Rucaparib Prostate Cancer Approval Sets Stage for Further Research

In Partnership With:

Partner | Cancer Centers | <b>Mayo Clinic</b>

Alan H. Bryce, MD, discusses the recent regulatory approval of rucaparib in metastatic castration-resistant prostate cancer and explains how the approval has underscored the importance of genetic testing in the field.

The FDA approval of rucaparib (Rubraca) in BRCA-mutant metastatic castration-resistant prostate cancer (mCRPC) marks the first designation for a biomarker-driven targeted therapy in the field, said Alan H. Bryce, MD, who added that the regulatory decision reinforces the importance of germline and somatic genetic testing for men with high-risk and metastatic disease.

“Until this approval, prostate cancer was 1 of the big 4 tumors without any biomarker-directed therapies. That has changed with rucaparib, which is an extremely effective and attractive treatment option for patients who harbor BRCA mutations,” said Bryce. “In summary, germline testing should be done unequivocally for all patients with high-risk, very high-risk, or metastatic prostate cancer. Somatic testing, either by tissue biopsy or cell-free DNA (cfDNA) should be done at the time of progression when you’re looking to make a treatment decision about what to put the patient on next.”

The accelerated approval is based on data from the ongoing phase 2 TRITON2 trial (NCT02952534), in which interim results showed that treatment with rucaparib led to a 44% confirmed objective response rate (ORR) in 62 evaluable patients with BRCA1/2-mutant mCRPC who had received prior androgen receptor (AR)–directed therapy and taxane-based chemotherapy.1 Among patients with BRCA1/2 alterations, 51.1% experienced a prostate-specific antigen (PSA) response to the PARP inhibitor.

At the 2019 ESMO Congress, updated findings with an additional 7 months of follow-up showed that rucaparib led to a 43.9% confirmed ORR and a confirmed PSA response of 52.0% in patients with BRCA1/2-mutant mCRPC. At 24 weeks, 60% of responses were still ongoing.2

The adverse effects (AEs) of the PARP inhibitor were consistent with the drug’s use in breast cancer and ovarian cancer, said Bryce. In the trial, the most common all-grade treatment-related AEs occurring in more than 20% of patients included asthenia/fatigue (55.3%), nausea (49.5%), anemia/decreased hemoglobin (37.9%), decreased appetite (27.9%), transient increased aspartate transaminase/alanine aminotransferase (24.7%), constipation (24.7%), vomiting (22.1%), and diarrhea (21.1%).

The approval is contingent on the results of a confirmatory trial. The phase 3 TRITON3 trial (NCT02975934), in which patients with homologous recombination deficient mCRPC who have received 1 prior AR-directed therapy will be randomized to rucaparib or physician’s choice of abiraterone acetate (Zytiga), enzalutamide (Xtandi), or docetaxel.3

In an interview with OncLive, Bryce, assistant professor of medicine and medical director of the Genomic Oncology Clinic, Mayo Clinic, discussed the recent regulatory approval of rucaparib in mCRPC and explained how the approval has underscored the importance of genetic testing in the field.

OncLive: How will the accelerated approval impact the treatment paradigm for patients with BRCA1/2-mutant mCRPC?

Bryce: This is an important moment in prostate cancer in that we finally have the first approval for a biomarker-selected targeted therapy. We’ve known for some time that BRCA1/2 mutations in prostate cancer are predictors for efficacy to PARP inhibitors. Multiple studies have shown that. Now, we have [data from the TRITON2 trial] which led to the FDA approval of the PARP inhibitor rucaparib for use in patients with BRCA1/2-mutant mCRPC.

Could you expand on the data from the TRITON2 trial, which served as the basis for the drug’s approval?

The accelerated approval was based on ORR in patients [with soft tissue disease] who had measurable disease by RECIST criteria. Only a proportion of patients with prostate cancer have measurable soft tissue disease. Approximately half of patients will have bone-only disease, which is not assessable by RECIST criteria. The approval was based on the patients who had some degree of soft tissue disease, whether lymph node or visceral disease.

In patients with RECIST measurable soft tissue disease, rucaparib led to an ORR of 44% by RECIST criteria, which is fantastic. It’s better than we expect to see with chemotherapy in the same setting. It’s a meaningful outcome. Patients’ soft tissue tumors shrank. In the overall population, we saw a PSA response rate of over 50%.

Regarding safety, what does the toxicity profile for the agent look like?

We know a lot about the toxicities of PARP inhibitors because of their extensive use in breast cancer and ovarian cancer. The only variable in prostate cancer is that these are male patients. Despite this, we didn’t see a difference [in toxicity]. The toxicity profile is the same as what oncologists are accustomed to from treating patients with breast cancer and ovarian cancer. Additionally, this is a slightly older patient population who have potentially had pelvic radiation and other myelosuppressive treatments over time.

[Even so], the main toxicity is anemia. Prolonged use of a PARP inhibitor can certainly cause some myelosuppression. You can see the blood counts drop. Other than that, there is some potential for gastrointestinal toxicity, nausea, diarrhea, and some fatigue. However, it’s fairly well tolerated and consistent with what the oncologist already knows from their use in patients with breast cancer and ovarian cancer.

The approval is contingent on results from the TRITON3 trial in patients who have not received prior chemotherapy. Could rucaparib be moved to earlier lines of therapy?

