NEW YORK – Among certain advanced prostate cancer patients with homologous recombination repair gene mutations, Pfizer's PARP inhibitor Talzenna (talazoparib) plus Pfizer and Astellas' androgen receptor pathway inhibitor Xtandi (enzalutamide) significantly improved progression-free survival versus just Xtandi, according to data presented at the American Society of Clinical Oncology's annual meeting.
This was not the first data readout from the TALAPRO-2 study. Back in February, Pfizer shared data from an initial all-comers cohort of metastatic castration-resistant prostate cancer (mCRPC) patients enrolled in the Phase III trial. TALAPRO-2's cohort 1 readout had suggested that Talzenna-Xtandi improved outcomes in the overall population regardless of biomarker status, leading Pfizer to announce that it would seek US Food and Drug Administration approval for the combination in the all-comer population. Pfizer said in a statement on Sunday that its supplemental new drug application in this broad population is under priority review with the FDA and that a decision should come this year. The drugmaker also submitted a marketing authorization application to the European Medicines Agency.
That said, oncologists today, as in February, remain concerned that perhaps the benefit among patients with HRR gene mutations — especially alterations in BRCA1 and BRCA2 — drove the benefit observed in the all-comers population.
An analysis focused on the biomarker-selected patients now shows that there was indeed a significant benefit among the HRR-mutant patients, though the benefit to patients without HRR mutations is still unclear.
Just last week, the FDA approved Merck and AstraZeneca's competing PARP inhibitor Lynparza (olaparib) plus the androgen receptor inhibitor Zytiga (abiraterone) in BRCA1/2-mutated mCRPC. Merck and AstraZeneca had also tried seeking approval in the larger all-comers population based on data from the Phase III PROpel trial. The agency, however, took the advice of its Oncologic Drugs Advisory Committee in approving Lynparza in the narrower, biomarker-defined indication.
Karim Fizazi, a medical oncologist at the Gustave Roussy Cancer Institute in France, presented results from the biomarker-selected TALAPRO-2 patient cohort on Sunday. "Based on these data, talazoparib plus enzalutamide, if approved, should become a standard of care for patients with mCRPC and HRR gene alterations, mostly those with BRCA alterations," Fizazi said.
The HRR-deficient cohort included patients whose cancers had gene alterations in ATM, ATR, BRCA1, BRCA2, CDK12, CHEK2, FANCA, MLH1, MRE11A, NBN, PALB2, or RAD51C, as determined using Foundation Medicine's next-generation sequencing tumor profiling test FoundationOne CDx or its liquid biopsy test FoundationOne Liquid.
"When we drafted the protocol, the goal was only to test this hypothesis in men with DNA repair defects," Fizazi said. "But as [the protocol draft] was ongoing, we decided to make it broader, and to test the same hypothesis in an all-comer population."
After the investigators finished enrolling patients to the all-comers cohort 1, they continued to accrue biomarker-selected patients to the HRR-mutant cohort 2. In this cohort, 399 patients with alterations in any of the above mentioned HRR genes were randomized to receive either Talzenna and Xtandi or placebo plus Xtandi.
"As you can see, TALAPRO-2 is not just one Phase III trial, but actually two Phase III trials asking two different questions in two different cohorts," Fizazi said.
Although these patients had not yet received treatment, including PARP inhibitors, for their metastatic castration-resistant prostate cancers, 8 percent in each arm had received prior abiraterone and around a third in each arm had received prior chemotherapy. Talzenna-Xtandi reduced the risk of progression or death by 55 percent versus just Xtandi. The median progression-free survival was not yet reached in the PARP inhibitor arm versus 13.8 months in the placebo arm.
The combination benefited patients across all subgroups, and Fizazi pointed out that even patients who had received prior abiraterone or chemotherapy fared well on the regimen. In a discussion following Fizazi's presentation, however, David VanderWeele, a genitourinary oncologist at Northwestern University, suggested the number of patients enrolled in the trial who had prior androgen receptor pathway inhibitors — just 8 percent — might not reflect the real-world population, among which he suspected a much higher population do receive prior androgen receptor pathway inhibitors.
Fizazi briefly shared interim overall survival analysis with the caveat that these data were only 24 percent mature. "Still, we see a clear trend separating the combination arm," he said, noting a 41 percent reduction in the risk of death with the addition of Talzenna. The median overall survival was not yet reached with the PARP inhibitor-Xtandi combination versus 33.7 months with only Xtandi at the time of the data cutoff. Patients also did better with the Talzenna combination when it came to the trial's other secondary endpoints, including time to prostate specific antigen (PSA) progression and time to cytotoxic chemotherapy. The objective response rate was also improved with the addition of the PARP inhibitor, with 67 percent of patients on Talzenna-Xtandi experiencing a complete or partial response, versus 40 percent of patients on Xtandi.
Importantly, Fizazi reported that the treatment-related adverse events were significantly higher with the PARP inhibitor than in the control arm. In the experimental arm, about two-thirds, or 66.2 percent, of patients, experienced grade 3 or 4 toxicities versus 37.2 percent of patients in the control arm. Anemia was the most common adverse event in the PARP inhibitor arm, followed by fatigue and neutropenia.
