A limited sampling procedure for estimating adriamycin pharmacokinetics in cancer patients

Abstract

(Iliadis et al., 1985; Favre et al., 1987). Moreover, sampling points are generally chosen arbitrarily or sometimes by using more sophisticated approaches like sensitivity functions or stepwise linear regression (Ratain & Vogelzang, 1987). But these approaches are not optimal, since they do not provide maximal precision in model parameters.Adriamycin (ADM) is a broad-spectrum antineoplastic drug which is active against a wide variety of tumours.Experimental studies have suggested that ADM concentration-time product (area under the curve, AUC) is a determinant factor for tumour cell lethality (Eichholtz-Wirth, 1980; Ritch et al., 1982 Included were 42% of patients with primary tumour only, 10% with primary and distal metastases, 38% with multifocal metastases and 10% with isolated metastases (liver and bone). There was no concomitant irradiation. Cumulative dose of ADM per patient was (mean, range) 372mg, 100-600 mg. Five patients had previously been treated by a chemotherapy regimen including anthracyclines and the total cumulative dose, including weekly ADM, was (mean, range) 640mg, 160-1,011mg; two other patients had received a chemotherapy protocol without anthracyclines. All patients had normal pretreatment serum bilirubin (5-17jmoll-1). In all cases, ADM was given by a 5-min infusion through a venous catheter. A complete pharmacokinetic study concerning ADM given on a weekly schedule on the same patients has been published (Frenay et al., 1988).Analttical method A complete pharmacokinetic profile was obtained for the first injection in 25 patients and once a month in 10 of them, i.e. in most cases at the 5th, 9th and 13th injections. Blood samples were drawn 5, 20, 40mmn and 1, 2. 4, 8 and 24h after the start of injection. As all patients were treated on an outpatient basis, it was not possible to get blood samples for more than the 24 h. From these patients and their cycles, 41 concentration-time curves were obtained. Blood samples were collected in EDTA tubes and immediately centrifuged.Plasma was stored at -20-C until analysis (within 2 weeks). ADM was quantified by HPLC. Extraction was performed on Sep-Pak C18 cartnrdges (Millipore, Waters) as previously described (Robert. 1980) with slight modifications: cartnrdges were conditioned by successive elutions with 2ml methanol and 5 ml phosphate buffer (Na2HPO4, 0.05 M; NaH2P04, 0.05 M; 2:1). One ml of plasma (patient or plasma for standard curve) spiked with 50 pi of internal standard (daunorubicin, 1 nmol ml -1) was passed through the cartridge, followed by I ml of phosphate buffer (discarded). Drug material was eluted by 3 ml of methanol in previously Br. J. Cancer O 989). 60, [89][90][91][92]

Application of pharmacokinetic modelling to the routine therapeutic drug monitoring of anticancer drugs

“…Bayesian estimators were developed for a large number of anticancer drugs: methotrexate [16,[43][44][45][46], carboplatin [47][48][49][50], cisplatin [51], etoposide [52,53], teniposide [54], anthracyclines (doxorubicin and piraru- bicin) [55][56][57], irinotecan [58,59], topotecan [28,60], vincristine [61], vinorelbine [62,63] and suramin [64][65][66]. Generally two or three blood samples were enough to predict individual pharmacokinetic parameter values with low bias and good precision.…”

Population pharmacokinetics of doxorubicin, etoposide and ifosfamide in small cell lung cancer patients: results of a multicentre study

“…In this first step, plasma concentrations are measured in a population of patients for whom the PK parameters (mainly CL, V d and AUC) are calculated. Developments in population pharmacokinetic software using Bayesian methodology allow the investigator to estimate PK parameters for some drugs with good precision, with limited sampling strategies [4, 5]. Inter‐individual and intraindividual course‐to‐course variability as well as the influence of clinical and biological covariates (e.g.…”

Therapeutic drug monitoring of cancer chemotherapy

“…MAP Bayesian methods have major advantages over other strategies, as they allow: The main drawback is that such an estimator is difficult to develop, but it is easy to use. Bayesian estimators were developed for a large number of anticancer drugs: methotrexate, [65][66][67][68] carboplatin, [69][70][71][72] cisplatin, 73 etoposide, 74,75 teniposide, 76 anthracyclines (doxorubicin and pirarubicin), [77][78][79] irinotecan, 80,81 topotecan, 43,82 vincristine, 83 vinorelbine, 84,85 and suramin. 86,87 Generally two or three blood samples were enough to predict individual pharmacokinetic parameter values with low bias and good precision.…”