*Clinic of Infectious and Topical Diseases, Department of Medicine, Surgery, and Dentistry; †Biochemistry Laboratory, Department of Medicine, Surgery, and Dentistry; ‡Infectious Diseases Molecular Diagnostics, Department of Medicine, Surgery, and Dentistry, San Paolo University Hospital, Milan, Italy
Supported by grant from Fondo Interno Ricerca Scientifica e Tecnologica 2008-Università degli Studi di Milano, and from Istituto Superiore di Sanità, “National research program on AIDS,” Italy.
To the Editors:
Atazanavir (ATV), a protease inhibitor of HIV, acts as an inhibitor of the UDP-glucuronosyltransferase (UGT)1A1, an enzyme responsible for bilirubin metabolism, leading to unconjugated hyperbilirubinemia.1 The frequency and the severity of hyperbilirubinemia correlates with the UGT1A1 gene polymorphisms, mimicking the mechanism underlying Gilbert Syndrome.2,3 Furthermore, bilirubin levels directly correlate with ATV plasma concentrations.3,4 To our knowledge, the interaction between ATV plasma concentrations and the UGT1A1 polymorphism has never been formally tested, assuming their effects to be additive. The aim of this study is to assess whether the impact of ATV plasma concentrations on serum bilirubin was different according to the presence of UGT1A1-TA7 allele.
Patients cared at our Institute treated with ATV 300 mg plus ritonavir 100 mg (ATV/r) and 2 nucleoside analogues for longer than 3 months were identified. Plasma ATV concentration was measured after 24 ± 4 hours drug intake [trough concentration (Ctrough)] by a validated high-performance liquid chromatography method.5 Genetic profile of the UGT1A1 gene was assessed by direct sequencing of DNA extracted from peripheral blood mononuclear cell.6
Descriptive results of continuous variables were expressed as median and interquartile range (IQR) values. Univariate and multivariate linear regression models were used to investigate factors correlated with total serum bilirubin. Sixteen of 55 patients (29.1%) were female, the median age was 43 years (IQR 39-49), and the body mass index (BMI) was 23.7 (IQR 21.8-25.5). The CD4 cell count was 468 cells per microliter (IQR 336-588). The median follow-up on ATV was 14.5 months (IQR 5-32). The median serum bilirubin was 2.53 mg/dL (IQR 1.31-3.94), and the proportion of patients with grade 3 (>2.5 × Upper Normal Limit [ULN]) and grade 4 (>5 × ULN) hyperbilirubinemia was 50.1% and 12.7%, respectively. The median ATV Ctrough was 1 mg/L (IQR 1-3). The distribution of UGT1A1 genotypes was as follows: TA6/TA6 in 43.6%, TA7/TA6 n 40%, TA7/TA7 in 16.4%. The median bilirubin level in each genotype was: 2.1 mg/dL (IQR 1.2-2.8) in TA6/TA6, 2.58 mg/dL (IQR 1.3-3.9) in TA6/TA7, 5.2 mg/dL (IQR 3.0-7.6) in TA7/TA7 (P = 0.05). There were no significant differences between the 3 genotypes in terms of age, gender, BMI, CD4 count, and ATV plasma level. When total serum bilirubin was regressed on ATV plasma concentration and UGT1A1 polymorphism, the expected difference in total bilirubin resulted: 1.49 (SE 0.38; P = 0.0003) per unit difference in ATV Ctrough, and increased of 1.39 factor (SE 0.51; P = 0.0009) if at least 1 UGT1A1-TA7 allele was present (rho2 = 0.26). To test whether the difference in total bilirubin per unit difference in ATV plasma concentrations changed according to the presence of the UGT1A1-TA7 allele, a polymorphism-by-ATV Ctrough interaction term was included in the model. The increase in total bilirubin per unit difference in ATV Ctrough resulted 0.80 (SE 0.39) in patients with TA6/TA6 genotype and 2.33 (SE 0.64) in patients carrying at least one TA7 allele (rho2 = 0.29; P for interaction = 0.04). The model did not change after adjusting for age, gender, CD4 count, and BMI. In addition, older age resulted independently correlated serum bilirubin level (Table 1).
