Patients with recurrent urothelial cancer lived longer when they received pembrolizumab (Keytruda) instead of chemotherapy as second-line treatment, according to long-term follow of a randomized trial.
After a median follow-up of 28 months, patients treated with pembrolizumab had a median survival of 10.3 months versus 7.3 months for those who received chemotherapy. Both 12- and 24-month survival was significantly better in the group treated with the immunotherapeutic drug, according to Joaquim Bellmunt, MD, of Dana-Farber Cancer Institute in Boston, and colleagues.
“Pembrolizumab is the first immunotherapy to demonstrate superior survival over chemotherapy in patients with advanced urothelial carcinoma after failure of platinum-based therapy,” Bellmunt said. “This study provides level 1 evidence that supports the use of pembrolizumab as a standard of care for this patient population.”
Data from the trial, known as KEYNOTE 045, provided the basis for approval of pembrolizumab for advanced urothelial carcinoma, irrespective of PD-L1 status, in the U.S., Europe, and Japan, he added. The 2-year follow-up data remained consistent with the data that supported the approval.
A different PD-1/PD-L1 inhibitor failed to demonstrate an advantage over chemotherapy for PD-L1-positive locally advanced/metastatic urothelial cancer that progressed or relapsed after initial platinum-based chemotherapy. As previously reported, patients treated with atezolizumab (Tecentriq) had a median overall survival (OS) of 11.1 months versus 10.6 months for investigator’s choice of chemotherapy. An intention-to-treat (ITT) analysis of all treated patients, irrespective of PD-L1 status, yielded a similar result, reported Thomas Powles, MD, of Barts Cancer Institute in London.
The pembrolizumab results should increase confidence in second-line use of the drug, said invited discussant Robert Jones, MD, PhD, of the University of Glasgow in Scotland.
“This helps our patients make an informed decision about whether or not to accept this treatment,” said Jones. “The results remain in keeping with the possibility of a long immunotherapy [survival] tail. None of these data support a role for second-line cytotoxics after failure of platinum in preference to a checkpoint inhibitor.”
The pembrolizumab data affirmed findings initially reported at the 2016 Society for Immunotherapy of Cancer, followed by publication in the New England Journal of Medicine. At that point, after a median follow-up of 14 months, the median OS was 10.3 versus 7.4 months for the pembrolizumab and chemotherapy arms, respectively.
KEYNOTE 045 involved 542 patients whose disease had progressed or relapsed after first-line platinum-based chemotherapy. Almost half the patients had two or more high-risk characteristics.
The patients were randomized to pembrolizumab or the investigators’ choice of three different chemotherapy options: paclitaxel, docetaxel, or vinflunine. The trial had coprimary endpoints of OS and progression-free survival (PFS), as assessed in the ITT population and according to PD-L1 status (using ≥10% PD-L1 expression in tumor cells, lymphocytes, and macrophages to define positivity).
The initial results in favor of pembrolizumab represented a 27% reduction in the survival hazard (P=0.0022). The updated data reflected a 30% reduction in the survival hazard (95% CI 0.57-0.85, P=0.00017). The 12-month survival was 44.4% with pembrolizumab and 29.8% with chemotherapy, and the 24-month survival was 27.0% versus 14.3% with pembrolizumab and chemotherapy, respectively.
“By 24 months, 60% of patients in the chemotherapy arm had received an immunotherapeutic agent, including those who received pembrolizumab at crossover,” said Bellmunt.
Subgroup analysis demonstrated a consistent survival advantage for patients treated with pembrolizumab.
Analysis by PD-L1 status showed a median OS of 8.0 months with pembrolizumab and 4.9 months with chemotherapy in the PD-L1-positive patients (n=124) and 10.8 versus 7.7 months in the PD-L1-negative group.
Median PFS did not differ significantly between treatment groups after 14 or 28 months of follow-up (2.1 vs 3.3 months) although the proportion of patients who remained progression free at 12 months (18.4% vs 9.5%) and 24 months (12.5% vs 2.5%) favored pembrolizumab.
Objective response rate was twice as high with the PD-1 inhibitor than with chemotherapy (21.1% vs 11.0%).
Pembrolizumab was associated with a more favorable adverse-event profile, as patients treated with chemotherapy had more fatigue, diarrhea, asthenia, anemia, constipation peripheral sensory neuropathy, peripheral neuropathy, decreased neutrophil count, neutropenia, and alopecia. Immune-related adverse events occurred more often with pembrolizumab: hypothyroidism, pneumonitis, hyperthyroidism, and colitis.
Investigators in the atezolizumab study, known as IMvigor211, performed extensive exploratory analyses to gain insight into the negative result. They found a correlation between DNA damage response (DDR) mutations tumor mutational burden (TMB). Additional analysis showed no association between DDR and efficacy. However, they identified a significant benefit of atezolizumab in the small subgroup of patients (about 100 of 931) who had high TMB and tested positive for PD-L1 expression (IC 2/3): median OS of 17.8 versus 10.6 months, representing a 50% reduction in the hazard ratio (95% CI 0.29-0.86).
In his review of the two trials, Jones concluded that neither provided compelling evidence of a biomarker to predict response to PD-1/PD-L1 inhibition.