An immunotherapy combination for metastatic castration-resistant prostate cancer (mCRPC) failed to hit a response target but still exceeded historical results, a small prospective study showed.
Ten of 35 (28.6%) patients responded to the combination of nivolumab (Opdivo) and ipilimumab (Yervoy). The activity fell short of the 40% threshold for a positive outcome but well above the historical response rate for immunotherapy in patients previously treated with docetaxel.
The safety profile of the combination was consistent with prior clinical experience, reported Mark Linch, MD, PhD, of University College London, at the American Association for Cancer Research virtual meeting.
“Responders were enriched with patients with mismatch repair or BRCA 1 and 2,” said Linch. “We believe that further study of nivolumab and ipilimumab i biomarker-selected patients with metastatic castration-resistant prostate cancer is warranted. The rate of incomplete treatment was high, suggesting that alternate dosing schedules may be required.”
Single-agent checkpoint inhibition has shown limited activity in mCRPC. However, studies of anti-CTLA4 therapy have shown induction of T-cell infiltration into prostate cancer, and combined anti-CTLA4 and anti-PD1 therapy has produced a signal of clinical activity in unselected patients with mCRPC, Linch said. In the CheckMate 650 trial, for example, nivolumab plus ipilimumab produced objective responses in 10% of patients previously treated with docetaxel, increasing to 25% in patients in the pre-docetaxel setting.
Previous studies showed that about 20% of patients with mCRPC have high levels of tumor-infiltrating lymphocytes in the primary tumor. Investigators hypothesized that mCRPC associated with an immunogenic signature — DNA damage repair (DDR), defective mismatch repair (dMMR), or high inflammatory infiltrate — would be more likely to respond to the combination of nivolumab and ipilimumab.
Linch reported initial findings from the ongoing phase II NEPTUNES trial. Eligible patients are receiving androgen deprivation therapy, have archival cancer tissue or disease amenable to biopsy, and have received at least one prior line of systemic therapy for metastatic prostate cancer (hormone-sensitive or CRPC).
The first cohort received nivolumab 1 mg/kg q21d plus ipilimumab 3 mg/kg for a maximum of four cycles. Following a 6-week break, patients received single-agent nivolumab once a month for up to a year. A second cohort that began enrollment last fall is evaluating nivolumab 3 mg/kg plus ipilimumab 1 mg/kg. Linch’s presentation was limited to the first cohort.
The primary endpoint was a composite response rate comprising radiologic response, PSA response (≥50% reduction), and conversion of circulating tumor cells (CTCs) at 9 weeks. Secondary endpoints included overall survival (OS), progression-free survival (PFS), and safety.
A response rate of ≥40% was considered a favorable outcome, and a response rate ≤20% was an unfavorable outcome. Assessment of the primary endpoint required a minimum of 35 patients with an immunogenic signature by at least one criterion, with objective responses in 12 to meet statistical requirements for a favorable outcome.
Investigators screened 184 patients, 35 (19%) with insufficient tissue to assess immunogenic signature, 62 (34%) with a positive immunogenic signature, and 87 (47%) with a negative immunogenic signature. Data analysis included 35 patients who received the immunotherapy combination.
The study population had a median age of 66.5. Half the patients had received three or more prior lines of therapy, 11% had received one prior systemic therapy, 89% had previous docetaxel treatment, and half had prior radiotherapy.
In addition to the 10 patients who achieved the composite response, eight patients had a PSA response, four had a radiologic response, three had CTC and PSA responses, and one had a CTC response. With a median follow-up of 7.2 months, most responses were ongoing at data censoring, Linch said. Median OS was about 18 months and median PFS was about 5 months.
In analysis of response by type of biomarker, four of five patients with dMMR responded, as well as three of four with BRCA1/2 mutations, and two of 10 with high inflammatory infiltrate (including one patient with CDK12 aberrations in addition to increased inflammatory infiltrate). No responses were observed in the remaining six patients with CDK12 alterations, eight with ATM alterations, two with CHD1 alterations, or one with CHEK2 alterations.
The most common treatment-related adverse events (TRAEs) were diarrhea (57%), rash (54%), fatigue (37%), increased liver enzymes (23%), and fever (20%). Grade 3/4 diarrhea occurred in 17% of patients; otherwise, no other grade 3/4 TRAE occurred in more than 9% of patients.
Eleven of the 35 patients completed all four cycles of immunotherapy. Some patients discontinued because of disease progression, but most did not complete planned therapy because of toxicity, said Linch.
During a discussion that followed the presentation, Linch said investigators are conducting additional analyses to help explain the higher response rates in patients with dMMR and BRCA1/2 mutations.
“The finding that four out of five mismatch repair patients responded wasn’t unexpected because we’ve seen that across the board in other tumor types,” he said. “It’s also previously been seen in prostate cancer, but it looks as though it’s more like 50% would be the expected response, at least to monotherapy. I think this is an interesting finding. The one patients [with dMMR] that didn’t respond actually had an incomplete partial response.”
Ongoing analyses include assessment of tumor status with respect to the androgen receptor splice variant 7, which predicts for resistance to second-generation androgen receptor inhibitors.
“We need more information about biomarkers. We should be directing our future studies in biomarker-selected groups,” said Linch.
The study was supported by Bristol Myers Squibb (BMS).
Linch disclosed relevant relationships with Astellas, AstraZeneca, BioNTech, Pfizer, Merck, BMS, Shionogi, and GlaxoSmithKline.