Hippocampus avoidance versus standard prophylactic cranial irradiation for small cell lung cancer: A systematic review and meta-analysis of randomized controlled trials.

Abstract

e20151 Background: Prophylactic cranial irradiation (PCI) for limited stage (LS-SCLC) or extensive stage small cell lung cancer (ES-SCLC) reduces incidence of brain metastases, but causes neurocognitive function (NCF) decline. Hippocampus is critical for updating cognitive representations to reflect changing social and environmental context. We aim to compare the effects of hippocampus avoidance PCI (HA-PCI) versus standard PCI on neurocognition, incidence of brain metastasis, overall survival and toxicity, in patients with small cell lung cancer. Methods: We searched PubMed, Embase, and Cochrane database for randomized control trials (RCTs) comparing HA-PCI, defined as 5mm Hippocampal avoidance zone (HAZ) versus standard PCI in patients with small cell lung cancer without prior brain metastasis. The primary outcome was decline in NCF, measured by the Hopkins Verbal Learning Test- Revised (HVLT-R) and Free and Cued Selective Reminding Test (FCSRT). The secondary outcomes include incidence of brain metastasis, overall survival and toxicity. Statistical analysis was performed using Review Manager (version 5.4) (Cochrane Collaboration). Heterogeneity was assessed with I 2 statistics. The protocol is registered in PROSPERO (CRD420251250284). Results: We included 711 patients from three RCTs, including 355 (49.9%) patients of HA-PCI group. Patients received radiation of 25 Gy in 10 fractions, with mean dose of ≤ 8.5 Gy to the HAZ across trials. In LS-SCLC (n = 500), 245 received HA-PCI and 255 received PCI; in ES-SCLC (n = 211), 111 received HA-PCI and 100 received PCI. Pooled analysis of binary outcomes showed there is a non-significant trend toward reduced risk of NCF decline in HA-PCI compared to standard PCI. This was observed across both verbal learning (total recall) (RR 0.89; 95%CI 0.56 - 1.43; p = 0.64; I 2 = 58%) and delayed memory (delayed recall) (RR 0.62; 95%CI 0.22 - 1.75; p = 0.36; I 2 = 82%). Subgroup analysis between two different tests measuring verbal learning showed significant difference (p = 0.03). The cumulative incidence of brain metastasis at two years was not significantly reduced in HA-PCI group (RR 0.97; 95%CI 0.67 - 1.42; p = 0.89; I 2 = 0%). No significant effect on overall survival (HR 1.02; 95% CI 0.83 - 1.25; p = 0.87; I 2 = 0%). Subgroup analysis for stage (LS-SCLC vs. ES-SCLC) showed no significant difference (p = 0.20). Toxicity analysis showed HA-PCI significantly had lower incidence of skin and subcutaneous tissue disorders compared to standard PCI (RR 0.68; 95%CI 0.49 - 0.95; p = 0.02; I 2 = 0%). Conclusions: In conclusion, hippocampus avoidance PCI demonstrated comparable oncologic outcomes to standard prophylactic cranial irradiation, with a non-significant trend toward HA-PCI, suggesting it may reduce neurocognitive failure across different assessments, in patients with SCLC.