Adverse Outcomes After Palliative Radiation Therapy for Uncomplicated Spine Metastases: Role of Spinal Instability and Single-Fraction Radiation Therapy

Abstract

Radiation therapy (RT) is frequently employed to palliate spine metastases. There are multiple randomized trials of single fraction (SF) versus multiple fraction (MF) palliative RT with equivalent pain outcomes in uncomplicated bone (including spine) metastases. However, there are little data to inform incidence and predictors of adverse outcomes after RT for uncomplicated spine metastases, including the possible role of spine instability and RT fractionation. In total 337 spine metastases in 258 patients treated with RT at a single institution from 2008 to 2013 were retrospectively reviewed. Previously irradiated sites, complicated spine lesions due to spinal cord compression (SCC) or requiring surgery were excluded. The primary outcome was the time to first adverse event (AE) at the treated site, including: symptomatic vertebral body (VB) collapse, hospitalization for site-related pain, salvage spine surgery, vertebroplasty or other interventional procedure for pain, and SCC. A competing risks survival model was used to simultaneously assess potential predictors, including demographic, disease [e.g., Spinal Instability Neoplastic Score, SINS (0-18)], and treatment (e.g., dose fractionation) factors. Death was the competing risk. Clustering was adjusted for with a random-effects model. Median survival was 160 days; 260 days for those receiving post-RT systemic therapies, 40 days for those who did not; 30Gy/10 was given to 49% of sites, 20Gy/5 to 22%, and 8Gy SF to 9%. Median SINS was 7 (IQR 5 to 9). AEs developed in 46 sites (14%), including 12 symptomatic VB collapses, 32 hospitalizations, 4 salvage spinal surgeries, 14 interventional procedures, and 16 SCCs. Median time to first AE - was 121 days (IQR 50-256 days), 75 days for SF RT and 139 days for MF RT. In competing risks MVA, SINS ≥11 [hazard ratio (HR) = 2.5 (95% CI 1.3-4.8) p = 0.006], SF RT [HR = 6.1 (2.5-15.4), p<0.001] and post-RT systemic therapies [HR = 3.4 (1.3-8.8), p = 0.01] were the only significant predictors of AEs. To confirm the SF findings, all SF RT cases (73 sites in 67 patients) in 2008-2013 were reviewed. One-to-one propensity-score matched case-control analysis (matched for demographics, performance status, histology, SINS, and post-RT systemic therapies) confirmed the high HR for AEs with SF [HR = 3.96 (1.1-14.8) p = 0.04] versus MF RT. At 90 days post-RT, AEs developed in 23% of SF RT cases versus 6% for MF RT. Single fraction RT and spinal instability were associated with greater risk of clinically significant adverse events in uncomplicated spine metastases. This suggests that for patients with expected survivals >2-3 months, 8Gy x 1 SF RT for uncomplicated spine metastases should be avoided; further studies are required.