In brief
- KEYNOTE-867 randomly assigned 448 patients with unresected stage I or II NSCLC to stereotactic body radiotherapy (SBRT) plus pembrolizumab or SBRT plus placebo.
- Median event-free survival (EFS) was 31.2 months with pembrolizumab and 28.3 months with placebo (hazard ratio 0.92; P = .2932) at the first interim analysis.
- Grade 3 to 5 treatment-related adverse events occurred in 20.4% of patients given pembrolizumab and in 3.7% of those given placebo.
- All five treatment-related deaths (2.2%) occurred in the pembrolizumab arm.
- On the basis of the benefit-risk profile, the trial was stopped early and all study treatment was discontinued.
Study at a glance
- Trial: KEYNOTE-867 (NCT03924869), phase III, randomized, double-blind
- Population: 448 adults with untreated stage I or II (T1 to limited T3, N0, M0) NSCLC that was inoperable, or who refused surgery
- Treatment: SBRT to the primary tumour over 2 weeks or less, with pembrolizumab 200 mg (n = 227) or placebo (n = 221) every 3 weeks for up to 17 cycles
- Primary endpoint: EFS by blinded independent central review
- Follow-up: median 20.6 months (range, 1.3 to 56.8) at the first interim analysis, data cutoff 11 June 2024
- Efficacy: median EFS 31.2 vs 28.3 months; hazard ratio 0.92 (95% CI, 0.69 to 1.24); P = .2932
- Safety: grade 3 to 5 treatment-related adverse events 20.4% vs 3.7%; five treatment-related deaths, all with pembrolizumab
- Not reported: overall survival, subgroup results and the specific adverse events, none of which appear in the abstract
SBRT is the standard for patients who cannot or will not have surgery
The authors describe SBRT as the standard of care for inoperable stage I or II NSCLC. KEYNOTE-867 evaluated whether adding pembrolizumab, given concomitantly with SBRT to the primary tumour, would improve event-free survival in this population.
Eligible adults had untreated stage I or II disease staged by the eighth edition of the AJCC system, limited to T1 to limited T3 tumours without nodal involvement or distant metastasis. Adults who refused surgery were also eligible. Participants were randomly assigned 1:1 and treatment was blinded.
A hazard ratio of 0.92 that did not separate the arms
At the prespecified first interim analysis, after a median of 20.6 months from random assignment, the 227 patients given pembrolizumab had a median EFS of 31.2 months (95% CI, 22.5 to 39.1). The 221 patients given placebo had a median EFS of 28.3 months (95% CI, 19.8 to 33.1).
The difference of about three months in medians came with a hazard ratio of 0.92 (95% CI, 0.69 to 1.24) and a P value of .2932. The interval spans 1, so the data are compatible with no effect of pembrolizumab on EFS. EFS was assessed by blinded independent central review.
Serious treatment-related toxicity rose more than fivefold with pembrolizumab
Among the 444 participants who received at least one dose of study treatment, 46 of 226 in the pembrolizumab arm (20.4%) had a grade 3 to 5 treatment-related adverse event. In the arm given SBRT with placebo the figure was 8 of 218 (3.7%).
Five patients died of treatment-related causes, and all five had received pembrolizumab, a rate of 2.2% in that arm. No treatment-related death occurred among patients given SBRT with placebo.
“The addition of pembrolizumab to SBRT did not improve EFS and was associated with a higher incidence of treatment-related AEs compared with SBRT alone in participants with inoperable stage I or II NSCLC.”Jabbour et al., Journal of Clinical Oncology (2026)
The trial was stopped early on its benefit-risk profile
According to the authors, KEYNOTE-867 was stopped prematurely on the basis of its benefit-risk profile, and all study treatment was discontinued. The efficacy result therefore comes from the first interim analysis, at a median follow-up of 20.6 months, and the range of follow-up ran from 1.3 to 56.8 months.
The design gives the negative result weight. The trial was randomized, double-blind and placebo-controlled, and its primary endpoint was read by blinded independent central review. In the abstract, the authors draw no conclusion about subgroups, and none is drawn here.
Sources
- Journal of Clinical Oncology. KEYNOTE-867: A Phase III, Randomized, Double-Blind Study of Stereotactic Body Radiotherapy With Pembrolizumab for Unresected Stage I or II Non–Small Cell Lung Cancer (2026-10-28). doi.org
Featuring Lung Summit faculty
This study was co-authored by Lung Summit faculty Nikolaj Frost.
This article was produced independently by the Lung Summit editorial team, without industry funding or input.
