HIv Articles  
Increase in serum bilirubin in HIV/hepatitis-C virus-coinfected patients on atazanavir therapy following initiation of pegylated-interferon and ribavirin
  AIDS:Volume 22(18)30 November 2008p 2535-2537
[Research Letters]
Rodriguez-Novoa, Soniaa; Morello, Judita; Gonzalez, Mara; Vispo, Eugeniab; Barreiro, Pablob; Gonzalez-Pardo, Gemaa; Jimenez-Nacher, Inmaculadaa; Gonzalez-Lahoz, Juanb; Soriano, Vincentb aPharmacokinetic and Pharmacogenetic Unit, Spain bDepartment of Infectious Diseases, Hospital Carlos III, Madrid, Spain.
Atazanavir use is associated with increases in serum bilirubin. Ribavirin, used to treat hepatitis-C infection, cause hemolysis and may worsen hyperbilirubinemia. We studied HIV/hepatitis-C virus-coinfected patients who initiated hepatitis-C therapy. Hyperbilirubinemia grade 3-4 increased from 9% to 45% after the start of hepatitis-C treatment in patients who used atazanavir concomitantly. Atazanavir use and hemoglobin (Hb) drops were predictors of increases in bilirubin. A substantial proportion of patients under atazanavir-therapy experienced significant hyperbilirubinemia and jaundice following initiation of hepatitis-C therapy.
Atazanavir (ATV) is a protease inhibitor used as part of HIV therapy. Hyperbilirubinemia is the most common side effect in patients treated with ATV, and it directly correlates with ATV plasma concentrations [1]. Increase in serum bilirubin results from an inhibitory competition by ATV of the uridinglucuronosyltransferase (UGT) 1A1 enzyme, which is responsible for bilirubin conjugation [2].
Around one-third of HIV-infected patients are coinfected with hepatitis-C virus (HCV) and may require treatment with pegylated-interferon (peg-IFN) and ribavirin (RBV) [3,4]. The main side effect of RBV is the occurrence of reversible hemolytic anemia [5]. Although drug interactions between RBV and anti-HIV nucleoside analogues have been described [6-8], no interactions have been reported so far between RBV and ATV. However, as both drugs may enhance serum bilirubin levels, the risk of hyperbilirubinemia and jaundice could be increased in HIV/HCV-coinfected patients on stable ATV therapy following initiation of hepatitis-C treatment.
Herein, we report changes in serum bilirubin in HIV patients on ATV who initiated hepatitis-C treatment. A second objective of the study was to assess the impact of the most common polymorphism at the UGT1A1 gene (allele*28, associated with the mild form of the Gilbert's syndrome) on serum bilirubin levels in this population [9,10].
A total of 72 HCV/HIV-coinfected patients who initiated hepatitis-C therapy (peg-IFN weekly and RBV 1000-1200 mg/day) were retrospectively selected, 36 on ATV and 36 on other antiretrovirals but indinavir (controls). Checking pharmacy records, clinical charts and plasma drug levels, poor drug adherence was noticed in 14 patients within the ATV group and in six controls and they were excluded from analyses. ATV and RBV concentrations were measured after 4 weeks following initiation of hepatitis-C therapy (around 12 h after intake) using a validated HPLC-ultraviolet method [11,12]. UGT1A1 genotypes were determined by sequencing. Biochemical parameters were recorded before and 1 month after beginning hepatitis-C treatment. All statistics were calculated using the SPSS package-v11.0 (SPSS Inc., Chicago, Illinois, USA), and differences were considered to be significant when P < 0.05.
Briefly, 27% were women; median (range) age, 44 (40-47) years; serum bilirubin, 1.25 (0.8-2.4) mg/dl. All patients were taken a combination of two nucleoside analogues and either ATV (31%, ATV 400 mg/day and 12%, ATV 300 mg boosted with ritonavir 100 mg) or another third agent (35% lopinavir and 22% other antiretrovirals). No changes in ATV dosing were made along the study period. There was no differences between groups in sex, age and zidovudine use. The number of patients carrying the allele *28 or with basal hyperbilirubinemia grade 3-4 (>3.26 mg/dl) were similar in both groups.
