Released: April 12, 2021
Expiration: April 11, 2022
Genetic testing is now the standard of care in prostate cancer, driving therapy selection and improving patient outcomes. There are several reasons genetic testing is now the standard of care. The first is that DNA repair alterations are very common. Whereas somatic DNA testing results can change over time, 23% of patients with mCRPC have DNA repair alterations and 11.8% have germline DNA repair defects.1-3 Genetic testing can also reveal prostate tumors with high microsatellite instability, which would drive the use of pembrolizumab. Ideally, the most recently collected tissue sample should be used when making a therapy selection. Tissue quality and the analytical platforms that are available are also important factors in deciding which samples to test.
Olaparib is a PARPi that received FDA approval in 2020 for patients with deleterious or suspected deleterious germline or somatic HRR gene–mutated mCRPC who have progressed following prior treatment with enzalutamide or abiraterone.4 Approval was based on the phase III PROfound trial (N = 387).5,6 Patients who were refractory to abiraterone or enzalutamide were randomized 2:1 to receive 300 mg olaparib twice daily or physician’s choice of AR-targeted therapy. There were 2 cohorts of patients based on whether patients had BRCA1, BRCA2, or ATM alterations (cohort A) or other HRR alterations (cohort B). The primary endpoint of this study was radiographic progression‑free survival (rPFS) in cohort A using Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 and Prostate Cancer Working Group 3 (PCWG3) consensus criteria by blinded independent central review (BICR). Secondary endpoints included confirmed radiographic overall response rate (ORR) in cohort A and overall survival (OS) in cohort A.
Cohort A patients treated with olaparib had a significant rPFS benefit with a 66% reduction in the risk of death (HR: 0.34; 95% CI: 0.25-0.47; P <.001) and a median rPFS of 7.4 months compared with 3.6 months for AR-targeted therapy. A subset of patients did have quite a long rPFS.
Within cohort A, patients treated with olaparib had a longitudinal reduction in the risk of death of 31% (HR: 0.69; 95% CI: 0.50-0.97; P = .02) and a median OS of 19.1 months vs 14.7 months for AR-targeted therapy.7 In the overall patient population there was a trend toward an OS improvement as well, but there is some nuance for the patients with less-common mutations. Olaparib was approved based on these data for patients with 14 different HRR mutations including BRCA1 and BRCA2.4
In the overall patient population, most toxicity from olaparib was hematologic and gastrointestinal related. In many ways, giving a PARPi is like giving an oral chemotherapy; hematologic counts need to be followed and patients should be given antiemetics to have at home. It is rare for patients to have nausea and vomiting to the extent that they require hospitalization or therapy discontinuation, but many require antiemetics. The most common hematologic toxicity was anemia that affected 46% of patients and 21% had grade ≥3 anemia. In patients who are developing anemia, doses can be held or reduced to prevent anemia from progressing, so this is not a major concern.
Rucaparib is another orally available PARPi and it has received accelerated approval from the FDA for the treatment of adult patients with a deleterious BRCA mutation (germline and/or somatic)–associated mCRPC who have been treated with AR-directed therapy and a taxane-based chemotherapy.8
There are several key differences from the olaparib approval. First, rucaparib is approved only for patients with BRCA‑mutated mCRPC. Previous treatment with AR‑directed therapy and taxane‑based chemotherapy is also required. Approval was based on the phase II TRITON2 study.9,10 An ongoing phase III study with rucaparib continues to accrue patients (TRITON3) and results of this confirmatory study may alter the indications of rucaparib.
The international, multicenter, open-label phase II TRITON2 trial enrolled 115 patients with mCRPC who had deleterious somatic or germline HRR alterations.9 Patients could have a wide range of mutations, including alterations in BRCA1, BRCA2, and ATM and had to have progressed on 1-2 AR-directed treatments and had 1 previous line of taxane-based chemotherapy and no previous PARPi therapy. All patients received 600 mg of rucaparib twice daily in 28-day cycles until radiographic or unequivocal clinical progression, unacceptable toxicity, or discontinuation for other reason.
The primary endpoint for patients with measurable disease at baseline was centrally assessed confirmed ORR per modified RECIST/PCWG3. The primary endpoint for patients without measurable disease at baseline was locally assessed, confirmed prostate-specific antigen (PSA) response, defined as a ≥50% decrease.
Patients in the BRCA1/BRCA2 cohort had a 44% confirmed ORR (n = 62), and this was the basis for the FDA filing. There was not a consistent association between germline or somatic alterations and response, as patients in both groups benefited.
The phase III TRITON3 trial evaluating rucaparib plans to enroll 400 patients.11,12 Rucaparib will be compared with physician’s choice of therapy in patients with mCRPC who are chemotherapy naive, have BRCA1/2 or ATM alterations, and have not previously received a PARPi. Enrollment is ongoing so we can expect data within the next few years. The primary endpoint is rPFS by independent radiology review. This is a confirmatory trial that, if positive, would move rucaparib into that prechemotherapy space and potentially demonstrate favorability over therapies like docetaxel.
