Abstract Tamoxifen (Tam), a standard therapy for estrogen receptor (ER)-positive breast cancer, is excessively metabolized mainly by enzymes of the CYP450 family to anti-estrogenic and estrogenic compounds. Whereas in breast cancer cells the anti-estrogenic metabolites endoxifen and 4-hydroxy-Tam inhibit the ER, Tam bisphenol (Bis) and both isomers (E)- and (Z)- of metabolite E (Met E) show full estrogenic properties, with (E)-Met E being the most potent ER agonist. Contrary to known clinical effects of ER antagonists, the influence of Tam Bis, (E)-, and (Z)-Met E on treatment outcome is not thoroughly explored. Therefore we analyzed whether interindividual differences in plasma levels of estrogenic Tam metabolites interact with Tam therapy in vivo. Plasma concentrations of estrogenic metabolites Bis, (E)-, and (Z)-Met E and their metabolic ratios (MR) to Tam were quantified in 306 premenopausal breast cancer patients mainly of European origin who were treated with 20 mg/day of Tam (Eccles D, et al. BMC Cancer 2007; 7: 160) using recently published LC-MS/MS methods (Johänning et al. Anal Bioanal Chem 2015; Mürdter et al. Clin Pharmacol Ther 2011). The endpoints recurrence-free interval (RFI) and event-free survival (EFS) were analyzed in hormone-receptor positive patients (N = 296) by Kaplan-Meyer and multivariate Cox regression adjusted for age, nodal status, tumor size, grade and plasma concentrations of endoxifen (as MR of desmethyl-Tam to endoxifen). Plasma concentrations were as follows: Bis 13-832 pM (median 187 pM), (E)-Met E 15-1029 pM (median 213 pM), (Z)-Met E 128-2484 pM (median 889 pM), and Tam 47-1061 nM (median 360 nM). Patients with higher (E)-Met E to Tam ratios showed a higher risk of breast cancer recurrence when classified into quartiles: compared to patients of the lower quartile, patients had shorter RFI when they grouped to the interquartile (P = 0.037) and the upper quartile (P = 0.005). There was a trend of higher recurrence risk for patients belonging to the upper quartile compared to the interquartile (P = 0.063). These findings were confirmed in multivariate Cox analyses: patients belonging to the upper quartile had a significantly increased hazard ratio (HR) of 2.77 (95% confidence interval (CI) 1.34-5.71; P = 0.006) and patients belonging to the interquartile showed a trend for increased HR of 1.75 (CI = 0.88-3.46; P = 0.1). For EFS, a linearly increased HR was confirmed for patients with higher (E)-Met E to Tam ratios (HR = 1.44, CI = 1.06-1.97; P = 0.021). Bis, (Z)-Met E and their MRs to Tam showed no significantly altered risk of cancer recurrence or death. These findings suggest that higher formation rates of the most potent estrogenic Tam metabolite (E)-Met E influence Tam outcome in breast cancer patients and may be linked to therapeutic failure. Citation Format: Janina Johänning, Diana M. Eccles, Bryony Eccles, Michel Eichelbaum, Matthias Schwab, Hiltrud Brauch, Thomas Mürdter, Werner Schroth. A high ratio of tamoxifen metabolite E to tamoxifen is associated with an increased risk of breast cancer recurrences in premenopausal women. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2029.

Tamoxifen is the standard-of-care treatment for estrogen receptor-positive premenopausal breast cancer. We examined tamoxifen metabolism via blood metabolite concentrations and germline variations of CYP3A5, CYP2C9, CYP2C19 and CYP2D6 in 587 premenopausal patients (Asians, Middle Eastern Arabs, Caucasian-UK; median age 39 years) and clinical outcome in 306 patients. N -desmethyltamoxifen (DM-Tam)/ (Z)- endoxifen and CYP2D6 phenotype significantly correlated across ethnicities ( R 2 : 53%, P <10 −77 ). CYP2C19 and CYP2C9 correlated with norendoxifen and (Z)-4-hydroxytamoxifen concentrations, respectively ( P <0.001). DM-Tam was influenced by body mass index ( P <0.001). Improved distant relapse-free survival (DRFS) was associated with decreasing DM-Tam /(Z)- endoxifen ( P =0.036) and increasing CYP2D6 activity score (hazard ratio (HR)=0.62; 95% confidence interval (CI), 0.43–0.91; P =0.013). Low (<14 n M ) compared with high (>35 n M ) endoxifen concentrations were associated with shorter DRFS (univariate P =0.03; multivariate HR=1.94; 95% CI, 1.04–4.14; P =0.064). Our data indicate that endoxifen formation in premenopausal women depends on CYP2D6 irrespective of ethnicity. Low endoxifen concentration/formation and decreased CYP2D6 activity predict shorter DRFS.

