Abstract 5065: Predicting response to chemotherapy in a mouse model of acute myeloid leukemia

Abstract We have previously shown that state transition theory is applicable to acute myeloid leukemia (AML) to predict disease evolution. This mathematical model represents AML evolution from health to disease as a state-transition of the transcriptome represented as a particle undergoing Brownian motion in a double-well quasi-potential, where critical points define states of perturbed hematopoiesis (c1), transition to AML (c2), and overt AML (c3). We successfully demonstrated that the state transition model predicted AML onset. We now test the applicability of AML state-transition model to predict disease response to chemotherapy. To this end, we performed a time-series experiment using the conditional Cbfb-MYH11 (CM) knock-in (Cbfb+/56M/Mx1Cre) mouse model recapitulating inv(16) AML. CM leukemic mice were treated with a combination of cytarabine (50mg/kg/day; 5 days) and daunorubicin (1.5mg/kg/day, 3 days) after detection of overt leukemia which is monitored by circulating leukemia blast (cKit+ > 20%) to model the 7+3 standard of care treatment for newly diagnosed AML. A total of 110 peripheral blood samples from 7 CM mice were collected weekly before, during, and following chemotherapy and subjected to RNA-sequencing. The singular value decomposition was used to construct the transcriptome state-space and identified dynamics of AML (c3) after administration of chemotherapy treatment. Gene expression profiles following treatment revealed dynamics consistent with the state-transition model, with the transcriptome particle moving from leukemia (c3) towards a state of perturbed hematopoiesis (c1), before eventual relapse, re-crossing the transition point (c2) back to overt AML (c3). All 7 CM mice achieved a partial response, defined as the transcriptome particle crossing the c2 critical point, with a mean time to relapse of 5 weeks, defined as the time of the first observation after the particle crosses back over c2 towards the leukemic state c3. Mean arrival time analysis was used to accurately predict the extent of response, defined by the transcriptome particle in the state-space, and the time to relapse. We successfully applied state-transition mathematical model to predict treatment response and the time to relapse in all CM mice, confirming the applicability of the model to post-chemo therapy disease dynamics. This predictive model has implications to improve therapeutic strategies by targeting transcriptome state-transition critical points in human AML. Citation Format: Lisa Uechi, Sergio Branciamore, David E. Frankhouser, Denis O'Meally, Lianjun Zhang, Ying-Chieh Chen, Man Li, Guido Marcucci, Ya-Huei Kuo, Russell Rockne. Predicting response to chemotherapy in a mouse model of acute myeloid leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5065.