Individual prediction of thrombocytopenia at next chemotherapy cycle: Evaluation of dynamic model performances

Abstract

Thrombocytopenia is a common major side-effect of cytotoxic cancer therapies. A clinically relevant problem is to predict an individual's thrombotoxicity in the next planned chemotherapy cycle in order to decide on treatment adaptation. To support this task, 2 dynamic mathematical models of thrombopoiesis under chemotherapy were proposed, a simple semimechanistic model and a comprehensive mechanistic model. In this study, we assess the performance of these models with respect to existing thrombocytopenia grading schemes. Methods: We consider close-meshed individual time series data of 135 non-Hodgkin's lymphoma patients treated with 6 cycles of CHOP/CHOEP chemotherapies. Individual parameter estimates were derived on the basis of these data considering a varying number of cycles per patient. Parsimony assumptions were applied to optimize parameter identifiability. Models' predictability are assessed by determining deviations of predicted and observed degrees of thrombocytopenia in the next cycles.Results: The mechanistic model results in better agreement of model prediction and individual time series data. Prediction accuracy of future cycle toxicities by the mechanistic model is higher even if the semimechanistic model is provided with data of more cycles for calibration. Conclusion:We successfully established a quantitative and clinically relevant method for assessing prediction performances of biomathematical models of thrombopoiesis under chemotherapy. We showed that the more comprehensive mechanistic model outperforms the semimechanistic model. We aim at implementing the mechanistic model into clinical practice to assess its utility in real life clinical decision-making. K E Y W O R D Sadverse drug reactions, biomathematical model, medical, oncology (oncology) pharmacodynamics, pharmacometrics, platelets (haematology)The notion of principal investigator is not applicable in our case. We performed no studies with human subjects/patients. All data are derived from published clinical studies.

Individual hematotoxicity prediction of further chemotherapy cycles by dynamic mathematical models in patients with gastrointestinal tumors

“…At later cycles, the prediction performances drop, in particular for very late cycles, which is not unexpected, because uncertainties in model parameter estimation result in accumulating prediction errors over multiple cycles. However, this is of minor practical importance, because the prediction of later chemotherapy cycles could easily be improved by considering all accomplished cycles of a patient (i.e., considering cycles 1-3 to predict cycle 4) as shown in our previous publication (Kheifetz and Scholz 2021). This issue was not investigated in the present study to reduce the complexity of analyses but could be considered in the future.…”

“…We assessed the predictive potential of the integrated model using a method proposed in a previous study (Kheifetz and Scholz 2021). In detail, we determined absolute deviations of predicted and observed degrees of hematotoxicity in all cycles except for the first one which was used for individual parameter estimation.…”

Individual hematotoxicity prediction of further chemotherapy cycles by dynamic mathematical models in patients with gastrointestinal tumors

“…However, this is of minor practical importance because the prediction of later chemotherapy cycles could easily be improved by considering all accomplished cycles of a patient (i.e. considering cycles 1-3 to predict cycle 4) as shown in our previous publication (Kheifetz and Scholz 2021). This issue was not investigated in the present study to reduce the complexity of analyses but could be considered in the future.…”

“…We assessed the predictive potential of the integrated model using a method proposed in a previous study (Kheifetz and Scholz 2021). In detail, we determined absolute deviations of predicted and observed degrees of hematotoxicity in all cycles except for the rst one which was used for individual parameter estimation.…”

Individual modelling of haematotoxicity with NARX neural networks: A knowledge transfer approach

“…Since pre-therapeutic risk models are not reliable, the idea is to improve individual predictions based on the observed data during the first therapy cycles. To solve this task, statistical risk models [1] , semi-mechanistic models of bone marrow haematopoiesis [3] , [4] , [5] , [6] and comprehensive mechanistic models [7] , [8] , [9] were proposed. The latter showed decent success in supporting individualised treatment decisions [9] .…”