Results from the phase 3 KEYNOTE-811 study indicated that pembrolizumab and trastuzumab plus chemotherapy yielded robust, long-lasting responses in patients with HER2-positive metastatic gastric or gastroesophageal junction cancer.
Pembrolizumab (Keytruda) and trastuzumab (Herceptin) plus chemotherapy have demonstrated robust, long-lasting responses vs trastuzumab and chemotherapy alone in patients with HER2-positive metastatic gastric or gastroesophageal junction cancer, according to an interim analysis of the phase 3 KEYNOTE-811 study (NCT03615326).1
Investigators reported an overall response rate (ORR) of 74.4% in the treatment arm, which was a statistically significant difference of 22.7% compared with the placebo arm (95% CI, 11.2-33.7; P = .00006). Findings from the phase 3 global trial were presented during the European Society for Medical Oncology 2021 World Congress on Gastrointestinal Cancer.
The interim analysis resulted in an accelerated approval of the immune checkpoint inhibitor combination for the first-line treatment of patients with locally advanced unresectable or metastatic HER2-positive gastric or gastroesophageal junction (GEJ) adenocarcinoma.2
In KEYNOTE-811, 692 patients with HER2-positive disease and an ECOG performance status 0 or 1 were randomized to receive pembrolizumab at 200 mg every 3 weeks and trastuzumab and fluoropyrimidine- and platinum-containing chemotherapy for up to 35 cycles vs placebo and trastuzumab plus chemotherapy. Dual primary end points were overall (OS) and progression-free survival (PFS). Key secondary end points were ORR and duration of response (DOR) per RECIST v1.1.
“These were relatively well patients with excellent functional status and organ function,” lead author Yelena Y. Janjigian, MD, chief, gastrointestinal oncology service at Memorial Sloan Kettering Cancer Center, said during presentation of the data. “It's important to note that patients were permitted to enroll regardless of PD-L1 status.”
Janjigian noted that the data presented was from the protocol-specified first interim analysis.
After the first 260 patients enrolled with ≥ 8.5 months of follow-up, investigators evaluated if adding pembrolizumab to trastuzumab and chemotherapy improved ORR based on the one-sided superiority boundary of P = .002. At the data cutoff of June 17, 2020, 434 patients had enrolled with the first 264 patients evaluated for efficacy. All patients who had received ≥ 1 dose of study medication were evaluated for safety.
In the efficacy population, after a median follow-up of 12.0 months (range, 8.5-19.4), 41% of patients were still enrolled and were continuing with therapy compared with 28% of patients continuing in the placebo arm. In patients undergoing treatment, 5% discontinued because of adverse events (AEs) compared with 8% in the placebo arm.
In the intent-to-treat population, with a median follow-up of 9.9 months (range, 0.1-19.4), similar dispositions were observed. In the treatment arm, 59% of patients continued, compared with 48% in the placebo arm. Treatment discontinuation rates were similar as well: 41% in the treatment arm and 52% in the placebo arm, respectively.
Overall, baseline patient characteristics were relatively well matched between the 2 arms in the efficacy and intent-to-treat populations, Janjigian said.
In the efficacy population, 97% of patients in the treatment arm demonstrated any decrease in size of tumor from baseline; further, 32% of patients demonstrated a decrease in tumor size of ≥ 80%. In the placebo arm, 90% of patients had any decrease in tumor size and 15% of patients had a decrease of ≥ 80%.
“We observed much more deeper responses and more meaningful responses [in the treatment arm] compared with the placebo arm, in which 15% of the population had deep responses of ≥ 80%,” Janjigian said.
The confirmed response rate in the efficacy population also favored the treatment arm. Investigators reported a disease control rate of 96% in the treatment arm vs 89.3% in the placebo arm. Median duration of response in the treatment arm was 10.6 months vs 9.5 months in the placebo arm, with patients in the treatment arm reporting a ≥ 6-month duration at rates of 70% vs 61% in the placebo arm. Nine-month DOR rates were 58% in the treatment arm vs 51% in the placebo arm.
“What is also notable is that in the pembrolizumab arm, in addition to this depth of response and overall response rate of 74%, 11% of patients had complete responses, which is quite remarkable for this combination and is not what we would typically expect in metastatic gastric cancer,” Janjigian said.
There were no new signals observed and the regimen was relatively well tolerated, Janjigian said. The most common AEs reported were chemotherapy-related events. In the treatment and placebo arms, these included diarrhea (53% vs 44%, respectively), nausea (49% vs 44%), and anemia (41% vs 44%).
The incidence of immune-mediated AEs and infusion reactions was higher in the treatment arm compared with the placebo arm (18% vs 13%, respectively), as were the incidence of pneumonitis (5% vs 1%), colitis (5% vs 2%), and hypothyroidism (5% vs 3%). Overall, AEs were similar between the arms of the study.
“An analysis of OS and PFS will be performed in the future, according to the analysis plan. We are already using this [regimen] in the clinic, and it’s well tolerated,” Janjigian concluded.