Change in pattern of relapse in newly diagnosed high-grade glioma following bevacizumab therapy.

Abstract

2020 Background: Local recurrence (LR) accounts for > 90% at the time of relapse in newly diagnosed high-grade glioma (HGG) following conventional therapy. Use of bevacizumab has shown impressive radiologic and clinical responses in both newly diagnosed and recurrent HGG. However, an apparent increase in diffuse invasive relapse (DIR) following therapy necessitated a detailed analysis of the pattern of recurrence. Methods: Seventy five patients with newly-diagnosed HGG were treated with radiation therapy and concomitant temozolomide (75 mg/m2 daily for 42 days) with bevacizumab (10 mg/kg every 2 weeks), 29 days following surgery, followed by 6 cycles of adjuvant TMZ (150 mg/m2 × 7d, q28 with bevacizumab at 10 mg/kg days 8 and 22 of each 28 day cycle. Recurrence was defined according to MacDonald Criteria. Pattern of recurrence and time to relapse were the primary aims of this study. DIR was defined as involvement of multiple lobes ± crossing the midline. Results: With a mean follow-up of 6.0 months (range 1-20), 40 patients have relapsed with 26 patients failing as DIR (65%). Seven patients who presented with multicentric HGG relapsed as DIR. The median progression-free survival (PFS) and overall survival (OS) were 8 (range 1-21 months) and 15 months (range 1-24 months) respectively for the entire group. The median OS for the patients who failed as DIR was 12 months vs. 4 months for patients with LR alone (p = 0.651). While the hazard risk for LR remained linear over time (R2 = 0.53), the risk of DR increased exponentially (R2 = 0.933). Biopsies done on five of these patients at the time of recurrence showed decreased expression of Olig-2 in 4 and increased expression of SDF-1 in 4 patients. No change in expression of HIF-1, p53, PDGF or CD163 was seen. Conclusions: Invasion remains a dominant process in newly diagnosed HGG treated with bevacizumab therapy. Unopposed angiogenic blockade resulting in increased invasiveness needs to be addressed in future clinical trials. No significant financial relationships to disclose.