Population Analysis of Myelosuppression Proﬁles Using Routine Clinical Data after the ICE (Ifosfamide/ Carboplatin/Etoposide) Regimen for Malignant Gliomas YOSHITAKA YANO, 1 TAKAAKI KODAWARA, 2 HARUYUKI HONGO, 2 IKUKO YANO, 2 YO KISHI, 3 JUN TAKAHASHI, 3 KEN-ICHI INUI 2 1 Center for Integrative Education of Pharmacy Frontier, Graduate School of Pharmaceutical Sciences, Kyoto University, 46-29 Shimoadachi-cho, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan 2 Department of Pharmacy, Kyoto University Hospital, Faculty of Medicine, Kyoto University, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan 3 Department of Neurosurgery, Graduate School of Medicine, Kyoto University, 54 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan Received 11 September 2008; accepted 28 January 2009 Published online 10 July 2009 in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.21731 ABSTRACT: We propose a simple and practical modeling approach for analysis of the data for myelosuppression after cancer chemotherapy, which can be applied when pharmacokinetic data are not available and several anticancer drugs were simulta- neously administered. The model equation is based on the probability density function for the Erlang distribution. The data for cell counts of leukocytes (white blood cell, WBC), platelets (PLT), and reticulocytes (RET) obtained in routine clinical laboratory tests after the ICE (ifosfamide/carboplatin/etoposide) regimen for cancer chemotherapy were retrospectively collected from 28 patients, and a population analysis was applied. The time course proﬁles could be well explained by the proposed model. The individual values of the time to reach the nadir were obtained by the Bayesian method, and their medians (days) were 16.8 for WBC, 12.8 for PLT, and 8.2 for RET. Such information would be useful to determine the day of visit for outpatients especially for additional treatment to prevent side effects such as infections. The model is simple and applicable to explain the time course proﬁles for myelosuppression irrespective of cell types, and also practical because it requires only the data from routine clinical laboratory tests without any additional burden to patients. ß 2009 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 98:4402–4412, 2009 Keywords: erlang distribution; pharmacodynamics; population analysis; myelosup- presion; cancer chemotherapy; ICE; modeling and simulation INTRODUCTION In the treatment of cancer with chemotherapy, hematological toxicity is often a dose-limiting factor and medical staff need to know the possible lowest counts (nadir) of blood cells such as leukocytes and the time to the nadir (T nadir ) after the administration of anticancer drugs. Under- standing such time course proﬁles and the variability of myelosuppresion is useful for determining the day of visit for outpatients and for planning additional treatments to prevent side effects such as infections. 1 Correspondence to: Yoshitaka Yano (Telephone: 81-75-753- 9254; Fax: 81-75-753-9254; E-mail: yano@pharm.kyoto-u.ac.jp) Journal of Pharmaceutical Sciences, Vol. 98, 4402–4412 (2009) ß 2009 Wiley-Liss, Inc. and the American Pharmacists Association 4402 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 98, NO. 11, NOVEMBER 2009