Inferring immunological control mechanisms from TKI dose alterations in CML patients

Abstract

Recent clinical findings in chronic myeloid leukemia (CML) patients suggest that the risk of molecular recurrence after stopping tyrosine kinase inhibitors (TKI) treatment substantially depend on an individual, leukemia-specific immune response. However, it is still not possible to prospectively identify patients that will most likely remain in a long-term treatment free remission (TFR). Here, we use a mathematical model for CML, which explicitly includes an anti-leukemic (presumably immunological) effect and apply it to a set of patients (n=60) for whom BCR-ABL/ABL time courses had been quantified before and after TKI stop. We demonstrate that such a feedback control is conceptually necessary to explain long-term remission as observed in about half of the patients. Based on simulation results we classify the patient data sets into three different groups according to their predicted immune system configuration. While one class of patients requires a complete CML eradication to achieve TFR, other patients are able to control the leukemia after treatment cessation. Among them, we identified a third class of patients, which only maintains TFR if an optimal balance between leukemia abundance and immunological activation is achieved before treatment cessation. Further, we demonstrate that the immune response classification of the patients cannot be obtained solely from BCR-ABL measurements before treatment cessation. However, our results strongly suggest that changes in the BCR-ABL dynamics arising after system perturbations, such as TKI dose reduction, holds the information to predict the individual outcome after treatment cessation.

Modelling of immune response in chronic myeloid leukemia patients suggests potential for treatment reduction prior to cessation

“…We hypothesized that these complete time courses reveal patient-specific features with respect to CML progression, TKI response and immunological control mechanisms. In order to quantitatively address these aspects, we applied an established mathematical model of CML treatment (12,13,31), which explicitly considers interactions between leukemic cells and the immune system (Figure 1B, see Methods) (29).…”

“…The results of the DESTINY trial also raise the question whether other dose reduction strategies prior to stopping could further increase the overall success rate of TFR. To address this, we have adapted our previously suggested mathematical model of CML treatment and immune response (29) to describe extended CML time course data from the DESTINY trial. Thereby we obtain an indirect estimate of the leukemiaimmune interaction of a larger patient cohort, which we further use to illustrate a novel modeling strategy allowing us to explore whether amended schedules of TKI application, including different schedules of dose reduction, can influence the overall success rate of TFR.…”