Perioperative Pembrolizumab Improves OS and EFS in Early NSCLC

Pembrolizumab (Keytruda) administered before and after surgery reduced the risk of death by 26% vs placebo in patients with early-stage non–small cell lung cancer (NSCLC), according to 5-year follow-up results of the KEYNOTE-671 study (NCT03425643) presented at the 2025 ESMO Congress.

The 5-year overall survival (OS) rate was 64.6% with the addition of neoadjuvant pembrolizumab to chemotherapy followed by adjuvant pembrolizumab compared with 53.6% for neoadjuvant chemotherapy plus placebo followed by adjuvant placebo (HR, 0.74; 95% CI, 0.59-0.92). Additionally, the 5-year event-free survival (EFS) rates were 49.9% and 26.5% for pembrolizumab and placebo, respectively (HR, 0.58; 95% CI, 0.48-0.69).

"These long-term findings support the use of pembrolizumab in the perioperative setting—given with neoadjuvant chemotherapy and then in the adjuvant setting—as a standard of care for patients with resectable, early-stage NSCLC," lead investigator Heather Wakelee, MD, a medical oncologist at Stanford Health Care, said during a presentation of the results. "There was no detriment in health-related quality of life observed with perioperative pembrolizumab."

What was the long-term follow-up of the KEYNOTE-671 study?

In the phase 3 study, 797 patients were randomly assigned to pembrolizumab (n = 397) or placebo (n = 400) in combination with neoadjuvant chemotherapy or alone as adjuvant therapy. Characteristics were balanced across arms. In the pembrolizumab group, the median age was 63 years, ECOG status was 0 (63.7%) or 1 (36.3%), and the pathological stage at baseline was II (29.7%) and III (70.3%). Tumor stage and nodal status was balanced across T1 to T4 and N0 to N2. Histologic features were also balanced, with 56.9% having nonsquamous tumors. A third of patients (34.8%) had PD-L1 staining of less than 1% and EGFR or ALK mutations were seen in roughly 6.5% of patients.

At the 5-year assessment, the median OS had not yet been reached in the pembrolizumab group compared with 70.7 months for those receiving placebo. OS benefits were seen across all key subgroups, with a statistically significant benefit noted for those with stage III disease (HR, 0.75; 95% CI, 0.58-0.96), N0 nodal status (HR, 0.68; 95% CI, 0.46-0.99), and for those with PD-L1 expression of 1% to 49% (HR, 0.67; 95% CI, 0.45-0.99).

"Although not all reached statistical significance, they all tend to favor pembrolizumab," Wakelee said. At the time of the analysis, 35.8% of events had occurred in the pembrolizumab arm and 45.5% had occurred in the placebo group.

The median EFS was 57.1 months with pembrolizumab compared with 18.4 months for placebo. Pembrolizumab induced a statistically significant improvement in EFS across all key subgroups. "This includes those with PD-L1 of less than 1%," Wakelee noted. In this group, the HR was 0.74 (95% CI, 0.55-0.98).

There was little change in patient-reported quality of life measures throughout the course of the study, with the only decline seen around the time of surgery. The mean change from baseline for both groups was near 0 at the 60.4-month follow-up assessment.

What was the safety profile of pembrolizumab in the KEYNOTE-671 study?

The adverse event (AE) profile remained similar to prior analyses, with no new AEs emerging over time, Wakelee noted. Grade 3-5 treatment-related AEs (TRAE) were experienced by 45.2% of those in the pembrolizumab arm compared with 37.8% of those in the placebo group. Serious TRAEs were seen in 18.4% and 14.8% of patients for pembrolizumab and placebo, respectively. TRAEs led to discontinuation of all study treatments for 13.9% and 5.3% of patients for pembrolizumab and placebo, respectively.

"The 5-year overall survival and event-free survival rates in this perioperative study again sealed the deal that perioperative therapy is the new standard of care," said invited discussant Julie Brahmer, MD, director of the Thoracic Oncology Program and professor of oncology at the Sidney Kimmel Comprehensive Cancer Center.

Brahmer pointed out several lingering questions on perioperative therapy, including whether the benefits were being derived from the neoadjuvant or adjuvant components and if treatment could be personalized. To potentially answer some of these questions, Brahmer called attention to the EA5231 CLEAR study for patients who do not achieve pathological complete response (pCR) and the SWOG S2414 INSIGHT study for those who did achieve a pCR (NCT06732401).

REFERENCE:

Wakelee H, Spicer J, Gao S, et al. Perioperative pembrolizumab in early-stage non-small- cell lung cancer (NSCLC): 5-year follow-up from KEYNOTE- 671. Presented at: 2025 ESMO Congress; October 17-21, 2025; Berlin, Germany. Abstract LBA67.