Leo M. L. Stolk

Dried Blood Spot Methods in Therapeutic Drug Monitoring: Methods, Assays, and Pitfalls

This article reviews dried blood spot (DBS) sampling in therapeutic drug monitoring. The DBS method involves applying whole blood obtained via a fingerprick to a sampling paper. After drying and transportation, the blood spot is extracted and analyzed in the laboratory. Assays of many medicines in DBS have already been reported in the literature and are reviewed here. The technique involved in and factors that may influence the accuracy and reproducibility of DBS methods are also discussed. DBS sampling ultimately seems to be a useful technique for therapeutic drug monitoring that could have many advantages in comparison with conventional venous sampling. However, its benefits must be weighed against the degree of potential errors introduced via the sampling method; there is evidently a need for more standardization, quality assurance, basic research, and assay development.

Therapeutic drug monitoring of everolimus using the dried blood spot method in combination with liquid chromatography–mass spectrometry

An assay of everolimus based on finger prick sampling and consecutive application as a blood spot on sampling paper has been developed. We explored several methods [K. Hoogtanders, J. van der Heijden, M. Christiaans, P. Edelbroek, J. van Hooff, L. Stolk, J. Pharm. Biomed. Anal. 44 (2006) 658-664; A. Allanson, M. Cotton, J. Tettey, et al., J. Pharm. Biomed. Anal. 44 (2007) 963-969] and developed a new method, namely the impregnation of sampling paper with a solution of plasma-protein, formic acid and ammonium acetate, in combination with the extraction of the blood spot by filter filtration. This kind of sample preparation provides new possibilities for blood spot sampling especially if analytes are adsorbed to the paper. The dried blood spot was analysed using the HPLC-electrospray-tandem mass spectrometry method, with 32-desmethoxyrapamycin as the internal standard. The working range of our study was 2-30 microg/l. Within this range, intra-and inter-assay variability for precision and accuracy was <15%. Everolimus blood spot samples proved stable for 3 days at 60 degrees C and for 32 days at 4 degrees C. Everolimus concentrations of one stable out-patient were compared after both blood spot sampling and conventional venous sampling on various occasions. Results indicate that this new method is promising for therapeutic drug monitoring in stable renal transplant patients.

Tacrolimus pharmacokinetics and pharmacogenetics: influence of adenosine triphosphate‐binding cassette B1 (ABCB1) and cytochrome (CYP) 3A polymorphisms

Tacrolimus, an immunosuppressant used after organ transplantation, has a narrow therapeutic range and its pharmacokinetic variability complicates its daily dose assessment. P‐glycoprotein (P‐gp), encoded by the adenosine triphosphate‐binding cassette B1 (ABCB1) and the cytochrome (CYP) 3A4 and 3A5 enzymes appears to play a role in the tacrolimus metabolism. In the present study, two different renal transplant recipient groups were used to examine the influence of ABCB1 and CYP3A polymorphisms on the daily tacrolimus dose and several pharmacokinetic parameters. In total 63 Caucasian renal transplant recipients divided into 26 early [median (range) of the days since transplantation – 16 (3–74)] and 37 late [median (range) of the days since transplantation – 1465 (453–4128)] post‐transplant recipients were genotyped for ABCB1 and CYP3A polymorphisms. The pharmacokinetic parameters of tacrolimus were determined for all renal transplant recipients and correlated with their corresponding genotypes. A significant difference in allele frequencies of the CYP3A4*1B (P = 0.028) and CYP3A5*1 (P = 0.022) alleles was observed between the early and late post‐transplant recipient groups. Significantly higher dose‐normalized trough levels (dnC0), dose‐normalized area under the curve (dnAUC0−12), and dose‐normalized maximum concentration (dnCmax) were observed for carriers of the CYP3A5*3 variant allele in both renal transplant patient groups. Except for the daily tacrolimus dose (P = 0.025) no significant differences were observed for carriers of the CYP3A4*1B variant allele. Neither the individual ABCB1 polymorphisms nor the ABCB1 haplotypes were associated with any pharmacokinetic parameter. We noticed that patients carrying a CYP3A5*1 allele require a twofold higher tacrolimus dose compared with homozygous carriers of the CYP3A5*3 variant allele to maintain the target dnAUC0−12. Therefore, genotyping for the CYP3A5*3 variant allele can contribute to a better and more individualized immunosuppressive therapy in transplant patients.

