Prospective observational study of nutritional and immunological indices as predictive biomarkers of response to immune checkpoint inhibitors in non-small cell lung cancer (ICI-PREDICT study).

Abstract

e20524 Background: Nutritional and immunological indices, including controlling nutritional status (CONUT) score, skeletal muscle area (SMA), and neutrophil-to-lymphocyte ratio (NLR) have been reported as predictors of response to immune checkpoint inhibitor (ICI) in non-small cell lung cancer (NSCLC). The aim of this prospective, multicenter study is to evaluate the clinical impact of pre- or early post-treatment changes in nutritional and immunological indices on treatment outcomes in patients with advanced NSCLC treated with first-line ICIs. Methods: We enrolled 301 advanced NSCLC patients treated with ICIs (monotherapy or combined with chemotherapy) as first-line therapy, without any oncogene mutations, and available data on programmed cell death-ligand 1 (PD-L1) expression (150 in training and 151 in validation cohorts) from March 2020 to November 2022. The CONUT score, SMA at the L3 level (cm2/m2), and NLR, were evaluated before treatment initiation and at the first response evaluation. Cut-off values were determined by receiver operating characteristic curves with objective response as the outcome in the training cohort. The primary endpoint was progression-free survival (PFS). For exploratory study, serum before treatment was collected to measure 15 muscle-derived cytokines (Apelin, BDNF, CX3CL1, EPO, FABP3, FSTL1, FGF21, IL-6, IL-15, Irisin, LIF, Musclin, Myostatin, OSM, and SPARC) (n = 123). Results: The median follow-up period and PFS was 17.5 (range: 7.3-39.4) and 6.9 (range: 0.1-31.5) months, respectively. In the training cohort, men (hazard ratio [HR]: 0.52, 95% confidence interval [CI]: 0.30-0.90, p = 0.0183), ΔCONUT < 0 (HR: 0.59, 95% CI: 0.35-0.98, p = 0.0431), and ΔSMA > −0.13 (HR: 0.56, 95% CI: 0.33-0.96, p = 0.0333) were independent predictors for PFS. In the validation cohort, PD-L1 ≥ 50% (HR: 0.54, 95% CI: 0.34-0.84, p = 0.0061), NLR < 3.65 (HR: 0.47, 95% CI: 0.30-0.74, p = 0.0011), and ΔSMA > −0.13 (HR: 0.61, 95% CI: 0.38-0.98, p = 0.0421) were independent predictors for PFS. In the exploratory study, multivariate analysis including muscle-derived cytokines (continuous variables) revealed that performance status = 0 (HR: 0.24, 95% CI: 0.12-0.48, p < 0.0001), without baseline prednisone (≥ 10 mg/day) (HR: 0.17, 95% CI: 0.05-0.58, p = 0.0213), smoking history (HR: 0.34, 95% CI: 0.14-0.81, p = 0.0269), and FSTL1 (HR: 0.97, 95% CI: 0.93-0.99, p = 0.0185) were independent predictors of PFS. Conclusions: ΔSMA was the only factor validated as an independent predictor of PFS. The underlying biological mechanisms may be related to skeletal muscle-derived FSTL1. Clinical trial information: UMIN000040450.