Optimization of chronic lymphocytic leukemia treatment using game theory

Abstract

The current strategy of chronic lymphocytic leukemia (CLL) treatment is based on genetic risk factors such as del(17p), TP53 mutations and/or unmutated variant of IGHV genes. Guidelines recommend the usage of targeted drugs, e.g. ibrutinib, in the first line for patients with unfavorable risk factors due to dismal results of other treatment options. Unfortunately, in real-life treatment decisions are often made without full knowledge of genetic risk factors in the treated patient. Our aim was to find the optimal therapeutic strategy for such patients, that is, those providing the best 5-year progression-free survival (PFS). Using a relatively simple game theory-based approach we here show, that currently, the used strategy is more advantageous (success rate 71%) compared to administration of immunochemotherapy to all patients (success rate with fludarabine + cyclophosphamide + rituximab — 45%, bendamustine + rituximab — 32%). However, the optimal strategy for CLL treatment in the conditions of unknown genetic risks is the administration of ibrutinib to all patients (success rate 73%). Our simple method can be used for optimization of treatment strategy of any oncologic disease and can be integrated into relevant clinical decision support systems.