Abstract
BACKGROUND: First-line immunotherapy combined with chemotherapy for extensive-stage small-cell lung cancer (ES-SCLC) has been consistently recommended by clinical guidelines, but the improvement in overall survival remains limited. There is an urgent need to identify reliable predictive biomarkers for immunotherapy to select patients who would benefit most. Serum ferritin (SF) is a key regulator in ferroptosis and plays a significant role in immunotherapy of lung cancer. Therefore, we hypothesized that the change rate of serum ferritin (ΔSF) during immunotherapy, combined with inflammation-related indicators, could serve as a useful predictive marker for treatment response in ES-SCLC patients. METHODS: We comprehensively reviewed the medical records of 550 ES-SCLC patients, divided into an experimental group (425 patients receiving immune checkpoint inhibitors (ICIs) plus chemotherapy) and a control group (125 patients receiving chemotherapy alone). The study analyzed the correlation between pre-immunotherapy SF levels and molecular subtypes, clinical stage, tumor location, and programmed death-ligand 1 (PD-L1) expression in ES-SCLC patients; the correlation of SF levels and ΔSF with objective response rate (ORR); the correlation of ΔSF combined with a multidimensional inflammation model-including neutrophil-to-lymphocyte ratio (NLR), lactate dehydrogenase (LDH), and C-reactive protein (CRP)-with ORR; and survival analysis for these parameters. RESULTS: Patients with lower SF levels before immunotherapy had a higher ORR (χ² = 4.837, P = 0.035) and longer progression-free survival (PFS) (median 6.9 vs. 4.1 months). Patients with a high ΔSF during immunotherapy showed a higher ORR (χ² = 6.475, P = 0.019). Patients with high ΔSF combined with low NLR and LDH levels before immunotherapy were more likely to achieve a higher ORR (P < 0.001). After integration, patients with low SF levels and high ΔSF before immunotherapy had the best PFS, whereas those with high SF levels and low ΔSF before immunotherapy had the worst PFS (median 8.9 vs. 4.5 months). Within the high ΔSF group, patients with lower NLR had longer PFS than those with higher NLR (median 9.8 vs. 5.2 months); similarly, patients with lower LDH levels had longer PFS than those with higher LDH levels (median 9.2 vs. 5.6 months). Multivariate analysis identified SF levels before immunotherapy (HR = 1.58, P = 0.026) and ΔSF during immunotherapy (HR = 0.52, P = 0.002) as independent prognostic factors. SCLC clinical stage (HR = 0.56, P = 0.037) and molecular subtype (SCLC-A: HR = 1.67, P = 0.003; SCLC-N: HR = 1.51, P = 0.012; SCLC-P: HR = 0.73, P = 0.004; SCLC-Y: HR = 0.64, P = 0.003) were also independent prognostic factors. However, NLR and LDH levels alone were not independent prognostic factors and required combined assessment with ΔSF. CONCLUSION: Our study suggests that the serum ferritin change rate combined with the NLR and LDH inflammation model can serve as a biomarker for predicting the efficacy and survival outcomes of immunotherapy in ES-SCLC.