibitors in men with BRCA1 mutations, rather than BRCA2. We know that the prevalence of BRCA1 mutations in metastatic prostate cancer is around 1% to 1.5%. BRCA2 is 8% to 10%. So really, when we are looking at the medians and confidence intervals of our trials, we are mostly looking at BRCA2-related data. We don't have that much data specifically for BRCA1. But when we go into the detail of these trials, we have seen in TRITON and PROfound that the responses in the BRCA1 population are less than for BRCA2.

Does it mean we should not use this drug? No. I think we have evidence from the fields of ovarian and breast cancer that biallelic loss of BRCA1 in the tumor leads to homologous recombination deficiency and sensitivity to PARP inhibitors. But the problem is that we may have scenarios in prostate cancer where the BRCA1 mutation does not associate with this phenotype.

For example, we have data from large series that show us that, often, BRCA1 mutations in prostate tumors do not go together with complete loss of the gene in the tumor. And that's very interesting because you can see here data from the Foundation Medicine database. So you have BRCA1 here. And blue are those patients with complete loss in the tumor associated with the original mutation. And that's very different from ovarian or breast cancer, where there is consistency between BRCA1 and BRCA2. There is a clear disconnection between BRCA1 and BRCA2 in the rate of biallelic loss. If we go to earlier settings and the cancer genetic role, it has been shown also that the cumulative incidence of prostate cancer among men with germline BRCA1 mutations is much lower than for those with germline BRCA2 mutations.

And together, these data suggest that there are things that we still don't understand about the role of BRCA1 in prostate carcinogenesis. And my recommendation would be that in front of a BRCA1 mutation, we need to understand the relevance of that mutation. So it's very important that, in clinical practice, we can communicate with the laboratories that generate the sequencing data to truly understand the impact of that finding on the disease for that particular patient.

Second, we have been asked for years whether the efficacy of PARP inhibitors is different in patients where the BRCA mutation is of germline versus somatic origin. We have data now from three trials consistently showing that response rates and durations of responses are almost identical, regardless of whether the origin of the mutation is germline or somatic. And if anything, we have seen in some of these studies that there is a subset of patients with homozygous deletion, which is a very particular way of losing the gene function, that have very, very long responses. But in general, germline and somatic mutations have similar outcomes for PARP inhibitors.

So Karim referred to this—do we use the PARP inhibitor before or after taxanes? We don't really have good data comparing the sequence of PARP followed by taxane or taxane followed by PARP in the BRCA population. I think it's going to be very difficult that this data ever exists. But we have subgroup analysis from PROfound and TRITON3 that suggest that there is benefit in using these drugs prior to docetaxel. But I think it's down to a patient-by-patient discussion because both options are reasonable.

And where we have seen more new data from APCCC 2022 to now is in the use of combinations. We have seen three large randomized trials reading out in these two years, comparing the combination of different ARPIs with different PARP inhibitors versus the use of ARPIs alone. And I'm only here showing the BRCA population in these trials because that's the focus of the talk. And I think it's clear from all the trials that these patients benefit from the addition of a PARP inhibitor.

But remember, these trials are comparing ARPI plus PARP versus ARPI alone. So basically, what these trials are telling us is that using a PARP inhibitor in BRCA-mutated prostate cancer is better than not using a PARP inhibitor—so they are validating what we are already seeing in the monotherapy trials. The question is, with these data, should we use combination as first-line CRPC for our BRCA metastatic prostate cancer patients? Well, again, we don't have data comparing the combination versus sequential use because that was not the question posed by these trials.

We have promising data by Maha Hussain and colleagues in the BRCAAway trial, but it's a small group of patients. My opinion, considering that these patients with BRCA mutations have poor outcomes on hormonal therapies, that we see consistent trends in all the trials, and that there is biological plausibility—not sure for synergy, but at least for additive effect—I think that it's reasonable to consider these combinations for patients who have not received an ARPI drug before, always acknowledging the lack of definitive data and discussing with the patient the risk of increased toxicities. But for me, that is not clear for patients who have already received an ARPI drug in the hormone-naive stage, which is now the standard of care in clinical practice.

As Karim mentioned, several trials discourage us from using an ARPI after ARPI in the CRPC space. Why would it be different in the hormone-naive to CRPC setting? The biological plausibility, for me, says that it's probably not different. And if it is different, it needs to be proven. And until that, if these patients with BRCA mutations have the option of receiving a PARP inhibitor as monotherapy, that would be, for me, the standard option.

In these trials, it is true that, because of the time when they were conducted, the proportion of patients that received an ARPI in the hormone-naive stage is very small. That's normal. But even if there were more patients in this scenario, in those trials, those trials are not the ones who would answer that question.

If we want to answer that question, really, what we have to do is run trials on patients that are progressing to an AR-targeting drug in the hormone-naive stage, and then at progression to CRPC, randomize them to what, for me, is the control arm—the PARP inhibitor monotherapy (we can discuss whether taxanes should also be here)—versus an experimental arm with a combination, not against a second ARPI, because that's not the question that we want to answer. And I'm going to use the time on the stage here, because I think this is a great audience to say that we should, together as a community, run these trials.

So I'm going to finish here with basic messages. BRCA1 and 2 mutations associate with HRD in prostate cancer and represent a group of patients with poor prognosis. These tumors are highly sensitive to PARP inhibition irrespective of the germline versus somatic origin of these mutations. And based on that, NGS profiling should be considered in all metastatic castration-resistant, at least, prostate cancer patients.

The trials that have confirmed the efficacy of this strategy in the BRCA population, both as monotherapy as well as in combination—so PARP inhibitors have a role in the treatment of these patients. It's probably less clear when is the right moment to treat it, but it's clear that using a PARP inhibitor at some point during the natural history of the disease is important for these men.

And in the absence of robust data comparing the combinatory versus sequential use of ARPI and PARP inhibitors, and while we wait to generate that data, I think that in the ARPI-naive CRPC space, it is reasonable to consider, based on the risk-benefit ratio, the use of combinations until definitive data is available. But I think that in patients who have already progressed on an ARPI, we should consider monotherapy as the standard of care until the right trials are conducted and tell us what's the best way forward for these patients. Thank you very much for your attention.