Eosinophilia Linked With Immune-Related Adverse Events in NSCLC Treatment

An increased rate of immune-related adverse events (irAEs) was found in patients with non–small cell lung cancer (NSCLC) who had pretreatment eosinophilia and were treated using immune checkpoint inhibitors (ICIs), according to a recent review.¹ Poorer survival outcomes and increased susceptibility for irAEs could be identified using eosinophilia as a biomarker.

NSCLC is the most common form of lung cancer; it is seen in 87% of all those diagnosed with lung cancer.² Even though lung cancer is one of the most commonly diagnosed cancers in the country, survival rates are poor for late-stage and metastatic NSCLC. ICIs have become a method of treating NSCLC that improves overall survival (OS) but often comes with irAEs. Studies in the past have evaluated the relationship between eosinophilia and clinical outcomes in those treated with ICIs. This review aimed to collect data from these previous studies to assess the association.

The researchers used PubMed, ScienceDirect, PMC, Web of Science, Google Scholar, and Cochrane Library to look for studies for the review. Included studies needed to have adult patients 18 years and older and to report pretreatment eosinophil levels. Studies that did not have sufficient data, did not separate NSCLC from other types of cancer, or only had measurements of eosinophilia that came after starting ICIs were excluded from the review. The researchers also excluded reviews, conference abstracts, animal studies, commentaries, articles not published in English, and case reports. Sociodemographic data were extracted by the researchers along with the country of the study, study design, total sample size, and outcomes.

There were 11 studies included in the review that reported data on 14,095 individuals, with studies coming from North America, Europe, and Asia. All studies were retrospective cohort studies. None of the studies had a high risk of bias, although 6 had an unclear risk.

The pooled unadjusted analysis found that there was not a significant association between overall survival and pretreatment eosinophilia (OR, 0.79; 95% CI, 0.42-1.51), which held true after adjusting (OR, 0.74; 95% CI, 0.53-1.03). The unadjusted pooled OR for progression-free survival (PFS) showed a trend towards reduced PFS in those with higher eosinophil levels (OR, 0.78; 95% CI, 0.54-1.13), but this was not significant. After adjusting, this association became significant (OR, 0.68; 95% CI, 0.58-0.80), with eosinophilia linked to lower odds of longer PFS.

An increased risk of immune-related toxicity was found in those with pretreatment eosinophilia in both the unadjusted (OR, 3.19; 95% CI, 2.11-4.83) and adjusted (OR, 3.35; 95% CI, 2.25-5.02) analyses.

There were some limitations to this analysis. All studies were observational, which could have led to selection bias. Causal inference could not be drawn due to all studies being retrospective. There were differences between studies in how eosinophilia was defined, which could have led to heterogeneity in the results. There was inconsistent reporting of follow-up duration. Survival dynamics would likely be better represented with HR rather than OR measures. Subtype-specific associations between irAEs would have been missed due to looking at irAEs collectively. Some studies did not adjust for confounders. The subtype of NSCLC was also not considered.

“The findings of this systematic review and meta-analysis support the growing recognition of pretreatment eosinophilia as a potentially valuable prognostic biomarker in patients with NSCLC being considered for ICI therapy,” the authors concluded.

References

1. Ganatra N, Thompson J, Desai R, et al. Pretreatment eosinophilia as a biomarker for adverse outcomes in non–small cell lung cancer patients receiving immune checkpoint inhibitors: a systematic review and meta-analysis. Semin Oncol. 2025;52(6):152431. doi:10.1016/j.seminoncol.2025.152431

2. Key statistics for lung cancer. American Cancer Society. Updated January 16, 2025. Accessed November 5, 2025. https://www.cancer.org/cancer/types/lung-cancer/about/key-statistics.html