SAN DIEGO -- The degree of event-free survival (EFS) after surgery in early-stage non-small cell lung cancer (NSCLC) depends on the amount of residual tumor, according to a post hoc analysis of the KEYNOTE-671 study.
The authors also found that perioperative pembrolizumab (Keytruda, Merck) continued to show a benefit over placebo in their new results, which David R. Jones, MD, presented at the World Conference on Lung Cancer (WCLC) 2024 on September 8.
The new analysis suggested that the degree of residual viable tumor (RVT) after neoadjuvant therapy is "associated with poorer EFS time, irrespective of treatment arm," said Jones, chief, Thoracic Service, Department of Surgery, Memorial Sloan Kettering Cancer Center, New York City. He underlined that the results do indicate the "clinically significant EFS benefits" of pembrolizumab in this population "extend to patients with residual viable tumors," with the drug associated with "greater pathologic regression" than placebo.
"This supports the benefit of a perioperative regimen of neoadjuvant pembrolizumab plus chemotherapy and adjuvant pembrolizumab for early-stage NSCLC," he said.
Earlier results of KEYNOTE-671 showed that compared with placebo, the use of pembrolizumab both before and following surgery in patients with resectable stage II, IIIA, or IIIB NSCLC was associated with a 42% reduction in the risk for progression, recurrence, or death. Those were published last year, prior to the US Food and Drug Administration (FDA) approving pembrolizumab with platinum-containing chemotherapy as neoadjuvant treatment and with the continuation of single-agent pembrolizumab as the postsurgical adjuvant treatment for resectable NSCLC.
An exploratory analysis suggested there was an EFS benefit with pembrolizumab vs placebo in patients who experienced a pathologic complete or major pathologic response, but there was a need to study the relationship in a more granular manner.
Jones and colleagues therefore conducted a post hoc analysis of KEYNOTE-671 to determine the association between EFS and the degree of pathologic regression after resection, defined as %RVT.
Jones reminded the audience that the study included 797 individuals with stage II, IIIA, or IIIB NSCLC with N2 stage disease, who had undergone no prior therapy and were able to undergo surgery.
The participants were randomized to up to four cycles of neoadjuvant pembrolizumab or placebo alongside cisplatin-based chemotherapy. Surgery was performed within 20 weeks of the first neoadjuvant dose (or 4-8 weeks after the last dose if less than four cycles were given). Then, 4-12 weeks after surgery, patients were continued with pembrolizumab or placebo for up to 13 cycles.
The median follow-up was 36.6 months, and the current analysis included 320 patients from the pembrolizumab arm who had pathologically evaluable tumors, alongside 300 from the placebo arm.
Jones explained that both the groups were "well matched, specifically for sex, tumor histology, tumor stage, and tumor PD-L1 [programmed death ligand 1] status."
He reported that following neoadjuvant therapy, the median %RVT in the pembrolizumab arm was 29.5% vs 52.0% in the placebo arm. Specifically, 17.5% pembrolizumab patients had a %RVT > 60% vs 35.0% of those in the placebo arm.
The team found that the duration of EFS in pembrolizumab patients was associated with post-resection %RVT.
Among those with a %RVT of 0% to ≤ 5%, 24-month EFS was 90.8%, which fell to 74.7% in those with a %RVT > 5% to ≤ 30%, 62.1% in patients with a %RVT > 30% to ≤ 60%, and 40.2% among those with a %RVT > 60%.
Similar findings were seen in the placebo group, at 24-month EFS rates of 83.3%, 62.5%, 46.2%, and 34.8%, respectively, and when examining EFS at 36 months.
Jones noted that while RVT was indicative of EFS, "further studies are needed to validate %RVT for evaluation of outcomes in this patient population."
Tina Cascone, MD, PhD, Department of Thoracic/Head and Neck Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, commented that there is an "unmet need" in early-stage NSCLC.
She noted that 3-year EFS/disease-free rates with neoadjuvant, perioperative, and adjuvant immunotherapy/chemotherapy approaches in stage II-III NSCLC are approximately 53%-65%.
"More effective personalized treatments and biomarkers are needed to improve upon these benchmarks," emphasized Cascone, who was not involved in the study.
Cascone highlighted that the current results show the EFS benefit of perioperative pembrolizumab extends to patients with RVT after the neoadjuvant phase, but only up to a cutoff of 60%RVT.
She pointed to data from CheckMate 816, which also looked at the relationship between pathologic regression and EFS with neoadjuvant nivolumab (Opdivo, Bristol Myers Squibb) but did not include lymph nodes in the analysis.
It is not clear, Cascone said, whether pathologic regression following neoadjuvant immunotherapy in lymph nodes may differ from that in the tumor and have an impact on the survival benefit.
A further question, she noted, remains as to the contribution of the adjuvant phase of treatment to the EFS outcome and whether that relates to %RVT.