Prophylactic cranial irradiation (PCI) versus active surveillance in patients with limited-stage small cell lung cancer: a retrospective, multicentre study

Background: The recommendation of PCI for limited-stage small cell lung cancer (LS-SCLC) is primarily based on evidence from the pre-magnetic resonance imaging (MRI) era. However, as MRI accuracy improves and stereotactic radiosurgery advances, the role of PCI for LS-SCLC has become uncertain. This study aims to compare the contemporary survival outcomes of patients with LS-SCLC treated with PCI versus active surveillance.

Methods: We conducted a retrospective cohort study in which 1068 patients with LS-SCLC who achieved a good response to first-line chemoradiotherapy were consecutively enrolled from 5 tertiary medical centres between June 2009 and June 2019. Of these patients, 440 received PCI, while 628 received surveillance without PCI. Propensity score matching with a 1:1 ratio was performed to balance the baseline characteristics of the two cohorts. The endpoints were overall survival (OS) and the incidence of brain metastasis (BM).

Results: In total, 648 patients were matched. The baseline characteristics were generally well balanced. At a median follow-up of 64.5 months (range 2-190), patients who underwent PCI had a significantly lower risk for BM than those who underwent surveillance. The 3-year cumulative incidence rate of BM was 28.2% (95% CI 22.5-33.8%) in the PCI cohort and 38.5% (32.6-44.5%) in the surveillance cohort (Gray's p = 0.002). However, the lower incidence of BM in the PCI cohort did not translate into a significant extension of OS. The median OS was 35.8 months (95% CI 27.6-44.0 months) in the PCI cohort versus 32 months (26.4-37.6 months) in the surveillance cohort (HR 0.90, 95% CI 0.74-1.10, p = 0.29). Multivariable analysis showed that disease stage, chemoradiotherapy sequence, and response to chemoradiotherapy were independent prognostic factors for BM or OS.

Conclusions: Overall, PCI reduces the risk for BM but does not substantially prolong OS compared with active surveillance. A phase 3, prospective clinical trial (NCT04829708) we initiated is currently underway, which is expected to corroborate our results.