Linear Quadratic Model and Beyond, for Combined Therapies Against Brain Cancers
Abstract
Purpose
Glioblastoma (GBM) remains a lethal brain cancer with a devastating median survival of about 15 months with the current standard of treatment involving surgery, radiation therapy, and chemotherapy utilizing temozolomide (TMZ). Radioimmunotherapy (RIT), combining RT with immunomodulatory agents like lenalidomide, as well as nanoparticle mediated radiotherapy are promising combination modalities. Accurate biophysical modeling of cell survival from in vitro assays is critical for optimizing such combined therapies. The Linear-Quadratic (LQ) model is the standard for interpreting clonogenic survival data following RT. This work investigates the application and limitations of the LQ model for characterizing the effects of single-agent and combined lenalidomide-RT treatments in GBM cell lines.
Methods
Clonogenic survival assays were performed on human GBM cell lines U87 and T98G treated with RT alone (0-50 Gy, Faxitron CellRad), lenalidomide alone, and their combination, as well as NPRT. Survival curves were fitted with LQ models.
Results
Interestingly, many of the survival curves had up to 6 decades on the survival fraction axis, leading to significant deviation of alpha/beta parameters from expected values for the literature. Classic explanation for such deviations exist: LQ model fails beyond about 3 decades of cell killing model’s quadratic component causes it to over-predict the lethality of large radiation doses. We will present both the LQ model results and alternative models such as the universal survival curve model.
Conclusion
Our work demonstrates a noted limitation of the LQ model and provides impetus for exploration of complementary/alternative radiobiological models to better capture experimental results.