Prophylactic cranial irradiation in small cell lung cancer patients in the era of magnetic resonance imaging: A systematic review and meta-analysis.

Abstract

e20636 Background: Patients with small cell lung cancer (SCLC) are at a particularly high risk for the development of brain metastases. Current guidelines recommend prophylactic cranial irradiation (PCI) for these patients to prevent brain metastases and prolong overall survival (OS), although the clinical evidence for recommendations stems from an era without routine use of brain imaging. We performed a systematic review and meta-analysis of all available studies reporting on PCI for SCLC patients with minimal inclusion restrictions to assesses the efficacy of PCI in patients with limited disease (LD) and extensive disease (ED) SCLC and focus on the impact of routine brain imaging. Methods: A literature search for studies reporting on PCI in SCLC patients performed in EMBASE, MEDLINE, CENTRAL, and grey literature sources identified 4,783 studies. Random-effects meta-analyses pooled hazard ratios (HR) for overall survival (OS) between PCI and no intervention groups. Results: 208 individual studies reported on outcomes related to PCI with 175 studies assessing survival differences associated with PCI, and 87 of these being amenable to meta-analysis. The majority (n = 79) were retrospective cohort studies. PCI was associated with improved OS in all SCLC patients (HR 0.60; 95% CI 0.55-0.65; n = 87 studies; n = 50,339 patients; I2= 77%), as well as in cohorts exclusively consisting of LD patients (HR 0.60; 95% CI 0.54-0.66; n = 58 studies; n = 22,417 patients; I2= 76%) or ED patients (HR 0.63; 95% CI 0.53-0.75; n = 23 studies; n = 24,997; I2= 75%). Only eight studies confirmed absence of IMD at restaging using magnetic resonance imaging in all patients and found OS no longer to be in favour of PCI (HR 0.75; 95% CI 0.50-1.12; n = 1,318 patients; I2= 77%). Pooled OS from randomized controlled trials was not associated with PCI (HR 0.88; 95% CI 0.65-1.12; n = 5 studies; n = 925 patients; I2= 62.5%). Among studies not amenable to meta-analysis for OS, 4 radiographically confirmed absence of IMD, but none found OS to be in favour of PCI. Conclusions: Administration of PCI is associated with a significant OS benefit, but not when considering studies that radiographically confirmed absence of IMD at restaging or randomized controlled trials, suggesting that patient selection may be contributing to this observation.