Dried blood spot measurement: application in tacrolimus monitoring using limited sampling strategy and abbreviated AUC estimation

Dried blood spot (DBS) sampling and high‐performance liquid chromatography tandem–mass spectrometry have been developed in monitoring tacrolimus levels. Our center favors the use of limited sampling strategy and abbreviated formula to estimate the area under concentration–time curve (AUC0–12). However, it is inconvenient for patients because they have to wait in the center for blood sampling. We investigated the application of DBS method in tacrolimus level monitoring using limited sampling strategy and abbreviated AUC estimation approach. Duplicate venous samples were obtained at each time point (C0, C2, and C4). To determine the stability of blood samples, one venous sample was sent to our laboratory immediately. The other duplicate venous samples, together with simultaneous fingerprick blood samples, were sent to the University of Maastricht in the Netherlands. Thirty six patients were recruited and 108 sets of blood samples were collected. There was a highly significant relationship between AUC0–12, estimated from venous blood samples, and fingerprick blood samples (r2 = 0.96, P < 0.0001). Moreover, there was an excellent correlation between whole blood venous tacrolimus levels in the two centers (r2 = 0.97; P < 0.0001). The blood samples were stable after long‐distance transport. DBS sampling can be used in centers using limited sampling and abbreviated AUC0–12 strategy as drug monitoring.

Evaluation of limited sampling strategies for tacrolimus

ObjectiveIn literature, a great diversity of limited sampling strategies (LSS) have been recommended for tacrolimus monitoring, however proper validation of these strategies to accurately predict the area under the time concentration curve (AUC0–12) is limited. The aim of this study was to determine whether these LSS might be useful for AUC prediction of other patient populations.MethodsThe LSS from literature studied were based on regression equations or on Bayesian fitting using MWPHARM 3.50 (Mediware, Groningen, the Netherlands). The performance was evaluated on 24 of these LSS in our population of 37 renal transplant patients with known AUCs. The results were also compared with the predictability of the regression equation based on the trough concentrations C0 and C12 of these 37 patients. Criterion was an absolute prediction error (APE) that differed less than 15% from the complete AUC0–12 calculated by the trapezoidal rule.ResultsThirteen of the 18 (72%) LSS based on regression analysis were capable of predicting at least 90% of the 37 individual AUC0–12 within an APE of 15%. Additionally, all but three LSS examined gave a better prediction of the complete AUC0–12 in comparison with the trough concentrations C0 or C12 (mean 62%). All six LSS based on Bayesian fitting predicted <90% of the 37 complete AUC0–12 correctly (mean 67%).ConclusionsThe present study indicated that implementation of LSS based on regression analysis could produce satisfactory predictions although careful evaluation is necessary.

Population pharmacokinetics and dosing of flucloxacillin in preterm and term neonates.

In total 235 flucloxacillin total (free+protein bound) plasma concentrations were determined in 55 neonates (gestational age 26 to 42 weeks, postnatal age 0 to 44 days) with reversed-phase HPLC in surplus plasma samples from routine gentamicin assays. Population pharmacokinetic parameters were calculated according to an one compartment open model with iterative two-stage Bayesian fitting (MWPHARM 3.50, Mediware, The Netherlands). Mean clearance corrected for weight was 0.18±0.10 L kg−1h−1 and volume of distribution corrected for weight was 0.54±0.17 L/kg. Pearson correlations between the individual pharmacokinetic parameters and covariates, like gestational age, plasma creatinine, and gentamicin clearance, were low and therefore not relevant for use in clinical practice. Total plasma concentrations above 200 mg/L were considered toxic and T>MIC (time above minimum inhibitory free plasma concentration) of more than 40% was considered effective. Protein binding was assumed to be 86.3% in all neonates, based on literature. The current dosage regimen, 25 or 50 mg/kg every 8 or 12 hours, did not result in effective plasma concentrations for the treatment of Staphylococcus aureus in 17 (31%) of the 55 neonates. Therefore, the authors suggest an initial dose of 25 mg/kg/4 h for all neonates, irrespective of their age, based on the breakpoint MIC value of flucloxacillin for Staphylococcus aureus (2.0 mg/L). After isolation of the causative agent of infection, flucloxacillin administration ought to be reconsidered based on the expected susceptibility pattern of the isolate. When oxacillin sensitive coagulase negative staphylococci are isolated, the initial dose should be reduced to 10 mg/kg/6 h, based on the breakpoint MIC value of 0.25 mg/L. Simulation with these new dosage regimens indicated that satisfactory plasma concentrations were reached in 52 of the 55 neonates. However, the regimens need prospective verification. Moreover, the exact role of neonatal protein binding needs to be further investigated.