The International Tamoxifen Pharmacogenomics Consortium was established to address the controversy regarding cytochrome P450 2D6 (CYP2D6) status and clinical outcomes in tamoxifen therapy. We performed a meta-analysis on data from 4,973 tamoxifen-treated patients (12 globally distributed sites). Using strict eligibility requirements (postmenopausal women with estrogen receptor-positive breast cancer, receiving 20 mg/day tamoxifen for 5 years, criterion 1); CYP2D6 poor metabolizer status was associated with poorer invasive disease-free survival (IDFS: hazard ratio = 1.25; 95% confidence interval = 1.06, 1.47; P = 0.009). However, CYP2D6 status was not statistically significant when tamoxifen duration, menopausal status, and annual follow-up were not specified (criterion 2, n = 2,443; P = 0.25) or when no exclusions were applied (criterion 3, n = 4,935; P = 0.38). Although CYP2D6 is a strong predictor of IDFS using strict inclusion criteria, because the results are not robust to inclusion criteria (these were not defined a priori), prospective studies are necessary to fully establish the value of CYP2D6 genotyping in tamoxifen therapy.

The therapeutic effect of tamoxifen depends on active metabolites, e. g., cytochrome P450 2D6 (CYP 2D6) mediated formation of endoxifen. To test for additional relationships, 236 breast cancer patients were genotyped for CYP2D6, CYP2C9, CYP2B6, CYP2C19, CYP3A5, UGT1A4, UGT2B7, and UGT2B15; also, plasma concentrations of tamoxifen and 22 of its metabolites, including the (E)-, (Z)-, 3-, and 4'-hydroxymetabolites as well as their glucuronides, were quantified using liquid chromatography-tandem mass spectrometry (MS). The activity levels of the metabolites were measured using an estrogen response element reporter assay; the strongest estrogen receptor inhibition was found for (Z)-endoxifen and (Z)-4-hydroxytamoxifen (inhibitory concentration 50 (IC50) 3 and 7 nmol/l, respectively). CYP2D6 genotypes explained 39 and 9% of the variability of steady-state concentrations of (Z)-endoxifen and (Z)-4-hydroxytamoxifen, respectively. Among the poor metabolizers, 93% had (Z)-endoxifen levels below IC90 values, underscoring the role of CYP 2D6 deficiency in compromised tamoxifen bioactivation. For other enzymes tested, carriers of reduced-function CYP2C9 (*2, *3) alleles had lower plasma concentrations of active metabolites (P < 0.004), pointing to the role of additional pathways.

Induction of the major drug metabolizing enzyme CYP3A4 by xenobiotics contributes to the pronounced interindividual variability of its expression and often results in clinically relevant drug-drug interactions. It is mainly mediated by PXR, which regulates CYP3A4 expression by binding to several specific elements in the 5' upstream regulatory region of the gene. Induction itself shows a marked interindividual variability, whose underlying determinants are only partly understood. In this study, we investigated the role of nuclear receptor binding to PXR response elements in CYP3A4, as a potential non-genetic mechanism contributing to interindividual variability of induction. By in vitro DNA binding experiments, we showed that several nuclear receptors bind efficiently to the proximal promoter ER6 and distal xenobiotic-responsive enhancer module DR3 motifs. TRalpha1, TRbeta1, COUP-TFI, and COUP-TFII further demonstrated dose-dependent repression of PXR-mediated CYP3A4 enhancer/promoter reporter activity in transient transfection in the presence and absence of the PXR inducer rifampin, while VDR showed this effect only in the absence of treatment. By combining functional in vitro characterization with hepatic expression analysis, we predict that TRalpha1, TRbeta1, COUP-TFI, and COUP-TFII show a strong potential for the repression of PXR-mediated activation of CYP3A4 in vivo. In summary, our results demonstrate that nuclear receptor binding to PXR response elements interferes with PXR-mediated expression and induction of CYP3A4 and thereby contributes to the interindividual variability of induction.

