Limited-Stage Small Cell Lung Cancer: Is Prophylactic Cranial Irradiation Necessary?

Purpose: Prophylactic cranial irradiation (PCI) reduces the incidence of brain metastases in patients with limited stage small cell lung cancer (LS-SCLC). However, PCI is associated with neurotoxicity. Previous studies have not consistently used pretreatment magnetic resonance imaging. Modern imaging improvements continue to enhance early metastasis detection, potentially decreasing the utility of PCI. We sought to determine whether PCI was associated with improved outcomes in LS-SCLC patients with modern imaging.

Methods and materials: We identified LS-SCLC patients with no intracranial disease who were treated between 2007 and 2018. Kaplan-Meier estimates of overall survival (OS) and progression-free survival (PFS) were calculated and multivariate Cox proportional hazards models were generated. The cumulative incidence of brain metastases was estimated using competing risks methodology.

Results: Ninety-two patients were identified without intracranial disease at initial staging, 39 of whom received PCI. Median follow-up was 56.7 months. The median OS for the cohort was 35.5 months (95% CI, 25.8-49.3), and median PFS was 19.1 months (95% CI, 12.3-30.5). Median OS with PCI versus observation was 37.9 months (95% CI, 31.8-not reached) versus 30.5 months (95% CI, 14.6-56.1; P = .07), whereas median PFS was 26.3 months (95% CI 19.1-not reached) versus 12.3 months (95% CI, 8.5-30.5; P = .02), respectively. Overall, at 2 years, the cumulative incidence of brain metastases was 10% with PCI and 29% without; this increased to 32% and 29% by 4 years (P = .66). In those patients who had negative magnetic resonance imaging of the brain after completing initial treatment, the 1-year cumulative incidence of brain metastasis was not significantly different at 8% versus 11% (P = .46) respectively. Both PCI and treatment response were independent predictors for PFS on multivariate analysis. Stratified by disease response, patients with a complete response did not benefit from PCI (P = .50), whereas those with partial response or stable disease experienced improved PFS (P = .01).

Conclusions: Overall, PCI was associated with improved PFS and reduced early incidence of brain metastases. Patients achieving a complete response to initial therapy did not experience a PFS benefit with PCI. This may indicate that subsets of LS-SCLC patients can potentially be spared from PCI in the era of modern imaging.