"I've shown you efficacy and safety data, but it's important to look at quality-of-life data to fuse the two," he added, sharing that the PARP inhibitor combination prolonged the time to definitive deterioration in quality of life based on patient-reported health status.
Progression-free survival outcomes by HRR mutation
Fizazi shared a breakdown of the different HRR gene mutations these patients harbored. Not surprisingly, he said, BRCA2 mutations were the most common, followed by ATM, CDK12, CHEK2, BRCA1, and then a longer list of rarer gene alterations.
Homing in on TALAPRO-2 outcomes by patients' specific gene mutations, Fizazi reported that patients with BRCA1/2 mutations derived the most benefit, with an 80 percent reduction in the risk of progression or death with the combination regimen. Fizazi called this "highly significant." Patients with alterations in HRR genes other than in BRCA1/2 derived a more modest benefit from the PARP inhibitor combination, with a 32 percent reduction in the risk of disease progression or death.
When it came to the genes in the "other" category, Fizazi said "there are two ways of looking at this": either gene-by-gene or using a cluster strategy.
Fizazi began with the latter, sharing a gene-by-gene analysis in BRCA genes. "I'm happy to show that apparently both patients with BRCA1 and BRCA2 [gene mutations] derived an important benefit from the combination treatment," he said.
Going through the other gene mutations, Fizazi then highlighted a few key differences in benefit. For instance, the patients with CDK12-mutant cancers had a median progression-free survival of 21.9 months with Talzenna-Xtandi versus 13.8 months with placebo-Xtandi.
Meanwhile, patients who only had ATM mutations, or patients who only had CHEK2 mutations, did not seem to benefit quite as much from the combination. For patients whose cancers only harbored ATM gene mutations, the median progression-free survival with the PARP inhibitor had not yet been reached, versus a 27.7 median progression-free survival with the control regimen. Among patients with only CHEK2 mutations, the median progression-free survival was 22.1 months versus not yet reached with Talzenna-Xtandi and Xtandi, respectively.
While it can be valuable to look at the gene-by-gene breakdown, Fizazi noted that patients can have multiple HRR gene mutations at once. In the TALAPRO-2 trial, these patients comprised about 15 percent of those enrolled to the biomarker-selected cohort. To account for these patients, the investigators found it valuable to look at the outcomes with a clustering hierarchy approach. Here, they saw that patients with any BRCA mutations in the BRCA1/2 cluster also tended to benefit most, regardless of the presence of other HRR mutations. The same was true for patients with CDK12-mutant tumors.
Limitations, lingering questions
One important consideration that VanderWeele highlighted in his discussion of the TALAPRO-2 data was the fact that patients who do not respond to androgen receptor pathway inhibitors do have the option to go on to receive a PARP inhibitor as monotherapy in the real world. Because only 17 percent of patients in the TALAPRO-2 study who progressed in the control arm received Lynparza after progression on Xtandi, he suggested the study data might not fully capture the actual mCRPC population.
This raises questions about the relative benefit of the combination versus sequential therapy, he suggested. The TALAPRO-2 trial, of course, was not designed to compare combination therapy versus sequential approaches. To this end, VanderWeele pointed to a much smaller Phase II study underway, dubbed BRCAaway, in which patients are randomized to one of three arms: Zytiga plus a steroid, Zytiga plus a steroid and Lynparza, or Lynparza alone. In that trial, patients could go on to receive Lynparza after progression or vice versa if they received the Zytiga-steroid combo first. So far, he said early data suggest a benefit with the combination of all three, although those data are still immature.
Ultimately, the remaining question on genitourinary oncologists' minds following the TALAPRO-2 readout, especially after the recent biomarker-defined Lynparza-Zytiga approval in this patient population, is whether allcomers, including patients without HRR gene mutations, should be offered a PARP inhibitor plus an androgen receptor pathway inhibitor. The FDA's decision for Lynparza was narrower than Merck and AstraZeneca had hoped, and it remains to be seen whether the same will play out for Pfizer's Talzenna-Xtandi application.
Fizazi acknowledged that this question is still difficult to answer.
"Scientific interpretation is not easy," he said, adding that it's still hard to tell if there is a true benefit in patients without HRR alterations based on a synergy between the PARP inhibitor and the androgen receptor pathway inhibitor. It's possible that non-HRR-mutant patients who benefit from both drugs may have alterations that weren't identified with the NGS tests.
In the absence of a clear answer on that scientific question, though, Fizazi said it's important to consider the medical question. "The medical question, beside the scientific interpretation, is whether the data are good enough for use," he said. "There are different opinions here because we do see some progression-free survival benefit [in the non-HRR-mutant patients], but whether this is sufficient to justify use in a context where overall survival … are not necessarily demonstrated, is something we may question in a context where toxicity is demonstrated."
As the FDA considers Pfizer's application, these questions may also be top of mind for reviewers, too.