NCOG-22. RETROSPECTIVE ANALYSIS OF VISUAL OUTCOMES AFTER BEVACIZUMAB-BASED THERAPY IN OPTIC PATHWAY GLIOMA

BACKGROUND Patients with optic pathway glioma (OPG) are vulnerable to debilitating visual impairment. Consequently, vision stabilization is a primary treatment goal. Bevacizumab has demonstrated promising effects on radiographic tumor burden, but less is known about its impact on vision. Our objective was to characterize visual outcomes associated with bevacizumab-based therapy (BBT) in OPG. METHODS This is a single-institution, retrospective review of patients treated with BBT for OPG from 2011 to 2020. Ophthalmologic and radiographic data were abstracted before and after treatment. Clinically significant visual acuity (VA) impairment was defined as logMAR > 0.5 and change in VA was defined as change from baseline of logMAR ≥ 0.2. RESULTS Sixteen patients (13 sporadic OPG, 3 NF1-associated OPG) with evaluable vision outcomes were identified. Treatment indications were radiographic progression (N=15) and vision deterioration (N=4). Prior to BBT, 15 (94%) had failed at least one chemotherapy regimen. BBT regimens included bevacizumab/irinotecan (N=12), bevacizumab monotherapy (N=3) and bevacizumab/vinblastine (N=1). Nine patients (56%) had baseline VA impairment. Thirteen patients (81%) had stable or improved vision after BBT, including 8 of 9 with baseline VA impairment and all 4 patients with vision deterioration as a treatment indication. Eleven patients (69%) had radiographic progression following BBT (Median time-to-progression 66 weeks, IQR 27 weeks), 9 of whom had stable vision at time of progression. There were no associations between VA and age at treatment, NF1-status, histology, or BBT regimen. CONCLUSIONS BBT was associated with favorable visual outcomes for most patients with OPG in this modest retrospective cohort. Consistent with prior research, radiographic and ophthalmologic outcomes were discordant; a majority of patients experienced progressive disease despite stable vision. Next steps include (1) assessing visual field and optical coherence tomography outcomes in the same cohort and (2) comparing outcomes for BBT with other common therapies including carboplatin/vincristine and vinblastine.