Compared with baseline, there was a significant increase at week 4 in median serum bilirubin following initiation of RBV (P = 0.014). This elevation in median bilirubin levels was more pronounced in patients on ATV than in controls [0.80 (0.08-1.43) vs. 0.15 mg/dl (-0.3-0.35); P = 0.003]. On average, serum bilirubin increases following initiation of peg-IFN and RBV were 1.9-fold higher in patients on ATV than in controls.
Compared with baseline, the proportion of patients experiencing serum bilirubin increases of more than 1 mg/dl was higher in the 22 patients on ATV than in the 30 controls (45 vs. 3%; P = 0.001). In the ATV-group, the proportion of patients with hyperbilirubinemia-grade 3-4 increased 2.5-fold after beginning hepatitis therapy, from 9% (2/22) to 45% (10/22); (P = 0.021). None of the patients in the control group developed hyperbilirubinemia grade 3-4.
The distribution of UGT1A1*28 genotypes was as follows: common allele (43%), heterozygous (53%) and homozygous (6%). There was no association between UGT1A1 genotypes and increases in serum bilirubin. Only one patient in the ATV-group was homozygous for UGT1A1*28; he experienced a serum bilirubin increase from 1.5 to 2.7 mg/dl following initiation of RBV therapy.
In the bivariate analysis in which sex, age, baseline Hb, RBV plasma trough concentration, ATV use, Hb drops and allele UGT1A1-*28 were included, only the use of ATV and Hb drops were associated with serum bilirubin elevations more than 1 mg/dl. Moreover, these two factors remained as independent predictors in the multivariate analysis (Table 1).
The present study shows that a large number of HIV-infected patients on stable ATV therapy may experience an increase in total serum bilirubin levels following initiation of hepatitis-C therapy. In fact, hyperbilirubinemia grade 3-4 increased from 9% to 45% in patients who were taking ATV as part of their antiretroviral regimen. The elevation in serum bilirubin levels was directly related to the Hb decline as a result of the use of RBV and hemolysis [5]. RBV enter erythrocytes via the equilibrative nucleoside transporter 1 [13]. Once RBV is inside the erythrocytes, it is phosphorylated to the corresponding triphosphate form, spending a relatively high amount of ATP, leading to a reduction in ATP intra-erythrocyte levels. Given that erythrocytes lack a nucleus and mitochondria, the only mechanism to produce ATP is through glucolysis. For this reason, the antioxidant mechanisms vanish more rapidly and the erythrocyte membrane can no longer be preserved and is subject to vascular phagocytosis [14]. Hemolysis accounts for enhanced Hb degradation and increased production of bilirubin by the reticuloendothelial cells. Finally, bilirubin is transported to the liver to be eliminated. Given that ATV inhibits the UGT1A1 enzyme in the liver, the physiological clearance of bilirubin is seriously compromised and hyperbilirubinemia develops.
Unexpectedly, we did not find an association between the UGT1A1*28 allele and increases in serum bilirubin in patients taking ATV and RBV concomitantly, despite the impact of this allele on UGT1A1 activity being well known. Most likely, most patients with Gilbert's syndrome had already been excluded from the study population, which only recruited patients on stable ATV therapy. Therefore, this fact explained the relatively low frequency of the homozygote allele *28 in our patient population.
In summary, a relatively large number of HIV patients on stable ATV therapy show an increase in serum bilirubin levels following initiation of hepatitis-C therapy. This increase appears to be related to the use of ATV and to the extent of the Hb decline due to RBV-associated hemolysis. The normal clearance of bilirubin might be compromised due to the inhibitory competition of ATV, causing jaundice in HIV patients following initiation of hepatitis-C therapy.
  iconpaperstack view older Articles   Back to Top