Identifying drug combinations for the treatment of resistant acute myeloid leukemia patients

Introduction: AML is the most common form of acute leukemia in adults and is classified into 14 different groups depending on its genetic makeup. The heterogeneity of AML is defined by a diverse genetic landscape and is one of the major challenges in finding an effective treatment option for patients that do not respond to current standard treatment, which remained unchanged for the past 20 years and is applied across all 14 different groups not considering the genetic diversity of this disease.

Material and method: Therefore, we aim to identify the characteristic properties behind the response and resistance of individual patients, who might not only develop resistance to standard treatment, but also to targeted therapy. The ability to predict the potential risk of resistance to a treatment and identify strategic and personalized treatment options for a group of individual patients is of substantial significance. We have screened a group of patients by assessing the sensitivity of primary patient-derived cancer cells ex vivo from individual AML patients and healthy donors to a panel of anticancer drugs.

Results and discussion: We observed that only a group of patient-derived cancer cells showed response to a number of drugs. However, the other group did not show any therapeutically relevant response to the most effective drugs indicating the potential of resistance in an eventual treatment. We are approaching this therapeutic issue through a systematic screening of synergizing drug combinations. We have been able to develop computational methods for personalized single and/or combinatorial drug-sensitivity screens from high-throughput experiments including randomized dispensing and automated deconvolution of big data for further qualitative downstream analysis.

Conclusion: We further aim to map potential genetic alterations and immunological phenotypes to drug response or resistance and to identify potential biomarkers through multidimensional data together with the implementation of automated high-throughput screening methods in the hope not only to predict response and resistance, but also identify strategic treatment options for individual AML patients.