Mikiko Nakamura

Assessment of intestinal availability (FG) of substrate drugs of cytochrome p450s by analyzing changes in pharmacokinetic properties caused by drug-drug interactions.

In this study, we developed the drug–drug interaction (DDI) method as a new assessment technique of intestinal availability (FG, the fraction of drug transferred from the intestinal enterocytes into the liver, escaping from intestinal metabolism) based on the clearance theory. This method evaluates FG from changes caused by DDIs in the area under the blood concentration-time curve and in the elimination half-life of victim drugs. Application of the DDI method to data from the literature revealed that the mean and S.D. of FG values for 20 substrate drugs of CYP3A was 0.56 ± 0.29, whereas that for 8 substrate drugs of CYP2C9, CYP2C19, and CYP2D6 was 0.86 ± 0.11. These results were consistent with the fact that intestinal metabolism is mediated predominantly by CYP3A. The DDI method showed reasonable correlations with the conventional i.v./p.o. method and the grape fruit juice (GFJ) method (coefficients of determination of 0.41 and 0.81, respectively). The i.v./p.o. method was more susceptible to fluctuations in the hepatic blood flow rate compared with the DDI and GFJ methods. The DDI method evaluates FG separating from the absorption ratio (FA) although it requires approximation of FA. Since preciseness of approximation of FA does not greatly affect the evaluation of FG by the DDI method, we proposed a reasonable approximation method of FA for the evaluation of FG in the DDI method. The DDI method would be applicable to a broad range of situations in which various DDI data are utilizable.

Physiologically Based Absorption Modeling to Explore the Impact of Food and Gastric pH Changes on the Pharmacokinetics of Alectinib.

Alectinib, a lipophilic, basic, anaplastic lymphoma kinase (ALK) inhibitor with very low aqueous solubility, has received Food and Drug Administration-accelerated approval for the treatment of patients with ALK+ non-small-cell lung cancer. This paper describes the application of physiologically based absorption modeling during clinical development to predict and understand the impact of food and gastric pH changes on alectinib absorption. The GastroPlus™ software was used to develop an absorption model integrating in vitro and in silico data on drug substance properties. Oral pharmacokinetics was simulated by linking the absorption model to a disposition model fit to pharmacokinetic data obtained after an intravenous infusion. Simulations were compared to clinical data from a food effect study and a drug-drug interaction study with esomeprazole, a gastric acid-reducing agent. Prospective predictions of a positive food effect and negligible impact of gastric pH elevation were confirmed with clinical data, although the exact magnitude of the food effect could not be predicted with confidence. After optimization of the absorption model with clinical food effect data, a refined model was further applied to derive recommendations on the timing of dose administration with respect to a meal. The application of biopharmaceutical absorption modeling is an area with great potential to further streamline late stage drug development and with impact on regulatory questions.

231Pα-FETOPROTEIN (AFP) RESPONSE BY GC33 CORRELATES TO PROGRESSION FREE SURVIVAL (PFS) IN PATIENTS WITH ADVANCED HEPATOCELLULAR CARCINOMA (HCC)

ABSTRACT Aim: GC33 is a first-in-class recombinant, humanized mAb that binds to glypican-3 which is highly expressed in HCC. The aims of the analysis were to investigate AFP response after administration of GC33 in patients with advanced HCC and to assess the relationship between AFP response and PFS. Methods: Placebo and GC33 1600 mg biweekly with two loading doses were administered to previously treated patients with advanced HCC in a double-blind manner with 1:2 randomization. Seventy-one patients (placebo: GC33 = 24:47) who had AFP measurements at baseline (AFPB) and 6 weeks post treatment (AFP6W) were evaluated. Patients with steady state GC33 trough concentration higher than230 µg/ml (the median of the projected trough concentration on cycle 3 day1) were included in GC33 high exposure group. AFP response was defined as a 20% decrease in AFP at 6 weeks post treatment. Results: Twenty seven out of 47 patients treated with GC33 had trough concentration higher than 230 µg/ml. AFP change (%) from baseline at 6 weeks post treatment (calculated as (AFP6W-AFPB)/AFPB) in high exposure group was significantly lower than that in placebo and GC33 low exposure group combined (p = 0.029). In addition, 2 out of 24 (8.3%), 2 out of 20 (10%), and 8 out of 27 patients (30%) were AFP responder in placebo, GC33 low exposure group, and GC33 high exposure group, respectively. In landmark analysis (n = 71), AFP response was a prognostic factor of PFS (HR = 0.63, 80%CI: 0.41-0.97) regardless of treatment. Within GC33 treatment group, hazard ratio of PFS comparing AFP non-responder (n = 37) vs. AFP responder (n = 10) was 0.56 (80% CI: 0.34-0.91). Conclusions: GC33 high exposure group had better AFP response than placebo and low exposure group. Early assessment of AFP response may potentially be a valuable biomarker for predicting antitumor response in patients with advanced HCC. Disclosure: Yuyan Jin, Jun Shi, Alex Phipps and Ruey-min Lee have declared: is an employee of Hoffmann-La Roche; Mikiko Nakamura and Toshihiko Ohtomo have declared: is an employee of Chugai Pharmaceutical Co. Ltd.Ya-Chi Chen is an employee of Hoffmann-La Roche and also holds stocks of Hoffmann-La Roche.