Biochemical parameters
As a part of standard hospital protocol haematology and biochemistry assessments were done before every chemo course: aminotransferase (ASAT), alanine-amino transferase (ALAT), gamma-glutamyltransferase (GGT), alkaline phosphatase (ALP), alpha-1-acidglycoprotein (AAG), albumin (ALB), total bilirubin, creatinine, haemoglobin (Hb), hematocrit (Ht), red cell count (RBC), platelat count, total white cell count (WBC) and diffferential white cell count.
Data and statistical analysis Data were expressed as median with their interquartile range (IQR). Comparison between median values obtained in females and males were made using the Mann-Whitney U test for unpaired data. The accepted significance level was p<0.05. Linear regression analysis was used to compare correlations of BSA, LBM and TBW with docetaxel clearance and exposure. Correlations were therefore evaluated by determining Spearman correlation coefficients with corresponding p values.
Different doses were simulated, based on individual anthropometric and body composition parameters and median docetaxel/BSA, docetaxel/LBM, docetaxel/TBW and docetaxel dose (referred as fixed dose). An optimal target AUC was assumed to be the recommended docetaxel dose of 75 mg/m2 divided by all individual clearance values corrected for BSA. The difference between the optimal target AUC and simulated AUC results was evaluated by calculating accuracy using mean absolute percentage error (MAPE):
\begin{equation} \mathrm{MAPE=}\frac{\mathrm{1}}{\mathrm{n}}\mathrm{\ }\sum{\mathrm{|}\frac{\mathrm{\text{AUCsim}}\mathrm{\ }\mathrm{-}\mathrm{\ }\mathrm{\text{AUCtarget}}}{\mathrm{\text{AUCtarget}}}}\mathrm{|\ x\ 100}\mathrm{,}\nonumber \\ \end{equation}
where AUCsim denotes the simulated AUC results and AUCtarget denotes the optimal target AUC.
Bias was calculated using the mean error (ME):
\begin{equation} \mathrm{ME=\ }\frac{\mathrm{1}}{\mathrm{n}}\mathrm{\ }\sum{\left(\mathrm{\text{AUCsim}}\mathrm{\ }\mathrm{-}\mathrm{\ }\mathrm{\text{AUCtarget}}\right)\mathrm{,}}\nonumber \\ \end{equation}
where AUCsim denotes the simulated AUC results and AUCtarget denotes the optimal target AUC.
Toxicity Toxicity due to chemotherapy was scored by the physicians during all treatment cycles according to the Common Terminology Criteria for Adverse Events (CTCAE) v 4.0 criteria.30 Only grade 3 and 4 toxicities were considered in the analysis. Premature therapy termination (patient does not complete the standard 6 or 10 cycles), dose delay (patient needs more than 3 weeks to recover from chemotherapy) and dose reduction because of toxicity, were recorded. Overall toxicity was defined when; any toxicity ≥ grade 3, and/or dose delay, dose reduction and premature treatment termination occurred due to toxicity.