The above results confirm a direct correlation between ATV plasma level and bilirubinemia that was influenced by the presence of homozygosis or heterozygosis for UGT1A1-TA7 allele. Other authors concluded that severe hyperbilirubinemia was further increased by the presence of the UGT1A1-TA7 allele,3 however, to the best of our knowledge, this is the first study that formally tested the interaction between these 2 major predictors of hyperbilirubinemia in patients treated with ATV. In particular, carrying UGT1A1- TA7 polymorphism accounted for a 2-fold increase in bilirubin level for a unit change in ATV plasma concentration (Table 1). In keeping with literature findings,5,7 severe hyperbilirubinemia and hepatic toxicity were rare in our study population; only 2 patients showed increased (>2.5 times ULN) liver transaminase leading to interpret the herein observed hyperbilirubinemia as an innocent phenomenon (data not shown). Indeed, bilirubinemia and/or transaminase increases may not be frequently detected in patients on stable therapy, given that premature drug discontinuation occurs in subjects nontolerating the compound. The key question is: would UGT1A1 testing be cost-effective in the presence of an apparently innocent side effect?
The occurrence of jaundice may be considered undesirable by some patients, thus limiting adherence to ATV and increasing the probability of treatment failure. Furthermore, studies linked the reduced glucuronidation activity in the presence of the TA7 allele, to cancer disposition8 and severe unwanted drug reactions.9 ATV, acting as an inhibitor of UGT proteins, might enhance these effects particularly in individuals who already suffer from reduced UGT activity. Findings that better characterize the interaction between ATV plasma concentration and serum bilirubin may help in dose adjustments under the guidance of therapeutic drug monitoring. In general, haplotype identification as pharmacogenomic risk factors might improve drug safety and establish individualized pharmacotherapy.
Paola Cicconi, MD, PhD*
Teresa Bini, MD*
Alessandra Barassi, MD, PhD†
Maddalena Casana, MD*
Olivia Turri, biologist‡
Francesca Pateri, biologist†
Giulia C. Marchetti, MD, PhD*
Maria Luisa Biondi, MD, PhD‡
Gianlodovico Melzi d'Eril, MD, PhD†
Antonella d'Arminio Manforte, MD, PhD*
*Clinic of Infectious and Topical Diseases, Department of Medicine, Surgery, and Dentistry
†Biochemistry Laboratory, Department of Medicine, Surgery, and Dentistry
‡Infectious Diseases Molecular Diagnostics, Department of Medicine, Surgery, and Dentistry, San Paolo University Hospital, Milan, Italy
1. Lankisch TO, Moebius U, Wehmeier M, et al. Gilbert's disease and atazanavir: from phenotype to UDP-glucoronosyltransferase halotype. Hepatology
2. Strassburg CP, Lankisch TO, Manns MP, et al. Family 1 uridine-5'-diphosphate glucuronosyltransferases (UGT1A): from Gilbert's syndrome to genetic organization and variability. Arch Toxicol
3. Rodríguez-Nóvoa S, Martín-Carbonero L, Barreiro P, et al. Genetic factors influencing atazanavir plasma concentrations and the risk of severe hyperbilirubinemia. AIDS
4. Cleijsen RM, Van de Ende ME, Kroon FP, et al. Therapeutic drug monitoring of the HIV protease inhibitor atazanavir in clinical practice. J Antimicrob Chemother
5. Guaraldi G, Cocchi S, Motta A, et al. A pilot study on the efficacy, pharmacokinetics and safety of atazanavir in patients with end-stage liver disease. J Antimicrob Chemother
6. Nauck M, Stein U, Von Karger S, et al. Rapid detection of the C3435T polymorphism of multidrug resistance gene 1 using fluorogenic hybridization probes. Clin Chem
7. Torti C, Lapadula G, Antinori A, et al. Hyperbilirubinemia during atazanavir treatment in 2,404 patients in the Italian atazanavir expanded access program and MASTER Cohorts. Infection
8. Girard H, Thibaudeau J, Court MH, et al. UGT1A1 polymorfisms are important determinants of dietary carcinogen detoxification in the liver. Hepatology
9. Ando M, Hasegawa Y, Ando Y. Pharmacogenetics of irinotecan: a promoter polymorphism of UGT1A1 gene and severe adverse reactions to irinotecan. Invest New Drugs