Abstract 2625 This icon denotes an abstract that is clinically relevant.

BACKGROUND AND OBJECTIVES:The contribution of transport in the small intestine by the apically located efflux pump P-glycoprotein to variable drug absorption in humans is still poorly understood. We therefore investigated whether inhibition of intestinal P-glycoprotein-mediated efflux by quinidine leads to increased absorption of the P-glycoprotein substrate digoxin.METHODS:Using a multilumen perfusion catheter, we investigated the impact of P-glycoprotein inhibition on absorption of two compounds: the P-glycoprotein substrate digoxin and the marker for passive transcellular absorption antipyrine. Two 20cm adjacent jejunal segments were isolated with the multilumen perfusion catheter in seven healthy subjects. Unlabelled and deuterated digoxin and antipyrine, respectively, were simultaneously infused into either of the intestinal segments. One of the segments was additionally perfused with the P-glycoprotein inhibitor quinidine. Intestinal perfusates were collected for 3 hours, and drug concentrations were determined in the intestinal perfusates, plasma and urine.RESULTS:Quinidine did not affect the disposition of antipyrine. In contrast, coadministration of quinidine into one jejunal segment caused a considerable increase in the amount of digoxin absorbed from this segment compared with the absorption from the other quinidine-free segment (22.3 +/- 8.9% vs 55.8 +/- 21.2% of the dose; p < 0.05). Accordingly, the area under the plasma concentration-time curve and the maximum plasma concentration of digoxin were considerably higher when luminal quinidine was coadministered (p < 0.05 and p < 0.001, respectively). Differences in digoxin absorption from the two intestinal segments were also reflected by pronounced differences in urinary digoxin elimination (5.5 +/- 3.3% vs 19.2 +/- 8.1% of the dose; p < 0.01).CONCLUSIONS:P-glycoprotein inhibition in enterocytes increases systemic exposure of orally administered drugs that are P-glycoprotein substrates. These data highlight the importance of the small intestine as an active barrier against xenobiotics.

Pharmacogenetics deals with inherited differences in the response to drugs. The best-recognized examples are genetic polymorphisms of drug-metabolizing enzymes, which affect about 30% of all drugs. Loss of function of thiopurine S-methyltransferase (TPMT) results in severe and life-threatening hematopoietic toxicity if patients receive standard doses of mercaptopurine and azathioprine. Gene duplication of cytochrome P4502D6 (CYP2D6), which metabolizes many antidepressants, has been identified as a mechanism of poor response in the treatment of depression. There is also a growing list of genetic polymorphisms in drug targets that have been shown to influence drug response. A major limitation that has heretofore moderated the use of pharmacogenetic testing in the clinical setting is the lack of prospective clinical trials demonstrating that such testing can improve the benefit/risk ratio of drug therapy.

