Aurelien Marabelle, MD

As reported in the Journal of Clinical Oncology by Aurelien Marabelle, MD, PhD, of Gustave Roussy, Institut National de la Santé et de la Recherche Médicale, Villejuif, and colleagues, the phase II KEYNOTE-158 trial has shown robust activity of pembrolizumab in patients with noncolorectal high microsatellite instability (MSI-H)/mismatch repair–deficient (dMMR) solid tumors.¹ Differences in objective response rates according to cancer type were observed.

Findings in the study supported the May 2017 accelerated approval of pembrolizumab for treatment of patients with unresectable or metastatic MSI-H or dMMR solid tumors progressing after prior treatment and who have no satisfactory alternative treatment options or with MSI-H or dMMR colorectal cancer progressing after treatment with fluoropyrimidine, oxaliplatin, and irinotecan.

Study Details

In the study, 233 patients with previously treated advanced noncolorectal tumors were enrolled between February 2016 and May 2018 from 55 sites in 18 countries. Overall, patients had 27 tumor types, with the most common being endometrial, gastric, cholangiocarcinoma, and pancreatic. Treatment consisted of pembrolizumab at 200 mg once every 3 weeks for 2 years (35 cycles) or until disease progression, unacceptable toxicity, or patient withdrawal. The primary endpoint was objective response according to Response Evaluation Criteria in Solid Tumors, version 1.1, on independent central radiologic review.

Among all patients, the median age was 60 years (37% ≥ 65 years); 59% were female; all had an Eastern Cooperative Oncology Group performance status of 0 (48%) or 1; 98% had disease stage M1; 2% had brain metastases; 24% had received (neo)/adjuvant therapy; and the number of prior lines of therapy for recurrent/metastatic disease was 0 in 3%, 1 in 37%, 2 in 26%, 3 in 18%, and at least 4 in 16%.

Responses

At data cutoff for the current analysis (December 2018), the median follow-up was 13.4 months (range = 0.4–34.2 months). Among all patients, an objective response was observed in 80 patients (34.3%), including a complete response in 23 (9.9%). The median time to response was 2.1 months. The median duration of response had not been reached at the time of analysis (range = 2.9 to 31.3+ months). The median progression-free survival was 4.1 months, and the median overall survival was 23.5 months.

Outcomes in patients with the most common tumors follow:

Endometrial cancer (n = 49): response in 20 patients = 57.1%, including complete response in 8.

• Median response duration = not reached (range = 2.9 to 27.0+ months)
• Median progression-free survival = 25.7 months; median overall survival = not reached (95% confidence interval [CI] = 27.2 months to not reached).

Gastric cancer (n =24): response in 11 patients = 45.8%, including complete response in 4

• Median response duration = not reached (range = 6.3 to 28.4+ months)
• Median progression-free survival = 11.0 months; median overall survival = not reached (range = 7.2 months to not reached)

Cholangiocarcinoma (n = 22): response in 9 patients = 40.9%, including complete response in 2

• Median response duration = not reached (range = 4.1+ to 24.9+ months)
• Median progression-free survival = 24.3 months; median overall survival = not reached (95% CI = 6.5 months to not reached)

Pancreatic cancer (n =22): response in 4 patients = 18.2%, including complete response in 1

• Median response duration = 13.4 months (range = 8.1 to 16.0+ months)
• Median progression-free survival = 2.0 months; median overall survival = 4.0 months

Small intestine cancer (n = 19): response in 8 patients = 42.1%, including complete response in 3

• Median response duration= not reached (range = 4.3+ to 31.3+ months)
• Median progression-free survival = 9.2 months; median overall survival = not reached (95% CI = 10.6 months to not reached)

Ovarian cancer (n = 15): response in 5 patients = 33.3%, including complete response in 3

• Median response duration = not reached (range = 4.2 to 20.7+ months)
• Median progression-free survival = 2.3 months; median overall survival = not reached (95% CI = 3.8 months to not reached)
• Among 13 patients with brain lesions, no responses were observed; median progression-free survival and overall survival were 1.1 months and 5.6 months, respectively.

Adverse Events

Treatment-related adverse events occurred in 64.8% of patients, with the most common being fatigue (14.6%), pruritus (12.9%), diarrhea (12.0%), and asthenia (10.7%). Grade ≥ 3 treatment-related adverse events occurred in 14.6% of patients; the most common grade 3 events were increased gamma-glutamyl transferase (1.7%) and pneumonitis (1.3%), and grade 4 events occurred in three patients (Guillain-Barre syndrome in one, increased alanine aminotransferase in one, and decreased neutrophils and enterocolitis in one). Treatment-related serious adverse events occurred in 7.7% of patients. Treatment-related adverse events led to pembrolizumab discontinuation in 9.4%.

Immune-mediated adverse events or infusion reactions of any grade occurred in 23.2% of patients, with the most common being hypothyroidism, hyperthyroidism, colitis, and pneumonitis. Grade ≥ 3 events occurred in 6.0%; grade 3 events consisted of pneumonitis, severe skin reactions, hepatitis, hyperthyroidism, colitis, type 1 diabetes, Guillain-Barre syndrome, and pancreatitis; grade 4 reactions consisted of Guillain-Barre syndrome and colitis. Immune-mediated adverse events led to treatment discontinuation in 5.2%. Grade 5 pneumonia occurred in one patient; no other fatal treatment-related adverse events were observed.

The investigators concluded: “Our study demonstrates the clinical benefit of anti–programmed death-1 therapy with pembrolizumab among patients with previously treated unresectable or metastatic MSI-H/dMMR noncolorectal cancer. Toxicity was consistent with previous experience of pembrolizumab monotherapy.” 

DISCLOSURE: The study was funded by Merck Sharp & Dohme. For full disclosures of the study authors, visit jco.ascopubs.org.

REFERENCE

1. Marabelle A, Le DT, Ascierto PA, et al: Efficacy of pembrolizumab in patients with noncolorectal high microsatellite instability/mismatch repair-deficient cancer: Results from the phase II KEYNOTE-158 study. J Clin Oncol 38:1-10, 2020.