ASCO 2019: Front-line Pembrolizumab as Alternative to Chemotherapy in Advanced Gastric and GEJ Cancer
In patients with newly diagnosed gastric cancers, pembrolizumab led to non-inferior OS rates compared to standard chemotherapy, while tolerability was improved. For patients harboring tumors with high PD-L1 expression, pembrolizumab showed clinically meaningful OS improvement.
The KEYNOTE-062 trial compared the efficacy of pembrolizumab versus standard chemotherapy in 763 previously untreated patients with PD-L1 positive, HER2-negative gastric or GEJ cancers (69 percent gastric cancers, 30 percent GEJ cancers). 26 percent had previously undergone gastric surgery.
PD-L1 expression was assessed using the combined positive score (CPS) that is estimated according to the number of PD-L1 positive cells in biopsy specimens as well as live tumor cells. In previous studies, a benefit from pembrolizumab had been seen in patients with CPS ≥1, while a CPS ≥10 indicates a higher probability of response. Patients with HER2 positive tumors were excluded. In the current trial, only patients with PD-L1 CPS ≥1 were included; 281 (37% of the participants) had a score of 10 or more.
Patients were randomized 1:1:1 to receive pembrolizumab monotherapy (P), chemotherapy (cisplatin plus 5-fluorouracil or capecitabine) plus placebo (C), or pembrolizumab plus chemotherapy (P+C). The primary endpoint was overall survival (OS) and progression-free survival (PFS) in patients with CPS ≥1 and ≥10.
Improved OS for pembrolizumab in patients with CPS ≥10
After a median follow-up of 11.3 months, the primary endpoint was met, with OS being noninferior in patients treated with pembrolizumab compared to those treated with standard chemotherapy. Median OS was comparable between the two arms, with 10.6 months with pembrolizumab, and 11.1 months with chemotherapy (HR 0.91).
In patients with a CPS ≥10, median OS with pembrolizumab was significantly and clinically improved compared to chemotherapy: 17.4 vs.10.8 months, (HR 0.69). At two years, 39 percent of patients with a CPS-score over ten were alive, compared with 22% who received standard chemotherapy.
Noteworthy, no OS or PFS difference was observed between the combination arm P+C and the chemotherapy arm (CPS ≥1 12.5 months, and CPS ≥10 12.3 months). According to lead author MD PhD Josep Tabernero, Vall d’Hebron University Hospital, Barcelona, it is not yet clear why there is no synergistic effect of chemotherapy and immunotherapy.
Reduced toxicity with pembrolizumab
The rate of severe adverse events (SAE) was lowest in the pembrolizumab arm: 17% of patients showed SAEs, 69% in the chemotherapy arm, and 73% in the combination arm 73%. The most frequent side effects were nausea and fatigue. The safety profile of pembrolizumab monotherapy corresponded to what has been previously observed.
Outlook
Further subgroup analysis is planned to specify which patient subgroups benefit most from immunotherapy. Tabernero noted that more biomarkers than PD-L1 alone are necessary to identify potential responders of pembrolizumab monotherapy or combination therapy.
Tabernero J et al.: Pembrolizumab with or without chemotherapy versus chemotherapy for advanced gastric or gastroesophageal junction (G/GEJ) adenocarcinoma: The phase III KEYNOTE-062 study; ASCO 2019, Abstract LBA4007