Single Institutional Experience with Tumor-Treating Field Compliance and Overall Survival in Patients with Primary Glioblastoma

Abstract

To determine predictors of tumor treating field (TTFields) use and overall survival in patients receiving combined modality therapy for primary glioblastoma (GBM). Between 2015 and 2019, 65 patients underwent maximal surgical debulking, completed external beam radiotherapy with concurrent TMZ chemotherapy, and initiated adjuvant TMZ. We performed this retrospective univariate and multivariate analysis of patient, tumor, and treatment related factors to assess predictors of TTFields use and its association with overall survival. Of the 65 patients, 47 were male, and the median age was 62 years (range, 34 to 87). ECOG performance status was 0 in 42 patients, 1 in 16, 2 in 5, and 3 in 2. Forty-two patients underwent gross total resection, while 14 had subtotal resection, and 9 had biopsy only. With a median follow up of 12.4 months, the median overall survival for all patients was 18 months, and the 1-year actuarial survival was 74%. The median compliance for TTFields use was 67.2% (range, 9.4 to 96%). We divided TTFields usage into 3 subgroups: 21 patients who declined to start TTFields, 28 patients with monthly usage of 1 to 75%, and 16 patients with usage greater than 75%. Of the 44 patients who elected to use TTFields, 31 were age 65 or younger, 40 had an ECOG status of 0-1, and 30 underwent gross total resection. There was a trend towards TTFields users being younger (mean age 64 vs. 59, p=0.08), but no other associations with TTFields use were noted. Univariate analysis of factors associated with 1-year survival revealed TTFields use (refused, 56% vs. usage 1-75%, 81% vs. usage >75%, 83%, p=0.007), ECOG performance status (ECOG 3, 0% vs ECOG 2, 80% vs ECOG 1, 86%, vs ECOG 0, 60%, p<0.001), MGMT status (unmethylated, 66% vs methylated, 84%, p = 0.012), and extent of resection (biopsy only, 40% vs. STR, 69% vs. GTR 83%, p=0.005) as significant. On multivariate analysis TTF compliance as a continuous variable (p<0.001), MGMT status (p=0.001), and extent of surgical resection (p=0.001) remained significant. Our single institutional results confirm the randomized trial data in that TTFields use increases OS and the magnitude of the benefit is dependent upon the degree of TTFields usage. We were unable to find any factor associated with TTFields use vs. non-use even though all of our patients were eligible. We are currently investigating methods of behavioral modification to increase both TTFields use and the degree of use so that more patients benefit from TTFields therapy.