Rifampicin (rifampin) is a potent inducer of cytochrome P450 (CYP) 3A4. It was recently identified as a substrate of the polymorphic organic anion transporting polypeptide 1B1 (OATP1B1) expressed on the sinusoidal membrane of human hepatocytes. The present study aimed to investigate the possible association of single nucleotide polymorphisms (SNP) in the SLCO1B1 gene encoding for OATP1B1 with the inducing effect of rifampicin on hepatic CYP3A4. A total of 38 healthy volunteers who had participated in drug interaction studies with rifampicin were genotyped for the g. - 11187G > A and c.521T > C SNPs in SLCO1B1, c.3435C > T SNP in ABCB1 and g.6986A > G SNP in CYP3A5. The plasma concentration of 4beta-hydroxycholesterol, an endogenous marker of CYP3A4 activity, was measured before and after administration of 600 mg rifampicin once daily for 9-11 days. Treatment with rifampicin significantly increased the mean +/- SD plasma concentration of 4beta-hydroxycholesterol from 55.2 +/- 17.9 ng/ml to 120.9 +/- 32.0 ng/ml (P < 0.001). A large intersubject variability existed in the induction of CYP3A4 by rifampicin, but no associations were observed between the variability in induction and any of the polymorphisms studied. These data suggest that SLCO1B1 polymorphism does not affect the extent of induction of hepatic CYP3A4 by rifampicin, probably because other uptake transporters, such as OATP1B3, can compensate for reduced uptake of rifampicin by OATP1B1. However, the present study had sufficient power to detect only a considerably smaller rifampicin-induced increase in 4beta-hydroxycholesterol in carriers of the SLCO1B1 c.521C allele compared to subjects with the reference genotype.

