Purpose: To improve the predictive capacity of a semi-mechanistic myelosuppression model for neutrophils as the model have shown to over-predict the nadir of neutrophils and, secondly, to develop a model describing the time-course of leukocytes and neutrophils simultaneously.
Experimental design: The study included 601 cancer patients treated with a 1 h infusion of docetaxel in monotherapy. A total of 3,549 pairwise observations of leukocytes and neutrophils from one treatment cycle were analyzed simultaneously in NONMEM.
Results: A basic model was developed consisting of a neutrophil and a non-neutrophil model, each with the same structure as the semi-mechanistic myelosuppression model. The leukocytes were modeled as the sum of the predicted neutrophils and non-neutrophils. The model described the time-course of the leukocytes well, but was not able to capture the nadir of the neutrophils. Hence the model was further refined and the included modifications (p < 0.001) in the final model are a sigmoid Emax functions for the drug effect, feedback functions on the cell maturation time in bone-marrow and an optimized number of transit compartments for each of the two cell types.
Conclusions: A joint semi-mechanistic myelosuppression model describing the time-course of leukocytes and neutrophils following docetaxel administration was developed. The data supported a more complex model compared to the previous model developed by Friberg et al. (2002), and increased the model's capacity to accurately describe the time-course of neutrophils following docetaxel therapy. The combined model also illustrates the differences between the cell types and allows prediction of neutrophil counts from leukocyte measurements.