MA22.09 Should Stereotactic Radiosurgery be Considered for Salvage of Intracranial Recurrence in Small Cell Lung Cancer

Prophylactic cranial irradiation (PCI) remains a standard of care for small cell lung cancer (SCLC) to improve overall survival (OS) and prevent recurrence in limited (LS) and extensive stage (ES) disease. Intracranial recurrence (IC) after PCI affects 12-33%. Limited published data describe outcomes of salvage reirradiation (ReRT). Our purpose was to review outcomes after IC post-PCI or therapeutic whole brain radiotherapy (WBRT). Consecutive patients with pathologically-confirmed SCLC assessed (01/2013-12/2015) at a tertiary cancer centre (catchment 1.5M) were retrospectively reviewed. Demographics, treatment and outcomes were abstracted and summary statistics calculated. Kaplan-Meier estimates and univariate and multivariate analysis (MVA) via the Cox proportional hazard model were performed. Median age was 66.1yrs, 53% female, and 80% ECOG 0-2 (N =372). Median survival (MS) was 24 months (95%CI 18.3-29.7 mos) for 103 LS, and 7 months (95%CI 6.1-7.9 mos) for 269 ES patients. 72/103 LS received PCI (69.9%), 84.7% of whom had radical thoracic radiotherapy (RT). 54/269 ES patients presented with brain metastases (BM); 98/215 of the remaining received PCI, and 72 of those thoracic RT (84.7% 25-30Gy/10). PCI dose was 25Gy/10 in 95.9%. PCI patients had better performance status (PS), and were more likely to receive chemotherapy (CT) and thoracic RT (all p<0.013). 18.8% (32/170) recurred post-PCI (13 LS; 19 ES) at a median of 11.5 mos. 45/54 presenting with BM received WBRT (83.3% 20Gy/5), 14 of whom recurred. MS after PCI was 28 mos vs 12 mos for LS and ES, respectively. For LS patients with IC post-PCI, MS was 20 mos vs 38 mos without IC (p=0.03). On MVA, interval between brain RT predicted OS after PCI (HR 0.87; p<0.001), while stage (HR 3.56; p=0.008) and cranial RT dose predicted IC (HR 0.65; p=0.047). At IC, 56.5% (26/46) had <5 BM, median 1.7cm (range 0.5-5cm), 39.1% had no extracranial disease, 6 were asymptomatic, and 50% had ECOG 0-2. 30/46 had ReRT: 27 WBRT and 3 stereotactic radiosurgery (SRS). In retrospect, 17/46 would have been candidates for salvage SRS: 5 LS post-PCI; 8 ES post-PCI; and 4 ES post-WBRT. This cohort seems to challenge the belief that in-brain progression is always: diffuse; associated with clinical deterioration; and synchronous with systemic failure. With potential for OS >6 months, repeat WBRT risks meaningful neurocognitive toxicity. Further data are required; however, approximately 1 in 3 SCLC patients who recur after PCI or WBRT appear clinically appropriate for salvage SRS.