The H1-receptor antagonist fexofenadine is a P-glycoprotein substrate [1]. However, the effects of genetic polymorphisms of the MDR1 (ABCB1) gene, which codes for P-glycoprotein on the disposition of fexofenadine are unclear [2, 3]. In addition, fexofenadine is a substrate of the intestinal uptake transporters organic anion transporting polypeptide 1A2 (OATP1A2, OATP-A) and 2B1 (OATP2B1, OATP-B) [1, 4]. Fruit juices such as grapefruit, orange and apple juice may decrease the oral bioavailability of fexofenadine by inhibiting the activity of OATP1A2 [5]. There seems to be no published data on whether fexofenadine is a substrate of the hepatic uptake transporter OATP1B1 (also known as OATP-C and OATP2), encoded by SLCO1B1. It has been recently demonstrated that certain single-nucleotide polymorphisms (SNP) of SLCO1B1 are associated with a decreased function of OATP1B1 in vitro. For example healthy carriers of the 521T > C allele [6] have an increased systemic exposure to pravastatin [7]. The aim of the present study was to characterize possible relationships between polymorphisms in the SLCO1B1 gene and the pharmacokinetics of fexofenadine in healthy Caucasian subjects. To this end, 20 Caucasian volunteers who had previously participated in a pharmacokinetic study with fexofenadine [3] were genotyped by TaqMan® 5′nuclease assays for the two SLCO1B1 SNPs that were significantly associated with increased pravastatin concentrations in our previous study [7], 521T > C (Val174Ala) and for the promoter SNP −11187G > A. Of the 20 subjects, two were homozygous and eight were heterozygous for the SLCO1B1521T > C SNP, the remaining 10 subjects having the reference genotype (521TT). Two subjects were heterozygous for the SLCO1B1−11187G > A polymorphism, and the remaining subjects had the reference (−11187GG) genotype. Both heterozygotes for the −11187G > A SNP were also heterozygous for the 521T > C SNP. In the two subjects with the 521CC (and −11187GG) genotype, the mean total AUC of fexofenadine was 76.0% higher (P < 0.05; anova followed by the Tukey test) than in the eight subjects with the 521TC genotype and 127% higher (P < 0.01) than in the 10 subjects with the 521TT genotype (Figure 1). The SLCO1B1521T > C polymorphism had no significant influence on the Cmax or t1/2 of fexofenadine. There was a trend (P = 0.13) towards a gene-dose effect regarding the influence of the 521T > C SNP on the apparent nonrenal clearance of fexofenadine. Thus, it was lowest (389 ± 185 ml min−1) in subjects with the 521CC genotype, intermediate (775 ± 401 ml min−1) in those with the 521TC genotype and highest in those with the reference genotype (957 ± 325 ml min−1). Mean (± SEM) plasma concentrations of fexofenadine in 20 healthy subjects with different SLCO1B1 521T > C genotypes after a single 180 mg oral dose of fexofenadine HCl. SLCO1B1 521CC subjects (n = 2) (▴); SLCO1B1 521TC subjects (n = 8) (▪); SLCO1B1 521T T subjects (n = 10) (○) A stepwise multiple regression analysis of data from subjects carrying the SLCO1B1521T > C and −11187G > A SNPs and the MDR13435C > T (exon 26), 2677G > T/A (exon 21), 1236C > T (exon 12) and 61 A > G (exon 2) SNPs indicated that only the SLCO1B1521T > C SNP was an independent predictor of the total AUC of fexofenadine (P < 0.01, r2 = 0.347). The MDR1 SNPs 2677G > T and 3435C > T were assigned into haplotype pairs according to Johne et al.[8] and data from subjects with these haplotypes were put into a stepwise multiple regression analysis with the SLCO1B1 SNP data. It was found that the MDR1 haplotypes had no independent effect on the AUC of fexofenadine (haplotype 2677T/3435C was present in one subject only). The results of this retrospective study demonstrated that the pharmacokinetics of fexofenadine were influenced by the SLCO1B1 genotype. Fexofenadine has been previously shown to be a substrate of OATP1A2 [1] and OATP2B1 [4]. Tamai et al. reported that OATP1A2 was not expressed in a single sample of human small intestine as analyzed by RT-PCR [9]. However, more recently, expression of OATP1A2 in human intestine was demonstrated [10], a finding supported indirectly by the work of Dresser et al.[5]. These investigators showed that certain fruit juices at 5% of normal strength markedly inhibited OATP1A2 mediated fexofenadine uptake in vitro and that the same juices considerably decreased the oral bioavailability of fexofenadine in vivo. A nonsynonymous SNP associated with a markedly impaired capacity for mediating the cellular uptake of fexofenadine was recently identified within the coding region of the SLCO1A2 gene (encoding for OATP1A2) [10]. Expression of OATP2B1 has been shown immunohistochemically at the apical membrane of human intestinal epithelial cells [11]. Known functionally significant SLCO2B1 SNPs are rare in Caucasians [7], and SLCO2B1 genotypes were not determined in the present study. OATP1B1 is exclusively expressed in the liver [12, 13]. That the SLCO1B1−11187G > A and 521T > C SNPs are associated with a higher pravastatin AUC [7] may be explained by decreased hepatic uptake of the drug, resulting in its slower elimination and lower systemic clearance. The pharmacokinetics of pravastatin and fexofenadine are similar in that these drugs do not undergo significant CYP-mediated biotransformation and are eliminated from the body unchanged mainly by biliary and urinary excretion [14, 15]. The influence of the SLCO1B1521T > C polymorphism on the AUC of fexofenadine suggests that the OATP1B1-mediated uptake of the drug into the liver is a prerequisite for its biliary elimination. This hypothesis is also supported by the trend towards a lower apparent nonrenal clearance of fexofenadine in subjects with the 521CC or 521TC genotype compared with those with the reference genotype. Studies on the effect of the 3435C > T SNP in the exon 26 of MDR1 on intestinal P-glycoprotein expression and on the disposition of P-glycoprotein substrates, including fexofenadine, have yielded conflicting results [16]. These might be partly explained by the influence of the SLCO1B1 genotype on fexofenadine disposition [2, 3]. In conclusion, we have demonstrated that the pharmacokinetics of fexofenadine are associated with a polymorphism of the SLCO1B1 gene (encoding OATP1B1). Further studies are required to characterize the interaction between fexofenadine and the hepatic uptake transporter OATP1B1. This study was supported by grants from the Robert Bosch Foundation (Stuttgart, Germany) and the Federal Ministry of Education and Research (Bonn, Germany). M. Niemi was supported by a fellowship from the Alexander von Humboldt Foundation (Bonn).

