Immunotherapy Lurches Forward in Prostate Cancer

SAN FRANCISCO — Combinations involving PD-1 checkpoint blockade showed moderate activity for patients with metastatic castration-resistant prostate cancer (mCRPC) in a pair of early-phase studies.

In CheckMate 650, the overall response rate was 17.7% with the anti-CTLA-4/PD-1 combination of ipilimumab (Yervoy) and nivolumab (Opdivo), reported Padmanee Sharma, MD, PhD, of MD Anderson Cancer Center in Houston.

And in Keynote-365, radiographic responses were seen in 7% of docetaxel-treated patients following the combination of PD-1 inhibitor pembrolizumab (Keytruda) plus PARP inhibitor olaparib (Lynparza), according to Evan Y. Yu, MD, of the University of Washington in Seattle.

Findings from both trials were presented here at the Genitourinary Cancers Symposium.

CheckMate 650

“In a malignancy where immune checkpoint monotherapy has shown limited activity, nivolumab plus ipilimumab demonstrated antitumor activity in patients with metastatic CRPC,” said Sharma, who explained that while single-agent ipilimumab had previously displayed some clinical activity in mCRPC, anti-PD-1/PD-L1 monotherapies have shown limited clinical benefit in this setting, likely due to an “immunologically cold tumor microenvironment.”

The phase II study included 90 patients split into two cohorts. Most patients (>50%) had already received treatment with abiraterone (Zytiga) or enzalutamide (Xtandi) prior to enrollment.

Among the evaluable patients in cohort 1, responses were seen in 25% of patients, with two patients achieving a complete response. This group was made up of asymptomatic or minimally symptomatic patients that had received second-generation androgen-directed therapies but no chemotherapy in the mCRPC setting.

In cohort 2, which included chemotherapy-treated patients, 10% responded, with two patients having complete responses.

Prostate-specific antigen (PSA) response was seen in 17.6% of patients in cohort 1 and 10.0% of those in cohort 2. Median radiographic progression-free survival (PFS) was 5.5 months in cohort 1 and 3.8 months in cohort 2, while overall survival (OS) was 19.0 and 15.2 months, respectively.

Adverse events (AEs) — an issue that has dogged this combination in other malignancies — were notable, with 51.1% and 44.4% of patients in cohorts 1 and 2, respectively, stopping treatment due to toxicities.

“The thing that’s dramatic about this is if the drug was tolerable or not,” said session discussant William Kevin Kelly, DO, of the Sidney Kimmel Cancer Center in Baltimore, noting that the median duration of therapy didn’t exceed 2.1 months.

He pointed to a trial of the ipilimumab-nivolumab in renal cell carcinoma where 79% of patients received all four treatments, and one in melanoma where 50% did.

In the current study, only 33% of mCRPC patients in cohort 1 and 24% of those in cohort 2 received the planned four treatments.

“For some reason prostate cancer patients are not tolerating therapy as well as the other populations,” he said. “As I teach all my fellows, if we can’t get the drug in, patients don’t respond.”

PD-L1 expression in the two groups combined did not appear to predict responses, but tumor mutational burden (TMB) did, with nearly all responses in the two cohorts coming in the high-TMB group (9 of 18 patients) rather than the TMB-low group (1 of 19). Radiographic PFS was prolonged in the TMB-high group as well (7.4 vs 2.4 months, P<0.0001).

Keynote-365

This study included 41 patients who had no more than two previous second-generation androgen-directed therapies, and no more than one chemotherapy other than docetaxel. Visceral disease was present in 41% of patients (12% with liver involvement, 29% without).

Yu highlighted that PSA declined in 12% of patients. Median radiographic PFS and OS were 4.7 months and 13.5 months, respectively.

Grade 3/4 AEs occurred in 49% of patients, with anemia (27%), fatigue (7%), and asthenia (7%) being the most common. No grade 3/4 immune-mediated AEs occurred in the study.

“Pembrolizumab plus olaparib seemed generally well-tolerated, and has some activity in the molecularly unselected post-docetaxel setting and post-second-generation hormone setting,” said Yu, who announced that a phase III study will study the combination following docetaxel and one second-generation agent in these patients.

“The results aren’t earth-shattering,” said moderator Nicholas John van As, MBBCh, of the Royal Marsden NHS Foundation Trust in London, following the presentation. He questioned why the investigators would continue on with a molecularly unselected cohort.

Yu acknowledged that previous studies had shown high responses in patients with homologous recombination deficiency (HRD) mutations (e.g., 88% in the TOPARP-A trial). But he said the Keynote-365 investigators couldn’t definitively find any patients who were actually HRD-positive, and noted that preclinical data have suggested mechanisms of response for HRD-negative patients as well.