Meta-Analysis Finds No Significant NSCLC Survival Differences With ICIs, but Trial Diversity Remains Limited

More representative trial populations and standardized subgroup reporting are needed in non–small cell lung cancer (NSCLC) research, according to a meta-analysis published in Cancers, which found no statistically significant differences in overall survival (OS) benefit from immune checkpoint inhibitors (ICIs) by sex, age, or race.¹

Uneven Representation Limits Insights Into ICI Efficacy

The researchers explained that ICIs, which block the PD-1 and PD-L1 pathways, have transformed NSCLC treatment and improved survival in advanced disease. However, prior studies have reported inconsistent findings regarding both trial enrollment disparities and treatment efficacy by age, sex, and race among patients with NSCLC treated with ICI therapy, as gaps in representation and outcome reporting persist.²

To address these limitations, the researchers conducted a comprehensive meta-analysis of relevant phase 3 clinical trials to systematically assess disparities in enrollment and subgroup reporting, as well as to clarify whether ICI efficacy differs by age, sex, or race in patients with NSCLC.¹

“These findings can inform clinical decision-making and guide future trial design toward more equitable cancer care,” the authors wrote.

Specifically, they performed a systematic review of completed prospective phase 3 trials that were initiated after 2015 and evaluated ICIs in patients with metastatic NSCLC (mNSCLC). Eligible trials were identified in September 2025 from PubMed and ClinicalTrials.gov using relevant keywords.

The researchers used random-effects meta-analyses to pool ratios of HRs (RHRs) for overall survival, evaluating treatment-by-subgroup interactions by sex (men vs women), race (White vs Asian), and age (< 65 vs ≥ 65 years). In addition, they conducted random-effects meta-analyses of ORs for death within each subgroup in the investigational treatment arm and across the overall study population.

No Significant OS Differences Found Across Sex, Race, and Age

The researchers identified 21 eligible trials comprising 10,950 patients. Two trials used multi-arm designs, resulting in 23 pairwise comparisons between investigational and control arms.

All trials reported sex distribution and OS subgroup analyses by sex, with men accounting for 68.5% of patients. Among the 23 comparisons, 11 (47.8%) showed a significant OS benefit for men, whereas 1 (4.5%) showed a significant OS benefit for women.

In a random-effects meta-analysis of the 20 comparisons reporting OS HRs by sex, women experienced a non–statistically significant smaller OS benefit from ICIs compared with men (RHR, 0.91; 95% CI, 0.80-1.04; P = .17). Also, a random-effects meta-analysis of the 9 studies reporting deaths by sex in the investigational arms found no significant difference in the odds of death for men (OR, 0.98; 95% CI, 0.73-1.32; P = .89).

Regarding race, 61.9% (n = 13) of trials reported distribution, and 33.3% (n = 7) reported OS subgroup analyses. Across these studies, 59.2% of patients were White, 34.8% were Asian, 2.7% were Native American or Alaskan Native, and 1.5% were Black; no trials reported outcomes for Black patients due to small sample sizes. Of the 8 investigational-control arm comparisons, 2 (25%) found a significant OS benefit from ICIs for White patients, while 1 of 6 comparisons (16.7%) found a significant survival benefit for Asian patients.

A random-effects meta-analysis of 5 comparisons reporting OS HRs for White and Asian patients showed no significant difference in OS benefit (RHR, 0.95; 95% CI, 0.72-1.24; P = .72). However, random-effects meta-analyses of 4 comparisons reporting deaths by race in the investigational arm found higher odds of death for White patients vs Asian patients (OR, 1.76; 95% CI, 1.00-1.39; P = .0496). Across both treatment arms, White patients also had higher odds of death than Asian patients (OR, 2.35; 95% CI, 1.42-3.88; P < .001).

Lastly, 85.7% (n = 18) of trials reported age distribution, with 55.9% of patients younger than 65 years. Of the investigational-control arm comparisons, 20 reported OS subgroup analyses among patients younger than 65, while 17 reported deaths among those aged 65 and older. Among the 20 comparisons, 7 (35%) found a significant OS benefit with ICIs for patients younger than 65 years, compared with 4 of 17 (23.5%) for those aged 65 and older.

A random-effects meta-analysis of 15 comparisons reporting OS HRs by age suggested a non–statistically significant smaller OS benefit with ICIs for patients aged 65 and older compared with younger patients (RHR, 0.92; 95% CI, 0.81-1.04; P = .19).

Improving Trial Design to Ensure Equitable ICI Evaluation

The researchers acknowledged several limitations, including the continued lack of trial population diversity. Their work was also limited by inconsistent data reporting across subgroups, as well as the lack of time-to-event survival data. Despite these limitations, the researchers concluded by expressing confidence in their findings.

“Our results highlight the need for more representative trial populations and standardized reporting of subgroup analysis to ensure equitable benefit and evaluation of ICIs in mNSCLC treatment,” the authors wrote.

References

1. Yaskolko M, Liu C, Barsouk A, Sussman JH, Barsouk AA. Disparities in non-small cell lung cancer (NSCLC) by age, sex, and race: a systematic review and meta-analysis of immune checkpoint inhibitor (ICI) trials. Cancers (Basel). 2025;18(1):128. doi:10.3390/cancers18010128
2. Girigoswami A, Girigoswami K. Potential applications of nanoparticles in improving the outcome of lung cancer treatment. Genes (Basel). 2023;14(7):1370. doi:10.3390/genes14071370