TRITON2 was a phase 2, single-arm study in patients who had already received chemotherapy. TRITON3 is the phase 3, randomized study in patients who have not received prior docetaxel but have had at least 1 of the AR inhibitor, abiraterone or enzalutamide. The TRITON2 population [enrolled patients with] later-stage disease. TRITON3 is going to [evaluate rucaparib] earlier in the disease course.

We expect to see the [same] drug development with rucaparib as we see with every other cancer drug in which it starts in the late-stage setting and is marched forward in the treatment paradigm. TRITON3 is moving [rucaparib] further forward.

Trials have already launched [with rucaparib] in the first-line CRPC setting. For example, rucaparib in combination with abiraterone or other AR inhibitors. We have already seen studies launch with combinations of PARP and AR inhibitors, such as abiraterone or enzalutamide in the first-line CRPC setting. Other smaller studies are evaluating [rucaparib] in the neoadjuvant or metastatic hormone-sensitive settings as well. You can expect to see PARP inhibitors try to cross the full spectrum of metastatic prostate cancer.

How will the approval impact the use of genetic testing in prostate cancer?

The National Comprehensive Cancer Network already recommends germline testing in every patient with high-risk, very high-risk, or metastatic prostate cancer. With germline testing, we test a patient’s inherited genetic profile and [determine what] they are passing onto their children and what is associated with any relatives.

Somatic mutations are the private mutations in the cancer cells themselves. PARP inhibitors have utility in patients with germline or somatic mutations. In the setting of any high-risk, very high-risk, or metastatic prostate cancer, patients should have germline testing to inform themselves and their family. Secondly, in the context of treatment decisions, somatic testing needs to be done on top of germline testing. It’s really not one or the other. At least half of patients will only have mutations in their tumor and not in their germline.

Tissue testing should be done on the most recent tissue possible. It’s a common scenario in prostate cancer that men will have had their prostatectomy or radiation 10 years prior, for example. If you test that old biopsy tissue or prostatectomy tissue, the DNA might be degraded, so you might not be able to get a good result.

Secondly, the tumor will evolve over time. Over those 10 years, the genetic signature of that tumor will change. It will be different today than it was at the time of initial diagnosis. It’s really important for treating clinicians to remember to test the current tumor to determine how to treat the patient, the gold standard being a repeat biopsy with a piece of tissue from any tumor. You don’t have to biopsy each tumor site—just 1 is ok.

If for whatever reason, a biopsy is not feasible or perhaps the risk is too high, the clinician can get a cfDNA blood test that will screen for the mutation in the circulating DNA in the blood. That was a major part of the TRITON studies, as well as other PARP inhibitor studies in other disease settings.

The correlation between cfDNA and tissue is extremely high. A cfDNA test is a good test to determine which patients will respond to PARP inhibitors. A positive cfDNA test for BRCA predicts for response to rucaparib in this setting. A positive test is a real result; a negative test result isn’t necessarily real. A negative test can mean that there’s just not enough cfDNA in the blood. [The amount of cfDNA could] fall below the level of detection.

Get the test as the cancer is advancing. You need growing tumor that is shedding DNA. You don’t want [to order] a cfDNA test when the PSA is 0 and the disease is responding to therapy.

As the clinician [exhausts] multiple therapies, they should be repeating that somatic testing. Every time we treat a patient with a new drug, the tumor is going to evolve in response to that treatment, so the patient might not have a BRCA mutation at initial diagnosis. They might not have [a BRCA mutation] when we’re treating hormone-sensitive disease or first-line castration-resistant disease, but they might [develop a BRCA mutation] in the second- or third-line castration-resistant setting, so it’s important to retest.

Are there any unanswered questions with rucaparib that future research efforts should address?

Outstanding questions include the use of PARP inhibitors in other DNA damage response mutations apart from BRCA1/2. We know PARP inhibitors work in patients with BRCA1/2 mutations, but what about ATM, CHK2, FANK, and RAD51? There’s a long tail of the curve of uncommon mutations that still fall on this pathway. Several publications are coming out, specifically looking at this. I really don’t think there is a role for PARP inhibitors in patients with ATM mutations right now. Those data are fairly unimpressive. A small proportion [of patients with ATM mutations] will respond to these agents, but the vast majority don’t seem to benefit much at all.

We’re seeing responses in other mutations, such as PALB2 and RAD51. However, they’re such uncommon mutations that it’s going to be hard to build a big enough data set moving forward. The field is going to have to put quite a bit of thought into how we develop a meaningful set of criteria around these rare mutations. There is more to come on that. It’s probably going to require meta-analyses from multiple studies and perhaps prospective registrational studies in order to gather enough data to answer those questions.

References

  1. Clovis Oncology highlights Rubraca (rucaparib) updated data from the ongoing TRITON2 clinical trial in patients with mCRPC and exploratory and integrated analyses in recurrent ovarian cancer at the ESMO Congress 2019 [news release]. Clovis Oncology: Boulder, CO; September 29, 2019. bit.ly/2u4iEPS. Accessed May 19, 2020.
  2. Abida W, Bryce AH, Vogelzang NJ, et al. Preliminary results from TRITON2: a phase 2 study of rucaparib in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) associated with homologous recombination repair (HRR) gene alterations. Ann Oncol. 2018;29(suppl_8):viii271-viii302. doi:10.1093/annonc/mdy28
  3. A Study of Rucaparib Versus Physician’s Choice of Therapy in Patients With Metastatic Castration-Resistant Prostate Cancer and Homologous Recombination Gene Deficiency (TRITON3). https://bit.ly/2zf65Et. Updated March 31, 2020. Accessed May 19, 2020.