Die Pharmakogenetik untersucht, inwieweit genetische Polymorphismen der Proteine, die die pharmakokinetischen und pharmakodynamischen Prozesse eines Arzneimittels kontrollieren, für die interindividuellen Unterschiede in Wirkung und Nebenwirkung verantwortlich sind. Im Gegensatz zur Pharmakogenetik nutzt die Pharmakogenomik einen genomweiten Ansatz zur Identifizierung von Genen bzw. Gennetzwerken, die an der Entstehung von Erkrankungen beteiligt sind bzw. als therapeutische Ziele für neue Arzneistoffe dienen können. Die zurzeit am besten charakterisierten pharmakogenetischen Polymorphismen betreffen die Arzneimittel metabolisierenden Enzyme Zytochrom-P450-2C9, -2C19 und -2D6 bzw. die Thiopurinmethyltransferase, für die in klinischen Studien relevante Konsequenzen für die Arzneimitteltherapie gezeigt werden konnten. Das ultimative Ziel pharmakogenetischer/-genomischer Forschung ist es, unter Verwendung einer neuen Krankheits- und Therapieklassifikation auf molekularer Ebene eine spezifische Arzneimitteltherapie bei genetisch definierten Untergruppen von Patienten durchzuführen.

ObjectivesOur objectives were to determine the content of cytochrome P450 (CYP) 3A4, CYP3A5, and P‐glycoprotein and to measure CYP3A4‐dependent catalytic activity in paired human small intestinal and liver specimens.MethodsSamples of duodenum or proximal jejunum and liver wedge biopsy specimens were obtained from 15 patients undergoing a gastrointestinal operation. Enterocytes were isolated from the intestinal samples. The contents of CYP3A4, CYP3A5, and P‐glycoprotein and CYP3A4‐mediated catalytic activities were determined in homogenized enterocyte and liver samples.ResultsThe CYP3A4 protein content was about 3 times (P < .01) and the P‐glycoprotein content about 7 times (P < .0001) higher in the enterocyte homogenates than in the liver homogenates. CYP3A5 protein was detected in all samples, but the levels were too low in most cases to allow quantification. The 2 cases with a quantifiable hepatic CYP3A5 content had the CYP3A5*1/*3 genotype; all other cases were homozygous for the CYP3A5*3 allele. No intraindividual correlations between the intestine and liver with respect to CYP3A4 content, P‐glycoprotein content, or the measured catalytic activities were present. Values for the maximum rate of metabolism (Vmax) of verapamil N‐dealkylation (formation of D‐617) and N‐demethylation (formation of norverapamil) activities correlated with the CYP3A4 protein content in both organs.ConclusionsThis work demonstrated a much higher content of both CYP3A4 protein and P‐glycoprotein in enterocytes isolated from human duodenal or jejunal mucosa than in paired specimens of liver tissue. These results lend support to the view that biotransformation in the gut wall substantially contributes to the overall first‐pass metabolism of many CYP3A4 substrates. Furthermore, the high content of P‐glycoprotein on the apical surface of enterocytes supports the theory that this efflux transporter may act in concert with CYP3A4 to limit oral drug bioavailability. Finally, these results indicate that neither CYP3A4 nor MDR1 (P‐glycoprotein) is coordinately regulated in the liver and intestine.Clinical Pharmacology & Therapeutics (2004) 75, 172–183; doi: 10.1016/j.clpt.2003.10.008

Transporter proteins, in particular P-glycoprotein (Pgp), are important determinants in absorption, tissue targeting, and elimination of drugs. In addition to physiological and environmental factors, its expression and function are modified by genetic polymorphisms of the MDR1 gene. So far, several MDR1 SNPs have been identified, and mutations at positions 2677 and 3435 were associated with alteration of Pgp expression and/or function. In contrast to drug-metabolizing enzymes (eg, CYP2D6), for which loss of function mutations or gene amplification manifests as distinct phenotypes in the population, the impact of MDR1 polymorphisms on pharmacokinetics and pharmacodynamics of Pgp substrates is moderate. Clinical studies on the effects of the C3435T polymorphism and drug treatment with cardiac glycosides, the immunosuppressants cyclosporine and tacrolimus, HIV protease inhibitors, and tricyclic antidepressants are discussed.

