Early and High‐Dose Glucocorticoid Use and Survival in Advanced NSCLC Receiving Immune Checkpoint Inhibitors: A Multicenter Target Trial Emulation

Despite the immunomodulatory mechanism of immune checkpoint inhibitors (ICIs), systemic glucocorticoids are commonly used in advanced non–small cell lung cancer (NSCLC). Evidence regarding their effects on ICI outcomes is inconsistent, in part due to confounding by indication and immortal time bias. We conducted a multicenter target trial emulation to assess the association of glucocorticoid timing and dose with survival outcomes while addressing these biases. Patients with advanced NSCLC who initiated ICI therapy between 2015 and 2021 at four Korean tertiary hospitals were included. The primary endpoint was 2‐year overall survival (OS), with disease progression as a secondary endpoint. To minimize biases, outcomes were compared between concomitant glucocorticoid users and non‐users using time‐dependent propensity score matching. Among 1,815 eligible patients, 1,128 were matched. Hazard ratios (HRs) from the four institutions were pooled using a random‐effects meta‐analysis. Concomitant glucocorticoid use was associated with an increased risk of mortality (pooled aHR 1.36; 95% CI, 1.06–1.75). Early exposure within 90 days of ICI initiation was associated with poorer OS (aHR 1.49; 95% CI, 1.18–1.87), whereas later initiation showed no association. Dose‐stratified analyses showed that high‐dose exposure (> 1 mg/kg/day) was associated with reduced survival (aHR 1.61; 95% CI, 1.31–1.99), whereas low‐dose exposure was not associated with adverse outcomes. After accounting for confounding by indication and time‐related biases, concomitant systemic glucocorticoid use during ICI therapy was associated with less favorable survival outcomes. This association was more pronounced with early and high‐dose exposure, underscoring the clinical importance of glucocorticoid timing and dose.