Introduction: Thrombocytopenia, prevalent in hematological malignancies and exacerbated by drugs, increases bleeding risks and constrains therapy dosages. Developing predictive modeling tools, capturing the system under various perturbations is important for clinical decision making.
Objective: Build a pragmatic systems model describing platelets dynamics in healthy individuals and cancer patients and evaluate its performance under different therapeutic interventions.
Methods: A minimal ODE model was developed to describe the mechanisms underlying thrombopoiesis and platelet distribution in the body. Thrombopoiesis was modeled using transit compartments [1], to capture megakaryopoiesis and platelet egress to the bloodstream. The bone marrow module consists of a megakaryocyte proliferating pool (pMK) with five transit compartments (1 day each) each.
The model included a regulatory mechanism on the proliferation of pMK by using the total amount of platelets in the circulation as proxy for thrombopoietin, in addition to the relative amount of pMK, which controlling the strength of the feedback signal to the bone marrow compartment.
The terminally differentiated megakaryocytes release platelets partitioning the output flux between spleen and systemic circulation in 1:2 ratio [2]. Circulating platelets have a lifespan of 9 days (9 transit compartments replicated in spleen and systemic circulation). The spleen works as platelets reservoir; first order release was implemented to represent return to homeostatic conditions.
The model’s parameters were mostly informed from literature. Chemotherapy toxicity was implemented in the model to understand the role of the spleen depot and to characterize the strength of the chemo-induced feedback.
Results: The model structure includes key cell populations with their reported lifespans and distribution to the spleen and circulation. The model was parametrized to capture 1) the system at equilibrium and 2) under perturbation conditions. CHOEP chemotherapy was implemented as a direct hit on each bone marrow cell stage (pMK and MKi). The model was fitted towards circulating platelets count over several CHOEP cycles. Spleen release was pivotal in countering the early toxicity by compensating for reduced platelet egress from the bone marrow. It mitigated the proliferation feedback during the off-drug period, facilitating a faster return to homeostasis without prolonged count oscillations.
Conclusion: This minimal model, parametrized from first principles and with literature data, integrates key aspects of thrombopoiesis and efficiently predicts post-chemotherapy platelet counts. The CHOEP study suggests that the spleen compartment plays a role in replenishing platelets in the circulation while waiting for the feedback to production and maturation to kick in. With no further complexity, this modeling framework will be adapted to a NLME approach and applied to multiple therapies (BETi-s and ADC-s.)
Citations: [1] Friberg et al, J Clin Oncol. 2002 Dec 15; 20(24): 4713-21
