Impact of previous corticosteroid exposure on outcomes of patients receiving immune checkpoint inhibitors for advanced non-small cell lung cancer: a retrospective observational study

Background Immunosuppressive dosing of corticosteroids at the start of treatment impairs immune checkpoint inhibitor (ICI) efficacy in advanced non-small cell lung cancer (NSCLC). Previous cumulative dose and intake modality of corticosteroids may result in different levels of immunosuppression. We sought to determine whether these aspects could predict disease control and survival in NSCLC patients receiving ICIs. Methods We conducted a retrospective single-center study, including patients treated with ICI as second-line therapy at our institution between July 2015 and February 2021. A prednisone equivalent threshold of 700 mg defined low (LCD) or high (HCD) cumulative dosing during the 90 days before starting ICIs. At least one 5-day course of ≥ 10 mg prednisone equivalent determined whether the intake was pulse (PCD, ≤ 5 days) or continuous (CCD, > 5 days). Results We included 113 consecutive patients. Durable control benefit and no clinical benefit (NCB) were reported in 53 (47%) and 60 (53%) of the cases, respectively. ECOG PS 1–2, negative PD-L1 expression, HCD, and CCD were significantly related to NCB in multivariate analysis. The median PFS was 4.6 months (95%CI: 3.9–6.3) and median OS was 6.9 months (95% CI: 6.0–8.9) after a median follow-up time of 43.5 months (95% CI: 32.6–54.4). Multivariate analysis of survival confirmed ECOG PS 1–2 ( p = 0.005), negative PD-L1 expression ( p = 0.002), and CCD ( p = 0.001) significantly correlated with an increased risk of mortality. Conclusions These findings imply that immunosuppressive corticosteroid dosing before ICIs, even of short duration, might affect survival. The constraints of this study and lack of reliable comparisons suggest a hypothesis-generating value of these results.