A pilot study on sublingual administration of tacrolimus

The immunosuppressant tacrolimus has a narrow therapeutic index and a high variability of its pharmacokinetics, e.g. oral bioavailability of tacrolimus is very variable [mean 25% (4–93%)]. Reams et al. showed that sublingual administration of tacrolimus results in therapeutic blood levels in patients after lung transplantation [1,2]. We questioned whether the results of this study were not biased by absorption of dissolved tacrolimus in swallowed saliva in the gut. Therefore we conducted a pilot study to confirm their findings in renal transplant candidates with rigid measures to avoid enteral resorption. Three renal transplant candidates, a patient with cystic fibrosis and a healthy volunteer had been treated sublingually with 0.04 mg/kg tacrolimus. The day thereafter, the renal transplant candidates received tacrolimus 0.1 mg/kg orally. Following the protocol of Reams et al., commercially available tacrolimus capsules were opened and the content was dispersed under the tongue. In contrast to the study of Reams et al., the person was explicitly instructed not to swallow during 15 min and afterwards to spit out saliva and rinse the mouth with water. The study was approved by the Medical Ethical Board. Blood levels were taken at t = 0, 1⁄4, 1⁄2, 1, 2, 4, 8, 12 and 24 h postdose and were analyzed by high-performance liquid chromatography and tandem mass spectrometry (LOQ = 1 lg/l; LD = 0.05 lg/l). In contrast to the study of Reams et al., [1,2] very low tacrolimus blood levels were measured after sublingual administration of tacrolimus. We have discussed these findings with Reams and suggested that in their study, enteral absorption could not be excluded after sublingual administration, because patients were not instructed not to swallow and also to rinse their mouth. Also, the patient with cystic fibrosis and the healthy volunteer in our study did not show sublingual absorption. A recent case report [3] of a kidney transplant patient treated sublingually with tacrolimus, showed similar blood levels after sublingual administration as after oral administration. However, the observed concentrationtime profile was compatible with absorption in the digestive tract instead of the sublingual mucosa. Goorhuis et al. [4] compared tacrolimus trough levels after buccal administration to pediatric liver transplant patients in the first week after transplantation to trough levels after administration by a nasogastric tube. Again, trough levels after both routes of administration were comparable but enteral absorption could not be excluded (Fig. 1). We conclude that sublingual administration in this way does not seem to be a suitable alternative for oral administration. In contrast to earlier studies, very low blood levels were measured after sublingual administration. In our opinion, enteral absorption after sublingual administration of tacrolimus in the earlier studies could explain the good absorption after sublingual administration.

Protein binding of flucloxacillin in neonates.