Thiopurine S-methyltransferase (TPMT) is a cytosolic enzyme that catalyzes the S-methylation of thiopurines. The TPMT locus is subject to genetic polymorphism, with heterozygous individuals (about 10% of Caucasians) having intermediate, and homozygous individuals (about 0.33% of Caucasians) having low TPMT activity. More than 95% of defective TPMT activity can be explained by the most frequent variant alleles TPMT*2, TPMT*3A, TPMT*3B, TPMT*3C and TPMT*3D. In the present study, we analyzed the association TPMT of genotype with minimal residual disease (MRD), before and after application of a four week cycle of 6-mercaptopurine (6-MP; 60 mg/m2/d) during induction consolidation treatment. Samples were derived from patients enrolled in our ongoing Berlin-Frankfurt-Muenster (BFM) trial on treatment of childhood ALL (ALL-BFM 2000) from 10/1999 to 03/2002. Routine bone-marrow samples were taken at diagnosis and after completion of induction (treatment day 33) and induction consolidation (treatment day 78). Semiquantitative estimation of MRD was performed by standardized PCR analysis of leukemia clone-specific immunoglobulin and T-cell receptor gene rearrangements and TAL1 deletions. In order to be included in the study presented here, patients had to have at least one MRD marker with a sensitivity of 10−4 and successful MRD monitoring at both follow-up time points (day 33 and day 78); MRD loads smaller than 10−4 were defined as negative. Genotyping was performed by a DHPLC method. We were able to analyze 814 patients with childhood ALL. 755 (92.8%) patients were homozygous for the TPMT wild-type allele, 55 (6.8%) were heterozygous, and 4 (0.5%) homozygous for a variant TPMT allele. Except for immunophenotype, no major differences with regard to characteristics known to be associated with treatment response were observed between patients homozygous for the wild-type allele and those being heterozygous. The four patients homozygous for a variant TPMT allele were already MRD negative after completion of remission induction on treatment day 33. Therefore, they were not included in further analyses. In patients homozygous for the wild-type allele or heterozygous, MRD levels on day 33 were equally distributed between the groups. However, when MRD levels were assessed on day 78, after administration of induction consolidation treatment, including a 4-week cycle of 6-MP, significant differences with regard to clearance of MRD were observed between patients homozygous for the TPMT wild-type allele and heterozygotes. For heterozygous patients, this distribution translated into a 2.4-fold reduction in risk of having measurable MRD after induction consolidation treatment (relative risk = 0.42; 95% confidence interval = 0.16–0.99; P = 0.047). This effect remained significant in multivariate analysis including immunophenotype (relative risk = 0.36; 95% confidence interval = 0.15–0.87; P = 0.023). Our results indicate that TPMT genotype has a substantial influence on MRD after administration of 6-MP during induction consolidation treatment of childhood ALL, most likely through modulation of 6-MP dose intensity. In addition, our data support a role for MRD in the assessment of treatment response to specific drugs.

BACKGROUND: The constitutive androstane receptor (CAR) plays a key role in the control of drug metabolism and transport by mediating the phenobarbital-type induction of many phase I and II drug metabolizing enzymes and drug transporters. RESULTS: We identified transcripts generated by four different alternative splicing events in the human CAR gene. Two of the corresponding ligand binding domain isoforms demonstrated novel functional properties: First, CAR(SV3), which is encoded by a transcript containing an lengthened exon 7, differentially transactivated target gene promoters. Second, CAR(SV2), which results from the use of an alternative 3' splice site lengthening exon 8, showed ligand-dependent instead of constitutive interaction with coactivators. Furthermore, alternatively spliced transcripts demonstrated a tissue-specific expression pattern. In most tissues, only transcripts generated by alternative splicing within exon 9 were expressed. The encoded variant demonstrated a loss-of-function phenotype. Correct splicing of exon 8 to exon 9 is restricted to only a few tissues, among them liver and small intestine for which CAR function has been demonstrated, and is associated with the induction of CAR expression during differentiation of intestinal cells. CONCLUSION: Due to their specific activities, CAR variant proteins SV2 and SV3 may modulate the activity of reference CAR(SV1). Furthermore, we propose that transcriptional activation and regulation of splicing of exon 9 may be coupled to ensure appropriate tissue- and differentiation state-specific expression of transcripts encoding functional CAR protein. Altogether, alternative splicing seems to be of utmost importance for the regulation of CAR expression and function.