SDPS-01 DEVELOPMENT OF A WEB-BASED CALCULATOR APPLICATION FOR PREDICTING SHORT- AND LONG-TERMSURVIVAL IN PEDIATRIC OPTIC PATHWAY GLIOMA: A POPULATION-BASED DATABASE ANALYSIS

Abstract

Abstract Optic pathway gliomas (OPG) are generally low-grade lesions arising from the astrocytes of the visual pathway, most commonly of the optic nerves and chiasm. Uncertainty remains regarding the optimal management of this patient population. We sought to leverage population-level data from the Surveillance, Epidemiology, and End Results program database to develop a predictive calculator for overall survival (OS) amongst pediatric patients with OPG. A total of 1,052 patients who had been diagnosed with OPG were included. Multivariate cox proportional hazard model demonstrated patients of “other” race (hazard ratio [HR]=4.109, 95% confidence interval [CI] 1.076-15.694, p=0.039) exhibited a decreased overall survival, while patients with a median household income ≥$55,000 (HR=0.172, 95% CI 0.065-0.457, <0.001) and those who underwent subtotal resection (HR=3.549, 95% CI 1.092-13.364, p=0.047) had a higher overall survival rate. An online, interactive survival prediction tool (https://spine.shinyapps.io/optic_glioma/) was created to estimate 5-, 10-, and 15-year overall survival probability as well as median survival months. C-index was 0.722 (95% CI 0.642-0.776) for the training set and 0.745 (95% CI 0.661-0.790) for the validation set. Good concordance between the predicted probability and the actual probability of the dataset was also observed. This calculator may serve as a useful clinical tool during patient counseling. Further investigations aimed at developing calculators of visual acuity preservation and progression-free survival are merited; progress toward this end will benefit from continued multicenter collaborations.