Adjuvant Pembrolizumab Results in Significant DMFS Improvement Over Placebo in Stage II Melanoma

Adjuvant pembrolizumab (Keytruda) was found to result in a significant improvement in distant metastasis-free survival (DMFS) compared with placebo, with a continued reduction in risk of recurrence and an acceptable safety profile in patients with resected stage IIB or stage IIC melanoma, according to data from the third interim analysis of the KEYNOTE-716 trial (NCT03553836) presented during the 2022 ASCO Annual Meeting.

After a median follow up of 27.4 months (range, 14.0-39.4), patients in the pembrolizumab arm experienced a better 24-month DMFS rate of 88.1% compared with 82.2% for those who received placebo. The 12-month DMFS also favored pembrolizumab with 94.7% vs 90.2% in the placebo arm.

The median DMFS was not yet reached in either the pembrolizumab or placebo arms. The number of events were fewer in the pembrolizumab arm than the placebo, affecting 63 patients (12.9%) vs 95 patients (19.4%), respectively (hazard ratio [HR] 0.64, 95% CI, 0.47-0.88; P = .0029).

“Fewer distant metastases as a first recurrence [were observed] with pembrolizumab vs placebo, and the most common location of first distant metastasis was the lung,” Georgina V. Long, MD, PhD, FRACP, of the Melanoma Institute Australia, The University of Sydney, Royal North Shore and Mater Hospital, said in a presentation on the data. “Sustained reduction in risk of recurrence [was observed] with pembrolizumab vs placebo with longer follow-up.”

The KEYNOTE-716 study enrolled 976 patients who were randomized 1:1 to receive pembrolizumab (n = 487) or placebo (n = 489). In part 1 of the study, investigators administered 200 mg in adults, or 2 mg/kg for pediatric patients, of pembrolizumab intravenously once every 3 weeks for 17 cycles. Placebo was given every 3 weeks.

The primary end point was relapse-free survival (RFS) per investigator assessment, and the secondary end points were DMFS, overall survival (OS), and safety. Investigators defined DMFS as the time from randomization to the first diagnosis of distant metastasis at the third interim analysis which was planned after 146 distant metastasis events. The exploratory end point was health-related quality-of-life. In part 2 of the study, pembrolizumab was administered to all patients whose disease recurred.

The minimum age of eligible patients was 12 years. Patients were eligible if they had newly diagnosed, resected, high-risk, stage II melanoma with a negative sentinel lymph node biopsy and ECOG performance scores of 0 or 1. Patients were stratified by T-category 3b, 4a, and 4b, as well as pediatric status. Long noted a type 1 error was controlled at a one-sided alpha of 2.5% with 2.5% previously allocated to the RFS end point because of this study measure’s statistical significance in the first interim analysis.

DMFS favored pembrolizumab across all 3 T-subcategories. The number of events in the T3b subcategory was 23 in patients receiving pembrolizumab and 31 in patients receiving placebo (HR 0.71; 95% CI, 0.41-1.22), for T4a it was 8 and 20, respectively, (HR 0.42; 95% CI, 0.19-0.96), and for T4b it was 30 and 41, respectively (HR 0.7; 95% CI, 0.44-1.13). No median DMFS rate was reached; however, the 12-month and 24-month DMFS rates favored pembrolizumab in all of these subgroups.

Fewer distant metastasis events were recorded for the pembrolizumab arm at first recurrence compared with placebo. Forty-five patients (9%) in the pembrolizumab arm and 79 patients (16%) in the placebo arm experienced distant metastasis at first recurrence. However, a similar number of patients across both arms experienced distant metastasis events after locoregional occurrence: 18 patients (4%) with pembrolizumab and 16 patients (3%) in the placebo arm. Most events could be found in the lungs (n = 100) affecting 49% of patients receiving pembrolizumab vs 73% of patients receiving placebo.

RFS favored patients who received pembrolizumab as well. At a median of 37.2 months, patients in the pembrolizumab arm had 95 events (19.5%) vs 139 events (28.4%) in the placebo arm (HR, 0.64; 95% CI, 0.50-0.84). The 24-month RFS rate favored pembrolizumab (81.2%) vs placebo (72.8%). Similarly, pembrolizumab had favorable RFS rates across all T-subcategories, and the 24-month RFS rates favored pembrolizumab vs placebo.

Median age of patients in the pembrolizumab arm was 60 years (range, 16-84) and the median age in the placebo arm was 61 years (range, 17-87). Both arms had similar distributions of age and sex among patients, and over 90% of patients in both arms had ECOG performance scores of 0.

No new safety signals were identified. Treatment-related adverse events (TRAEs) affected 83% of the pembrolizumab arm and 64% of the placebo arm. More grade 3 or worse TRAEs were reported in pembrolizumab patients (17%) than patients receiving placebo (5%). Seventy-seven (16%) patients in the experimental arm discontinued treatment because of TRAEs vs 12 patients (2%) in the placebo arm. No patients died as a result of TRAEs.

The TRAEs reported were typical of this treatment, but Long commented, “Please note that fatigue in the placebo arm was 19% and 21% in the pembrolizumab arm, as was pruritis 11% in the placebo arm and 25% in the pembrolizumab arm.”

Pembrolizumab performed favorably across DMFS and RFS variables for patients with resected stage IIB and stage IIC melanoma. Safety signals were consistent with the previous 2 interim analyses, and the quality-of-life data will be presented at a later time during the ASCO meeting.

Reference

1. Distant metastasis-free survival with pembrolizumab versus placebo as adjuvant therapy in stage IIB or IIC melanoma: The phase 3 KEYNOTE-716 study. J Clin Oncol 2022;40(suppl 17):LBA9500. doi10.1200/JCO.2022.40.17_suppl.LBA9500