Background: KEYNOTE-062 (NCT02494583) is a global phase 3 study of pembrolizumab (pembro) as monotherapy and in combination with chemotherapy (chemo) versus chemo as first-line therapy for PD-L1–positive (combined positive score [CPS] ≥1) advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma. At the time of the protocol-specified final analysis, pembro was noninferior to chemo, with fewer adverse events (AEs) observed. Pembro or pembro + chemo was not superior to chemo for the overall survival (OS) and progression-free survival (PFS) end points tested. We present the results of KEYNOTE-062 after ̃25 additional months of follow-up (cutoff: April 19, 2021). Methods: Patients with previously untreated gastric or GEJ cancer were randomly assigned 1:1:1 to pembro 200 mg Q3W, pembro + chemo (cisplatin 80 mg/m²/day on day 1 + 5-FU 800 mg/m²/day on days 1-5 Q3W [or capecitabine 1000 mg/m² twice daily on days 1-14 Q3W per local guidelines]), or placebo Q3W + chemo. Primary end points were OS in the CPS ≥1 and CPS ≥10 populations for pembro + chemo versus chemo and pembro versus chemo and PFS (RECIST v1.1; central review) in the CPS ≥1 and CPS ≥10 populations for pembro + chemo versus chemo. Safety was also evaluated. Results: At the time of data cutoff, 689 of 763 patients (90.3%) had died. Median follow-up (defined as time from randomization to data cutoff) was 54.3 months (range, 46.8-66.1). Pembro was noninferior to chemo for OS in the CPS ≥1 population (median, 10.6 vs 11.1 months; HR, 0.90; 95% CI, 0.75-1.08) but had a clinically meaningful OS benefit in the CPS ≥10 population (median, 17.4 vs 10.8 months; HR, 0.62; 95% CI, 0.45-0.86). 24-month OS rates (pembro vs chemo) were 26.6% versus 18.8% in the CPS ≥1 population and 39.1% versus 21.1% in the CPS ≥10 population. Pembro + chemo was not superior to chemo for OS in the CPS ≥1 (median, 12.5 vs 11.1 months; HR, 0.85; 95% CI, 0.71-1.02) or the CPS ≥10 (median, 12.3 vs 10.8 months; HR, 0.76; 95% CI, 0.56-1.03) population. 24-month OS rates (pembro + chemo vs chemo) were 24.5% versus 18.8% in the CPS ≥1 population and 28.3% versus 21.1% in the CPS ≥10 population. Pembro + chemo did not significantly prolong PFS versus chemo in the CPS ≥1 (median, 6.9 vs 6.5 months; HR, 0.84; 95% CI, 0.70-1.01) or the CPS ≥10 (median, 5.8 vs 6.2 months; HR, 0.71; 95% CI, 0.52-0.96) population. Grade 3-5 treatment-related AEs rates were 17.3% (pembro), 73.2% (pembro + chemo), and 69.3% (chemo). Conclusions: After ̃25 additional months of follow-up, efficacy and safety outcomes with first-line pembro or pembro + chemo versus chemo in patients with gastric or GEJ adenocarcinoma enrolled in KEYNOTE-062 were consistent with the final analysis data. Clinical trial information: NCT02494583.