Abstract CT160: Modeling long-term survival outcomes in glioblastoma patients treated with TTFields plus temozolomide and temolozomide monotherapy

Abstract

Abstract Background: Glioblastoma is the most common and aggressive primary brain tumor. Survival times for GBM have improved significantly since 2005. Researchers in 2005 demonstrated that adding TMZ, an oral alkylating agent, to surgery and radiation extended median overall survival from 12 to 15 months, and 2-year survival rates from 11% to 27%. More recently the addition of Tumor Treating Fields (TTFields), a loco-regionally delivered anti-mitotic treatment, to TMZ demonstrated an increase in median overall survival of 5 months, and survival rates of 43% and 17% at 2-years and 4-years respectively. Trial data for TMZ and TTFields both demonstrated increasing conditional probabilities of survival as patients survived longer from diagnosis. Increasing conditional probability of surviving GBM has also been reported in epidemiological literature. Accordingly, standard parametric survival models based on regression analysis will not provide reliable estimates of lifetime survival benefits and an alternative modelling method is provided here. Clinicians and policy makers need to understand the lifetime survival estimates for GBM patients, as the rate of GBM patients surviving to 4-years and 5-years is now significant. Methods: Lifetime patient survival was simulated using a synthesis of parametric survival curves fit to the recently completed EF-14 Trial’s 4-year follow-up data and conditional survival probabilities from a SEER state cancer registries database analysis. The 4-year parametric curves were extrapolated to estimate 5-year survival. We then used GBM-specific conditional survival estimates to model patient survival at 10 and 15 years given survival to year 5, and assumed patients surviving beyond 15 years had survival equivalent to population averages. We performed one-way and probabilistic sensitivity analyses to assess result uncertainty due to parameter variability. Results: Our modeled analysis showed that TTFields plus TMZ resulted in 3.63 (95% credible range [CR], 3.32 to 3.93) life-years versus 2.53 (95% CR, 2.36 to 2.69) life-years for TMZ monotherapy, a difference of 1.10 years (95% CR, 0.96 to 1.23) discounted; the difference was 1.56 years (95% CR, 1.33 to 1.80) undiscounted. The results were most sensitive to uncertainty in the 10- and 15- year survival estimates conditional on survival to years 5 and 10, respectively. Conclusions. This analysis presents a new approach to model lifetime survival benefits of adding TTFields to TMZ for GBM by combining long-term clinical trial data and epidemiological data. The model demonstrates that adding TTFields to TMZ for GBM increases lifetime survival substantially. Citation Format: Greg Guzauskas, Bruce Wang, Marc Salzberg. Modeling long-term survival outcomes in glioblastoma patients treated with TTFields plus temozolomide and temolozomide monotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT160. doi:10.1158/1538-7445.AM2017-CT160