Pembrolizumab Appears Safe Across Melanoma, NSCLC, RCC in Pooled Analysis

Pembrolizumab (Keytruda) induced treatment-related adverse effects (TRAEs) that were generally mild or moderate in severity, according to finding from a pooled analysis of more than 4000 patients with melanoma, non–small cell lung cancer (NSCLC), or renal cell carcinoma (RCC).

In data from a poster presented at the 2022 ESMO Immuno-Oncology Annual Congress, investigators found that 78.6% of patients assigned to pembrolizumab (n = 2060) experienced any-grade TRAEs compared with 58.7% of those assigned to placebo (n = 2065). In the pembrolizumab arm, 16.3% of patients experienced grade 3 to 5 TRAEs and 11.6% experienced serious TRAEs compared with 3.5% and 1.5%, respectively, in the placebo arm. Among patients assigned to pembrolizumab, 15.8% experienced TRAEs leading to discontinuation and 0.2% had TRAEs leading to death, compared with 2.1% and 0%, respectively, with placebo.

“The safety profile of adjuvant pembrolizumab in this analysis was consistent in severity and management with the established profile of pembrolizumab in advanced disease,” Jason Luke, MD, FACP, an associate professor of Medicine in the Division of Hematology/Oncology and the director of the Cancer Immunotherapeutics Center at the UPMC Hillman Cancer, and colleagues, wrote. “These data show that pembrolizumab has a manageable safety profile when used in the adjuvant setting and thus support the use of adjuvant pembrolizumab in patients with resected solid tumors.”

Investigators cautioned that although serious AEs are rare, physicians should be aware of the risk to inform and monitor their patients. They also noted that it is not possible to assess long-term irreversible toxic effects or late-onset AEs because the study reporting criteria did not fully support evaluation of all AE resolution or onset beyond 90 days after stopping treatment.

Investigators pooled patients from 4 phase 3 clinical trials:

• KEYNOTE-716 (NCT03553836): Patients with completely resected stage IIB or IIC melanoma (N = 969)
  • Median time from randomization to data cutoff: 27.4 months (range, 14.0-39.4)

• Median duration of pembrolizumab treatment: 11.1 months (range, 0-16.4)

• KEYNOTE-054 (NCT02362594): Patients with completely resected stage IIIA, IIIB, or IIIC melanoma (N = 1013)
  • Median time from randomization to data cutoff: 59.1 months (interquartile range, 57.3-61.6)
  • Median duration of pembrolizumab treatment: 11.8 months (range, 0-15.7)
• PEARLS/KEYNOTE-091 (NCT02504372): Patients with completely resected stage IB, II, or IIIA NSCLC (N = 1161)
  • Median time from randomization to data cutoff: 35.6 months (range, 16.5-68.0)
  • Median duration of pembrolizumab treatment: 11.7 months (range 0-18.9)
• KEYNOTE-564 (NCT03142334): Patients with post-nephrectomy/metastasectomy RCC with a clear-cell component at high risk of recurrence (N = 984)
  • Median time from randomization to data cutoff: 30.1 months (range, 20.8-47.5)
  • Median duration of pembrolizumab treatment: 11.1 months (range, 0-14.3)

Patients in KEYNOTE-564 and KEYNOTE-716 received adjuvant pembrolizumab 200 mg (2 mg/kg up to 200 mg for pediatric patients) or placebo every 3 weeks for up to 17 cycles. In KEYNOTE-054 and PEARLS/KEYNOTE-091, patients received the same dose for up to 18 cycles.

The analysis population included all randomly assigned patients who received at least 1 dose of study treatment. Investigators monitored for AEs for up to 30 days after treatment completion and up to 90 days for serious AEs. Investigators determined whether AEs were treatment related. Immune-mediated AEs and infusion reactions were based on a list of terms prepared by the sponsor and included AEs regardless of attribution to study treatment by the investigator.

In the pembrolizumab arm, 48.2% of patients had melanoma, 28.2% had NSCLC, and 23.7% had RCC. In the placebo arm, the distribution was 47.8%, 28.1%, and 24.0%, respectively.

The median patient age was 61.0 years (range, 16-88) in the placebo group including 36.8% aged 65 years or older. In the placebo group, the median patient age was 61.0 years (range, 17-87) including 38.4% who were 65 years or older. Men made up about 65% of patients in both groups.

In the pembrolizumab arm, 36.9% of patients experienced any immune-mediated AE or infusion reaction, 8.8% of which were grade 3 to 5. The most common such events were hypothyroidism (18.5%), hyperthyroidism (11.0%), and pneumonitis (4.0%).

In the placebo arm, 9.3% of patients experienced any immune-mediated AE or infusion reaction, 1.1% of which were grade 3 to 5. The most common such events were hypothyroidism (3.7%), hyperthyroidism (1.3%), and pneumonitis (1.4%).

Median time to onset of first immune-mediated AE or infusion reaction was 2.6 months with pembrolizumab vs 4.1 months with placebo.

More than 90% of patients in both treatment groups experienced any-cause AEs. In the pembrolizumab arm, 31.7% of such AEs were grade 3 to 5 and 23.0% were serious, compared with 20.9% and 15.7%, respectively, with placebo. The rate of discontinuation due to any-cause AEs was 18.1% with pembrolizumab and 4.1% with placebo. Fifteen (0.7%) and 12 (0.6%) patients died due to any-cause AEs in the pembrolizumab and placebo arms, respectively.

Reference

Luke JJ, Robert C, Carlino MS, et al. Safety profile of adjuvant pembrolizumab (pembro) in melanoma, non-small cell lung cancer (NSCLC), and renal cell carcinoma (RCC): Pooled analysis of phase III clinical trials. Ann Oncol. 2022;16(suppl_1): 100102-100102. doi:10.1016/iotech/iotech100102