Patterns of Failure Following Stereotactic Radiosurgery Salvage for Recurrent High-Grade Glioma

Abstract

<h3>Purpose/Objective(s)</h3> Stereotactic radiosurgery (SRS) is an increasingly used treatment for patients with progressive high-grade glioma (HGG). We sought to determine which demographic and clinical factors are associated with pattern of recurrence and outcome following SRS to better predict which patients would benefit from local salvage. <h3>Materials/Methods</h3> This retrospective analysis of a prospectively maintained database included continuous patients treated with SRS salvage at a single institution between January 2008 and May 2020. Patterns of failure were classified according to the distance from the prescription isodose line. Local failures included in-field (within the prescription isodose line), marginal (< 2 cm), and regional (2-5 cm). Distant failures were > 5 cm. Survival was estimated using Kaplan-Meier methods. The Wilcoxon rank-sum test, Kruskal-Wallis test, and Spearman correlations were used to examine associations between patterns of failure and demographic and clinical variables at first failure. Multilevel mixed-effects parametric regressions were used to identify predictors of SRS failure. <h3>Results</h3> One hundred forty-eight patients (males, 56.8%; non-Hispanic White, 72.3%) ages 20 to 83 years (median, 58 years) with HGG (WHO Grade IV, 92.6%) underwent salvage to a total of 412 targets. The median (interquartile range) size of target was 3.4 (1.0-8.7) cm³. The median dose was 18 (16-20) Gy. The median time from diagnosis to salvage was 12.1 (7.6-19.3) months. With a median follow up of 8.2 months, 62 (41.9%) patients developed progression following SRS. Progression-free survival at 6 and 12 months were 64.4% (95% confidence interval [CI]: 55.7-71.7) and 36.8% (95% CI: 28.7-45.0), respectively. Fifty (80.7%) patients developed local, two (3.2%) distant, and 10 (16.1%) concomitant local and distant failure. Actuarial local control at 6 and 12 months were 80.8% (95% CI: 72.4-86.8) and 61.8% (95% CI: 51.4-70.7), respectively. Of 60 local failures, 31 (51.7%) were in-field, 25 (41.7%) marginal, and four (6.7%) regional. Median overall survival was 26.6 months from diagnosis and 13.0 months from first GK treatment. Non-White race (HR = 0.32, 95% CI: 0.15-0.66) and GK dose (HR = 0.87, 95% CI: 0.79-0.95, P = 0.002) were independent predictors of SRS failure. Non-White race (r_s = .31, P = 0.012) and GK target size (X² (3, N = 62) = 10.09, P = 0.017) were positively associated with a more distant pattern of failure. Age at diagnosis, sex, and glioma grade were not associated (P > 0.05) with GK failure or pattern of failure. No Grade 4 or 5 toxicity was observed. <h3>Conclusion</h3> SRS salvage is a safe and effective treatment for select patients with progressive HGG. The dominant pattern of failure following salvage was local; approximately half of local progressions were in-field. Higher dose decreased the hazard for progression, indicating that some patients may benefit from dose escalation.