(596at) Modeling Hematopoiesis: Clinical Applications of Population Balance Models

In humans, blood cells (erythrocytes, leukocytes, platelets) are the most dynamic population of cells with high growth rates.  Blood cells are derived from the long-term pluripotent stem cells residing in the bone marrow through hematopoiesis. Hematopoiesis is a complex process during which the stem cells are committed to a particular cell lineage, proliferate and mature to a specialized cell type according to the biochemical signals received from the periphery. Aberrations in these signaling pathways lead to several acute and chronic cyclical blood disorders which can be treated by growth factor cytokines. As the stem cells mature in an age-dependent manner, they progressively lose their proliferating capability and enter the peripheral blood as a fully functional cell. Thus, bone marrow is one of the most active and tightly regulated systems in the human body.

In addition to the blood disorders mentioned above, hematopoiesis is severely affected by chemotherapy, given the nature of most of the cytotoxic drugs in targeting fast growing cells. In both instance, i.e. therapy with growth factors and chemotherapy, peripheral blood count monitoring has become a routine to assess the bone marrow response. However, this is a reactive rather than a proactive approach because the bulk of the effect is sustained by the early bone marrow precursors which usually take between 5 to 12 days to present themselves in the periphery. This marked delay in the realization of the actual treatment effects in the periphery hinders the decision making process during therapy administration. Thus, accurate prediction of normal and diseased bone marrow cellular dynamics would help to forecast the oncoming spike/decline in the cell population and adopt corrective measures before the patient succumbs to the disease/side-effect.

Population balance modeling (PBM) provides an ideal framework for representing this age-dependent maturation and death process together with dose-dependent proliferation (with growth factor) and death process (with chemotherapy) during various stages of hematopoiesis. In this work, we employ PBM framework to model the hematopoiesis process under normal as well as diseased conditions. Influence of growth factors and chemotherapy on hematopoiesis is also considered. The model is validated using data from actual patients undergoing treatment with growth factor cytokines and chemotherapy. Based on the average patient response, dosing conditions are defined such that the cell population is maintained above a critical level.  Utilization of such predictive models in clinical settings will augment the decision making capabilities of the treating physicians.