At ASCO GU 2021, de Bono and colleagues13 presented data from an interim analysis of the phase II TALAPRO-1 trial.14 At the time of analysis, 128 patients with mCRPC were enrolled. Eligibility for this study includes the presence of a monoallelic or biallelic DDR alteration, 1-2 previous chemotherapy regimens, and progression on abiraterone acetate or enzalutamide. Talazoparib is given at 1 mg/day until radiographic progression or discontinuation for other reason. In the interim analysis, the efficacy population contained 104 patients with predefined DDR alterations and the safety population contained 127 patients. The primary endpoint is ORR, and secondary endpoints include rPFS, OS, and safety.
In the efficacy population, 50.0% of patients had BRCA2 alterations and 3.8% had BRCA1 mutations. The ORR was highest for patients with BRCA1/2 alterations at 45.9%. Responses were lower in patients with alterations in either PALB2 or ATM. The clinical benefit rate is impressive for a heavily pretreated population. A PSA decrease ≥50% from baseline was observed in 63.9% with BRCA1/2 alterations. Although the number of patients with PALB2 mutations was small (n = 4), 3 of the 4 experienced PSA decreases of ≥50%. This is certainly encouraging and may be another measure of efficacy worth exploring for some patient populations.
These waterfall plots are colored by mutation type showing the effect of each mutation type on both tumor response and PSA levels.
Common adverse events from talazoparib included anemia, thrombocytopenia, neutropenia, and gastrointestinal-related issues like nausea and decreased appetite. Treatment-emergent adverse events that resulted in death included disease progression, pulmonary embolism, cardiorespiratory arrest, and neoplasm progression. It is important to note that none of the deaths was deemed related to talazoparib. These early phase data are promising for single-agent talazoparib in prostate cancer.
Evidence from the TRITON2 and PROfound studies led to FDA approvals allowing the use of olaparib and rucaparib in selected men with mCRPC. But when should we use them and in whom should we use them? Tumors with BRCA1 and BRCA2 mutations appear to be most sensitive. The numbers on the other, less common mutations are a bit smaller, so it is hard to make significant, sweeping generalizations about response in those contexts.
I want to highlight some nuances in PARPi therapy that one might encounter in the clinic. Different DNA repair alterations respond very differently to PARPi. On the top left, one can see the most frequent alterations and the difference in PFS between olaparib and control in the PROfound study. Patients with alterations in BRCA2 had significantly improved PFS with olaparib (10.84 vs 3.48 months; HR: 0.21) and patients with CDK12 also had an impressive improvement (5.09 vs 2.20 months).
It is puzzling that patients with ATM mutations do not seem to have a similar benefit. It is not that patients with ATM mutations never respond to PARPi, but it is less than we thought it would be. This is surprising, because ATM alterations were assumed to be PARPi sensitive. This may result from ATM sensing DNA damage and sounding the alarm rather than actually repairing the DNA, so we may need other strategies for these mutations like ataxia telangiectasia and Rad3-related (ATR) inhibitors. After BRCA2, ATM is the most common alteration in prostate cancer, affecting 5% or 6% of the CRPC patient population, so we clearly need effective therapies for this group of patients.
The number of patients enrolled on PROfound with BRCA1, CHEK2, PPP2R2A, RAD51B, and RAD54L alterations was very small. But I think one can see that some mutations benefit from olaparib whereas others do not appear to benefit.
Although the approval for rucaparib is for BRCA1 and BRCA2 alterations only, TRITON2 enrolled patients with other mutations and collected response data, showing more mixed results.15 For instance, one of the 2 enrolled patients with an alteration in FANCA had a great response while the other did not. A mixed response was also observed for patients with CDK12 alterations. As an academic and clinical researcher, I look at this and think it is great that some patients respond very well, especially patients with alterations in BRCA1/2. But why are there not responses in patients with other HRR mutations who we thought would respond?
Factors related to tumor biology and patient biology may drive the response. For example, deletion of RB1 and alterations in other tumor suppressor genes may override the benefit of PARPi therapy. These data demonstrate that patients with BRCA2 alterations with RB1 deletion or MYC amplification have a worse prognosis.16 This is an area of ongoing investigation and I think we can expect more data.
Based on these data, I would expect a better response for a patient with an isolated BRCA2 mutation than for a patient with a complex set of deletions and mutations that includes BRCA2. Patients with complicated tumor subsets may have worse prognoses, so I advise considering the whole genomic profile when setting expectations with patients.
Another key point is that there are more targets in DNA repair than PARP. The protein kinases ATM, DNA‑PK, and ATR also regulate DNA damage response and can be mutated in prostate cancer. Clinical trials are underway for inhibitors of ATR and DNA‑PK.
Most data for single‑agent PARPi are from patients with BRCA1 or BRCA2 alterations. Patients with less common HRR mutations should receive olaparib based on the FDA-approved indication, and patients with BRCA1 or BRCA2 alterations and previous AR-directed therapy can receive olaparib or if they also had previous docetaxel, rucaparib would be an option. Olaparib is currently the only PARPi indicated without previous taxane therapy. I encourage you to consider HRD mutations in the full context of disease biology, as other alterations, high volume disease, and high Gleason grade are associated with reduced efficacy with PARPi monotherapy. Genomic testing should also be considered as prognostic, not just something to determine PARPi eligibility.
Back
