Neoadjuvant nivolumab with or without relatlimab in resectable non-small-cell lung cancer: a randomized phase 2 trial

Lung cancer remains a leading cause of cancer death globally. While advancements in precision therapies targeting specific oncogenic mutations and immune checkpoint inhibitors (ICIs) have improved outcomes in metastatic disease, these advancements are being translated to localized disease treated with curative surgery. Preoperative or perioperative systemic therapies offer advantages, including avoiding delays due to postoperative morbidity and enabling response monitoring via imaging and histopathology. The reinvigoration of a suppressed immune response may be more effective in the native tumor context. Studies have explored preoperative ICI therapy targeting PD-1, PD-L1, CTLA-4, and other targets in resectable non-small-cell lung cancer (NSCLC). More recently, preoperative PD-1/PD-L1 antibodies combined with platinum-based chemotherapy have been explored; however, this approach may obscure the ICI's contribution and exposes patients to unnecessary chemotherapy toxicity. Dual ICI blockade, such as targeting LAG-3 and PD-1, is a rational approach given their distinct and potentially synergistic mechanisms of action. Preclinical models demonstrate synergistic enhancement of T cell function and anti-tumor activity with dual blockade. A phase 3 study in melanoma showed that the combination of nivolumab and relatlimab was superior to nivolumab alone. This study aimed to explore the feasibility and safety of preoperative dual targeting of PD-1 and LAG-3 in patients with resectable NSCLC, assessing pathological and radiographic responses, survival endpoints, and quality of surgical resections. Exploratory analyses investigated specific biological processes associated with response or resistance.

Numerous studies have investigated the use of preoperative immunotherapy in resectable NSCLC, utilizing various agents and combinations, including PD-1/PD-L1 inhibitors alone or with chemotherapy. While these studies demonstrated varying degrees of success in terms of pathological response rates and improved survival, there remained a need to identify more effective and less toxic regimens. The combination of PD-1 and LAG-3 blockade has shown promise in preclinical models and in melanoma, making it a compelling strategy for NSCLC as well. The existing data on preoperative combined ICI therapies, particularly those involving PD-1/PD-L1 axis plus another checkpoint, suggested a potential benefit but also highlighted the need for further investigation to optimize treatment strategies and reduce toxicity. This necessitates investigation into less toxic, potentially synergistic combinations. The combination of nivolumab and relatlimab was chosen based on the promising results from melanoma trials.

This open-label, randomized phase 2 trial (NEOpredict-Lung, NCT04205552) enrolled 60 treatment-naive patients with resectable NSCLC (stages IB, II, or selected IIIA). Patients were randomized (1:1) to receive two preoperative doses of nivolumab (240 mg every 14 days) or nivolumab plus relatlimab (240 mg and 80 mg, respectively, every 14 days). The primary endpoint was the feasibility of surgery within 43 days. Secondary endpoints included pathological and radiographic response rates, pathologically complete resection rates, disease-free survival (DFS), overall survival (OS), and safety. Exploratory analyses included metabolic response assessment (PERCIST), immune cell phenotyping in peripheral blood and resected tumors by flow cytometry, and gene expression profiling using NanoString technology. Whole-exome sequencing was performed on pretreatment and post-treatment tumor samples to analyze mutational spectra and subclonal dynamics. Statistical analyses included descriptive statistics for secondary endpoints and exploratory analyses to assess correlations between translational endpoints and patient outcomes.

The primary endpoint was met by all 60 patients, demonstrating the feasibility of both treatment arms. Curative resection was achieved in 95% of patients. Major pathological response (MPR, ≤10% viable tumor cells) rates were 27% (nivolumab) and 30% (nivolumab plus relatlimab). Objective radiographic response rates were 10% (nivolumab) and 27% (nivolumab plus relatlimab). At 12 months, DFS rates were 89% (nivolumab) and 93% (nivolumab plus relatlimab), and OS rates were 93% (nivolumab) and 100% (nivolumab plus relatlimab). Both treatments were generally well-tolerated, with grade ≥3 treatment-emergent adverse events reported in 10% and 13% of patients, respectively. Exploratory analyses showed an increase in CD8+ GrzB+ effector T cells in peripheral blood of responders, and fewer suppressive immune cells in tumors with MPR. Gene expression analysis revealed distinct patterns of immune-related gene modulation in response to the two treatment arms. Whole-exome sequencing demonstrated genomic remodeling in immunotherapy responders, with enrichment and depletion of subclones. In some cases, enrichment of cancer gene mutations was observed in residual tumor cells.

This study demonstrates the feasibility and safety of preoperative nivolumab with or without relatlimab in patients with resectable NSCLC. The higher response rates and survival observed in the combination arm suggest a potential benefit of dual PD-1/LAG-3 blockade. The exploratory analyses provide valuable insights into the biological mechanisms underlying response and resistance to immunotherapy. The observed genomic remodeling in responders highlights the dynamic interplay between immunotherapy and tumor evolution. Limitations include the moderate sample size, which precludes definitive conclusions regarding clinical efficacy, and the lack of stratification for PD-L1 status, which may have influenced the observed response rates. The excellent surgical outcomes may also be attributed to the exclusion of patients with extensive mediastinal lymph node metastases.

This phase 2 study establishes the feasibility and safety of preoperative nivolumab with or without relatlimab in resectable NSCLC. The early signals of enhanced clinical activity and modulation of the tumor microenvironment warrant further investigation of dual PD-1/LAG-3 blockade in larger, controlled trials to confirm the benefits and optimize treatment strategies.

The moderate sample size limits the power to definitively conclude on the superiority of the combination therapy over nivolumab monotherapy. The lack of stratification for PD-L1 status might have introduced bias into the results. The exclusion of patients with extensive mediastinal lymph node involvement might limit the generalizability of the findings to all patients with resectable NSCLC.