Two-Year Follow-up of Keynote-087 Study: Pembrolizumab Monotherapy in Relapsed/Refractory Classic Hodgkin Lymphoma

Introduction: The near-universal genetic amplification events at 9p24.1 in classic Hodgkin lymphoma (cHL) results in overexpression of the programmed death 1 (PD-1) ligands and betrays an unusual dependence on the PD-1 pathway. Inhibition of this pathway by use of pembrolizumab has shown effective antitumor activity and acceptable safety in patients with relapsed or refractory cHL (R/RcHL) in the multicohort KEYNOTE-087 study. This led to FDA approval of pembrolizumab for the treatment of adult and pediatric patients who have refractory cHL or who have relapsed after ≥3 prior lines of therapy. A critical remaining question is the durability of responses, specifically whether a subgroup of patients can have durable remission with PD-1 blockade. Therefore, we present the results for the total population and by cohort, with an additional ~12 months of follow-up from last presentation.

Methods: The multicenter, single-arm, phase 2 KEYNOTE-087 (NCT02453594) study was conducted to evaluate pembrolizumab in patients with R/R cHL that progressed after autologous stem cell transplantation (ASCT) and subsequent brentuximab vedotin (BV) therapy (cohort 1); salvage chemotherapy and BV (cohort 2); or ASCT but not treated with BV after ASCT (cohort 3). Patients received pembrolizumab 200 mg intravenously every 3 weeks. Response was assessed every 12 weeks per 2007 Revised Response Criteria for Malignant Lymphomas. Primary end points were safety and overall response rate (ORR) per blinded independent central review in all patients and in each cohort; secondary end points were complete remission rate (CRR), progression-free survival (PFS), duration of response (DOR), and overall survival (OS). All patients who received at least 1 dose of pembrolizumab were included in the analyses.

Results: At data cutoff (Mar 21, 2018), median follow-up was 27.6 mo (range, 1.0-32.9), and 5 of 210 enrolled patients were still in treatment. Baseline characteristics were previously presented (Chen et al. J Clin Oncol. 2017; 35(19):2125-2132). In the total population, ORR was 71.9% (95% CI, 65.3-77.9); CRR, 27.6% (95% CI, 21.7-34.2); partial response (PR), 44.3% (95% CI, 37.5-51.3). Response rates by cohort were as follows: cohort 1 (n=69): ORR: 76.8%; CRR: 26.1%; cohort 2 (n=81): ORR: 66.7%; CRR: 25.9%; cohort 3 (n=60): ORR: 73.3%; CRR: 31.7%. Median DOR was 16.5 mo overall (range 0.0+ to 27.0+; [+, no progressive disease at last assessment]). Of 151 responders, 87 (75.6%) had response ≥6 mo; 61 (58.5%) had response ≥12 mo; 16 (42.5%) had response ≥24 mo; 37 (24.5%) pts had ongoing response. Median DOR by cohort was 22.1 mo in cohort 1, 11.1 mo in cohort 2, and 24.4 mo in cohort 3 (Table). In patients with CR (n=58), median DOR was not reached (NR) in the total population and was 25.0 mo in cohort 1, 19.2 mo in cohort 2, and NR in cohort 3. In patients with PR (n=93), median DOR was 10.9 mo overall, 19.5 mo in cohort 1, 7.9 mo in cohort 2, and 13.9 mo in cohort 3. Median PFS for all patients was 13.7 mo (95% CI, 11.1-17.0) (Table); 24-mo PFS rate was 31.3%. In patients with CR, median PFS was NR in the total population, 27.6 mo in cohort 1, 21.9 mo in cohort 2, and NR in cohort 3 (Table). In patients with PR, median PFS was 13.8 mo in the total population, 22.2 mo in cohort 1, 13.4 mo in cohort 2, and 19.4 mo in cohort 3. Median OS was not reached in the total population or in any cohort; 24-mo OS rate was 90.9% in the overall population, 92.5% in cohort 1; 90.6% in cohort 2, and 89.4% in cohort 3 (Table). Any-grade treatment-related adverse events (AEs) occurred in 153 (72.9%) patients; those occurring in ≥10% patients were hypothyroidism (14.3%), pyrexia (11.4%), fatigue (11.0%), and rash (11.0%). Grade 3/4 treatment-related AEs occurred in 25 (11.9%) pts, most commonly, neutropenia (5 [2.4%]) and diarrhea (3 [1.4%]); none resulted in death. Treatment-related AEs led to discontinuation in 14 (6.7%) patients.

Conclusions: With more than 2 years of median follow-up, pembrolizumab continued to demonstrate effective antitumor activity with high ORR, durable response, and manageable safety in patients with R/R cHL in 3 cohorts of patients in whom treatment history was different. ORR, DOR, and PFS seemed higher in patients in cohorts 1 and 3 than in cohort 2, in which patients likely had more chemoresistant disease; nonetheless, pembrolizumab is still an effective treatment in this hard-to-treat patient population.

Table.

View largeDownload slide

Table.

View largeDownload slide

Close modal