Original article
Introduction Docetaxel is a semi-synthetic taxane derivate neoplastic agent which is used in the treatment of breast and castration-resistant prostate carcinoma (CRPC), and several other cancers.8 Pharmacokinetics of docetaxel shows a high inter-individual variability in clearance, which may result in under- or overdosing.1 In order to reduce this variability, doses are currently based on body surface area (BSA).2
Dosing of anticancer drugs based on BSA is common practice since the 1950’s.3 Dosing on BSA is the method for predicting a safe starting dose in phase 1 human trials translated from animal toxicology data.3,4 However, many argued whether this approach results in an optimal dose for each individual. BSA-based dosing has shown to result in high inter-individual variability in drug exposure for most anticancer drugs leading to undesirably side effects or insufficient tumour response. Therefore, BSA-based dosing is much debated as the method of choice for dosing chemotherapeutics.1,5 Pharmacokinetic parameters such as clearance and area under the curve (AUC) are known markers for predicting therapeutic responses.2 A study of Engels et al showed that the application of therapeutic drug monitoring (TDM) significantly decreased the inter-individual variability in docetaxel exposure when compared to BSA based dosing.6 Although TDM is an elegant way of dose optimisation, it is very labour intensive and costly in the day-to-day clinical setting. Therefore, alternative anthropometric parameters that correlate better with drug exposure may be considered to optimize dosing of anticancer drugs.7,8
Docetaxel doses vary from 75-100 mg/m2 given once every three weeks during an one hour intravenous infusion. Docetaxel is metabolised in the liver via oxidation by CYP3A4, CYP3A5 and is bound to albumin for 95% without significant renal clearance.9-11 The pharmacokinetics of docetaxel can be best described by a three-compartment model with half-life times of 4.5 minutes, 38.3 minutes and 12.2 hours, respectively. The area under the plasma concentration time curve (AUC) increases proportionally with increasing doses. Docetaxel is distributed in tissues with a mean volume of distribution (VD) of 74 L/m2.9Docetaxel is characterized by highly interindividual pharmacokinetic variation, with up to 10-fold differences in drug clearance in patients with normal hepatic function.12 Bruno et alfound a median clearance of 36.6 L/h (5th to 95th range 17.5 L/h to 59.3 L/h).13 This variability may lead to adverse effects or to suboptimal treatment or even treatment failure. The major adverse effect is neutropenia, which is dose-limiting most of the times.9 Other frequently occurring side effects of docetaxel are anaemia, alopecia, nausea, asthenia, peripheral neuropathy, fluid retention and nail toxicity.10,11
Lean body mass (LBM) could be an alternative dosing parameter to BSA, since LBM has shown to correlate better with drug clearance of cisplatin, paclitaxel and troxacitabine when compared to BSA or total body weight (TBW) in obese patients.14 LBM as a dosing parameter has been investigated for several anticancer drugs.7,15-23
Patients with comparable BSA values showed a wide variety in liver volume and LBM. Since liver volume is strongly correlated with LBM and docetaxel is mainly metabolized by the liver, it is hypothesized that individual dosing on LBM should be preferred over BSA.24,25
In our study we examined which of the anthropometric and body composition parameters BSA, LBM and TBW correlates best with docetaxel clearance (CL) and exposure (AUC). In addition, LBM, TBW and a fixed dose were compared to BSA as dosing parameters for dose individualisation of docetaxel.
Methods Patients and study We performed a multi centred prospective study in patients using docetaxel. Patients that received chemotherapy with docetaxel for breast cancer or castration-resistant prostate carcinoma (CRPC) were included. Docetaxel in breast cancer treatment is part of a combined therapy with cyclophosphamide and doxorubicin. In patients with CRPC docetaxel was given as monotherapy. Other criteria for inclusion were: absolute neutrophile count: > 1.5x109 /L, serum creatinine ≤ 2x ULN, total bilirubin < 1.5 ULN. Exclusion criteria were docetaxel use in the last year, moderate or severe liver impairment ([ALAT and/or ASAT ≥ 1.5 ULN] and [AF ≥ 2.5 ULN]), current therapy with any drug, dietary supplements, or other compounds known to inhibit or induce CYP3A4. Every patient received 75 or 100 mg/m2 docetaxel dissolved in a saline solution and infused over one hour. Estimation of the study population size, 36 participants, was derived from studies by Gusella and Prado.7,15 The study was conducted in accordance with the Declaration of Helsinki and all study participants provided written informed consent before study entry.
Body composition measurements TBW was measured using a medical body weight scale in kg. A fixed stadiometer was used to determine the patients height, standing barefoot against a straight wall. LBM was measured by a DEXA-scanner. In the Deventer Teaching Hospital patients were scanned by a GE Lunar Scanner (GE healthcare, Little Chalfont; United Kingdom), while patients in Radboud University Medical Centre were scanned with a Hologic Discovery scanner (Hologic, Bedford, U.S.A.).
PK sampling and analysis Pharmacokinetic blood samples were obtained at t=0 (just prior to infusion), t=30 minutes after start of infusion, t=55 minutes after start of infusion (i.e. just prior to the end of infusion) and a last sample was obtained at t=180 minutes after start of infusion, according to a validated limited sampling strategy.26-28Docetaxel plasma concentrations were quantified using a HPLC-UV method. First liquid-liquid extraction was performed with tert-butylmethylether as extraction fluid/organic layer. The plasma layer is frozen on a cryo bath and the organic layer is evaporated with compressed air. HPLC-UV analysis was done with paclitaxel as internal standard. Methanol/phosphate buffer (65:35 v:v) was used as mobile phase, flow: 1.0 mL/min, detection 230 nm. The method was validated in line with the European Medicines Agency Guideline on bioanalytical validation.29 NONMEM® software (ICON, Ireland) was used to determine the individual clearance (CL) and distribution volume (Vd) by Bayesian analysis using a population model described by Engels et al.6 The area under the curve (AUC) was calculated with the formula:
\begin{equation} \mathrm{AUC=\ }\frac{\mathrm{Dose\ (D)}}{\mathrm{Clearance\ (CL)}}\nonumber \\ \end{equation}