Population pharmacodynamic model of the effects of recombinant human interleukin 21 on platelets in humans
Dodds, M.G. (1), J.E. Visich(1), I. Nestorov(1), R. Overgaard(2), M.C. Rogge(1)
(1) ZymoGenetics, Inc., USA; (2) Novo Nordisk A/S, Denmark.
Objectives: Interleukin-21 is a cytokine that is produced by activated CD4+ T cells and increases proliferation, activity and survival of T, B and NK cells. Recombinant human interleukin-21 (IL-21) is under investigation for the treatment of metastatic melanoma (MM) and renal cell carcinoma (RCC). A pharmacodynamic (PD) effect of the drug is the reduction of platelet counts in blood during administration followed by a rapid recovery beyond baseline counts during recovery. The aim of this study was to characterize this pharmacodynamic effect of IL-21 using population analysis approach.
Methods: Forty-three patients diagnosed with stage IV MM or RCC were enrolled in a phase 1 dose-escalation study with doses of 3, 10, 30, 50 or 100 ug/kg of IL-21, administered intravenously once daily for five days followed by nine or more days of rest before readministration. Platelet counts were available from standard hematology laboratory tests taken throughout the treatment course. These data were described by an indirect population PD model of enhanced clearance of platelets from the blood via a delayed, nonlinear drug effect. Megakaryocyte synthesis of platelets was assumed to be sensitive to the blood counts of platelets, and was described by a sigmoid function parameterized by the current and baseline blood platelet count, platelet half-life and coefficient.
Results: The model adequately described the data over all dose levels. Three random-effects were selected based on the NONMEM objective function values and goodness of fit considerations. The population pharmacodynamic parameters and coefficient of variation were: half-life of platelets, 3.31 d (13.5%), IL-21 mean effect time, 2.33 d (9.44%), baseline platelet counts, 231*10^9/L (4.50%), maximum possible fold-increase in platelet clearance by IL-21, 5.49 (25.7%), dose that induced half-maximal platelet clearance, 21.9 ug/kg (37.4%) and sigmoid coefficient, 1.71 (20.0%). Between-subject variability, expressed as coefficient of variation was introduced for: half-life of platelets, 36.1 %, baseline platelet counts, 28.9%, and maximum possible fold-increase in platelet clearance by IL-21, 34.5%.
Conclusion: This model accurately described data collected in different patient groups and dose levels. The interaction of delayed appearance of de novo platelets and the regulatory loop model components captured the observed initial decrease in platelets and subsequent rebound. These results may be useful in designing future studies.