The isoxazolyl penicillins, including flucloxacillin, have the highest levels of plasma protein binding among the semisynthetic penicillins. Because only the free fraction of the penicillin is pharmacologically active, it would be useful to measure both protein-bound and free flucloxacillin to determine its protein binding. Until now, flucloxacillin protein binding in newborn infants has been investigated in only two studies with relatively small populations. In the present study, flucloxacillin protein binding was investigated in 56 (preterm) infants aged 3 to 87 days (gestational age, 25-41 weeks). Surplus plasma samples from routine gentamicin assays of each infant were collected and combined to obtain a sufficiently large sample for analysis. Free flucloxacillin was separated from protein-bound flucloxacillin using ultrafiltration. Reversed-phase high-performance liquid chromatography with ultraviolet detection was used to measure free flucloxacillin concentrations in ultrafiltrate and total flucloxacillin concentrations in pooled plasma. Flucloxacillin protein binding was 74.5% ± 13.1% (mean ± standard deviation) with a high variability among the infants (34.3% to 89.7%). High Pearson correlations were found between protein binding and the covariates-plasma albumin concentration (r = 0.804, P < 0.001, n = 18) and plasma creatinine concentration (r = −0.601, P < 0.001, n = 45). Statistically significant but less striking correlations were found between protein binding and gestational age, postconceptional age, body weight, and triglyceride concentration. Because of the high variability of protein binding among infants, it is difficult to devise a flucloxacillin dosage regimen effective for all infants. Individualized dosing, based on free flucloxacillin concentrations, might help to optimize treatment of late-onset neonatal sepsis, but practical obstacles will probably prevent analysis of free flucloxacillin concentrations in newborn infants on a routine basis.

Pharmacokinetics of intravenous rifampicin (rifampin) in neonates.

Few reports have addressed neonatal rifampicin plasma concentrations and data on neonatal rifampicin pharmacokinetics are completely lacking. Therefore, plasma concentrations of rifampicin and its main metabolite 25-O-desacetylrifampicin (DES) were measured in 123 surplus plasma samples from routine vancomycin monitoring in 21 neonates using reversed-phase HPLC. Rifampicin peak and trough plasma concentrations were 4.66 ± 1.47 mg/L and 0.21 ± 0.20 mg/L, respectively, after a dose of 8.5 ± 2.1 (mean ± SD) mg/kg per day. A significant linear relationship between rifampicin dose and peak plasma concentrations was found, but inter-patient variability was high. Pharmacokinetic parameters of rifampicin were calculated according to a one-compartment open model with iterative two-stage Bayesian fitting (MW\PHARM 3.60, Mediware, The Netherlands). First-order elimination constant, volume of distribution corrected for weight, total body clearance corrected for weight (CL/W), and elimination half-life were 0.16 ± 0.06 h−1, 1.84 ± 0.59 L/kg, 0.28 ± 0.11 Lkg−1h−1, and 4.9 ± 1.7 h, respectively. A high Pearson correlation was found between CL/W rifampicin and the covariates plasma creatinine and CL/W gentamicin of a preceding gentamicin treatment course, r = 0.728 (n = 17) and r = 0.837 (n = 12), respectively. DES was detected in each plasma sample. Therefore, rifampicin seems to be eliminated by both renal and metabolic pathways in neonates. In 8 study patients, plasma concentrations of rifampicin and DES were measured again after two weeks of therapy. CL/W rifampicin was significantly higher (67 ± 50%). The authors suggest maintaining the current dose regimen of 10 mg/kg once a day. Because of the large inter-patient variability in rifampicin plasma concentrations and CL/W increase during therapy, the authors suggest monitoring rifampicin peak and trough plasma concentrations to avoid low plasma concentrations. More research is needed to determine well-founded dosing guidelines.

Therapeutic Drug Monitoring in Clinical Research

The development of a new drug is characterized by distinct developmental stages, usually described as phases I to IV. Dose tolerance and dose response exploration studies are undertaken in phase II or III. Pharmacokinetic studies are often involved in these phases, but frequently only as an objective of minor importance. The usefulness of therapeutic drug monitoring (TDM) is not consequently investigated for new drugs. Usually the need for TDM is only discovered much later, when the drug is already on the market. TDM is particularly valuable under the following circumstances: (i) if there is a stronger relationship between the drug concentration and effect than between the dose and effect; (ii) if there is no simple and clear clinical parameter available to evaluate the clinical efficacy of the drug; (iii) if the therapeutic window is small; (iv) to document interactions; (v) to monitor drug compliance; and (vi) if there is large intra- and interindividual variability and unpredictability in pharmacokinetic parameters. Our recommendation is that randomized concentration controlled trials should be performed during the early stages of drug development and that it should be obligatory for drug licensing.