A semi-mechanistic population Pharmacokinetic-Pharmacodynamic (PKPD) model of thrombocytopenia characterizing the effect of trastuzumab-DM1 (T-DM1) on platelet counts in patients with HER2-positive metastatic breast cancer
B. Bender(1,2), F. Schaedeli-Stark(3), R. Koch(3), A. Joshi(1) , Y-W Chu (1), H. Rugo (4), I.E. Krop(5), S. Girish(1), L.E. Friberg(2), M. Gupta(1)
(1)Dept of Pharmaceutical Biosciences, Uppsala University, Sweden, (2)Genentech Inc, San Francisco, CA, (3)F. Hoffman-La Roche Ltd, Basel, Switzerland, (4)UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; (5)Dana Farber Cancer Institute, Boston, MA.
Objectives: T-DM1 is an antibody-drug conjugate in development for HER2-positive metastatic breast cancer. In clinical trials of single-agent T-DM1, thrombocytopenia (TCP) was the dose-limiting toxicity[2,3]. In some patients, platelet time profiles drifted slowly down after multiple cycles of T-DM1. A population PKPD model was developed to describe the time course of patient platelet response to T-DM1, to support mechanistic hypotheses for platelet response, and to predict patient platelet response in future clinical trials.
Methods: A semi-physiologic population PKPD model with transit compartments mimicking platelet development and circulation was fit to the concentration-platelet-time data from a phase I dose escalation study (N=52) and a phase II study of T-DM1 (3.6 mg/kg every 3 weeks (q3w); (N=112). Overall, 4340 platelet counts were used in the analysis, with dose levels from 0.3-4.8 mg/kg q3w and from 1.2-2.9 mg/kg weekly. The rate and extent of platelet drift was modeled as a time and drug concentration effect on the platelet proliferation pool. NONMEM software was used for the modeling analyses, with a mixture model approach used to identify two patient groups with different rates of platelet drifts. Patient baseline characteristics were tested as covariates on the final model parameters. A separate phase II study (3.6 mg/kg q3w; N=110) was used for model evaluation.
Results: The PKPD model described the data well and predicted the 9% incidence of grade ≥3 TCP observed in the evaluation dataset. Two populations of patients were identified by the model, with platelet profiles from 25% of the patients drifting downward more rapidly before stabilizing by 8 treatment cycles (cycle = 3 weeks) to typically 50% of the original baseline platelet count. Patient baseline characteristics were not significant covariates for platelet kinetics.
Conclusions: The model accurately predicts clinical observations of platelet counts during T-DM1 treatment and supports T-DM1 3.6 mg/kg q3w as a well-tolerated dose, requiring minimal dose delays/ reductions for TCP. In addition, the model supports partial depletion of the platelet proliferation pool as a mechanism of the downward drift in platelet count. Platelet response to T-DM1 could not be predicted a priori from baseline characteristics.
 Lewis-Phillips GD, Li G, Dugger DL et al (2008) Targeting HER2-positive breast cancer with trastuzumab-DM1, an antibody-cytotoxic drug conjugate. Cancer Res 6:9280-9290.
 Krop IE, Beeram M, Modi S et al (2010) Phase I Study of Trastuzumab-DM1, a HER2 Antibody-Drug Conjugate, Given Every 3 Weeks to Patients With HER2-Positive Metastatic Breast Cancer. J Clin Oncol 28: 2698-2704.
 Burris H, Rugo H, Vukelka S, et al (2011) Phase II study of the antibody drug conjugate Trastuzumab-DM1 for the treatment of Human Epidermal Growth Factor Receptor 2 (HER2)-positive breast cancer after prior HER2-directed therapy. J Clin Oncol 29: 398-405.