An Evaluation of Prognostic Factors of Elderly Patients With Low Grade Glioma

Abstract

Low grade gliomas (LGG) include a heterogeneous group of tumors generally diagnosed in younger individuals. Understanding how the clinical and pathologic features should influence our treatment recommendations may differ in older patients (≥60 years) compared to younger individuals. With limited data available at individual institutions, we utilized the National Cancer Data Base (NCDB) to examine the impact of clinical and treatment factors on outcomes. We identified patients in NCDB diagnosed with low-grade glioma from 2004-2013 with age >60 years and known treatment records with associated outcomes. Univariate and multivariate Cox proportional hazard models were used to assess covariates related to overall survival (OS). Overall, 2,993 patients met the inclusion criteria with the diagnosis of low-grade glioma. Median age was 69 years at diagnosis. Sixty-eight% were astrocytoma, 22% were oligodendroglioma, and 10% were mixed with a median tumor size of 4.0 cm. For those with known surgical status, gross total resection was achieved in 29%, subtotal in 27%, and biopsy in 44%. Forty-six% received radiation therapy (RT), 28% received chemotherapy, and 21% received chemoradiotherapy. On multivariable analysis worse OS was significantly associated with older age (HR=1.05, p<0.001), higher Charlson/DeyoScore (HR=1.50, p<0.001), community reporting facility (HR=1.25, p<0.001), earlier year of diagnosis (HR=1.38-1.44, p<0.001), astrocytoma/mixed histology (HR=1.66-2.36, p<0.001), eloquent location (HR=1.14, p=0.036), and those not undergoing resection (HR=1.51, p<0.001) were significantly associated with worse OS on multivariate analysis. With respect to overall survival, tumor size and chemotherapy were not significant on univariate analysis whereas RT was significant on univariate but not on multivariate analysis. Patients who were treated at an academic center were more likely to undergo surgical resection (49% vs 38%, p<0.001) and more likely to undergo gross total resection (9% vs 6%, p<0.001). The importance of surgery, and where a patient undergoes evaluation/treatment, should not be underestimated in elderly patients. While previous factors to determine treatment strategies such as age, medical comorbidities, surgical resection, histology, and location are significant in older patients, tumor size may not be. Further studies are needed to better define how to optimize treatment outcomes for elderly patients with low-grade gliomas.