Infection

Clinical and Epidemiological Study

Hyperbilirubinemia during Atazanavir Treatment in 2,404 Patients in the Italian Atazanavir Expanded Access Program and MASTER Cohorts
C. Torti, G. Lapadula, A. Antinori, T. Quirino, R. Maserati, F. Castelnuovo, F. Maggiolo, A. De Luca, G. Paraninfo, F. Antonucci, G. Migliorino, A. Lazzarin, G. Di Perri, G. Rizzardini, R. Esposito, G. Carosi
Abstract
Background: Although the mechanism of atazanavir (ATV)-related hyperbilirubinemia is well identied, its prevalence, risk factors, and association with transaminase ares have rarely been assessed in a large population from the real life setting. Methods: Prospectively collected data on 2,404 patients from the Italian MASTER Cohort and the Italian ATV expanded access program database were examined. Uni- and multivariable Cox proportional hazards regression models were conducted to identify risk factors for grade III hyperbilirubinemia during the administration of ATV. The risk of increased levels of serum alanine aminotranferase (ALT) was compared between patients with or without grade III hyperbilirubinemia in a Cox regression analysis stratied by hepatitis C virus (HCV) serostatus. Results: Grade III and IV hyperbilirubinemia were observed in 1,072 (44.6%) and 174 (7.2%) of the patients, respectively. Higher CD4+ T-cell counts, abnormal bilirubinemia at baseline, and ritonavir co-administration were associated with a higher risk of developing grade III hyperbilirubinemia. In contrast, female gender, clinical class C, and non-nucleoside reverse transcriptase co-administration appeared to be protective. Higher bilirubinemia at baseline and the use of ritonavir were associated with a higher risk of grade IV hyperbilirubinemia. The occurrence of grade III hyperbilirubinemia was not associated with severe hepatotoxicity (hazard ratio 1.00, 95% condence interval 0.641.57; p = 0.997). Conclusions: Hyperbilrubinemia is a common side effect of an ATV pharmacotherapeutic regimen. However, grade IV increase in bilirubin was rarely found. In most cases, ATV hyperbilirubinemia appeared to be an innocent phenomenon as far as the risk of a subsequent increase in liver enzyme level is concerned.
Infection 2009; 37: 244249 DOI 10.1007/s15010-008-8010-6

hyperbilirubinemia with a prevalence ranging from < 20% to 47% [16]. These studies were conducted on a limited number of patients, with a low prevalence of hepatitis C virus (HCV) coinfection. Moreover, possible co-determinants of hyperbilirubinemia (including hepatitis co-infection and stage of liver disease) are unclear, and the predictive value of ATV-related hyperbilirubinemia for the risk of elevated liver transaminase levels needs to be further investigated.

Patients and Methods Patients

An analysis of prospectively collected data from the Italian MASTER (Management Standardizzato di Terapia Antiretro-

C. Torti (corresponding author), G. Lapadula, G. Carosi School of Medicine, Institute of Infectious and Tropical Diseases, University of Brescia, P.le Spedali Civili, 1, 25123, Brescia, Italy; Phone: +(39/030) 3996624, Fax: 303061, e-mail: torti.carlo@libero.it A. Antinori, F. Antonucci National Institute of Infectious Diseases, Rome, Italy T. Quirino, G. Migliorino Ospedale di Circolo, Busto Arsizio, Italy R. Maserati Institute of Clinical Infectious Diseases, IRCCS Policlinico S. Matteo, Pavia, Italy F. Castelnuovo, G. Paraninfo Spedali Civili, Brescia, Italy F. Maggiolo Ospedali Riuniti, Bergamo, Italy A. De Luca Universit Cattolica del Sacro Cuore, Rome, Italy A. Lazzarin S. Raffaele Hospital, Milan, Italy G. Di Perri Turin University, Turin, Italy G. Rizzardini Sacco Hospital, Milan, Italy R. Esposito Modena University, Modena, Italy C. Torti and G. Lapadula equally contributed to this study.

Introduction
Hyperbilirubinemia is a frequent side effect of treatment with an atazanavir (ATV)-containing antiretroviral regimen. Previous studies have reported cases of severe

This work was partially presented during the 11th European AIDS Conference, 2427 Oct 2007, at Madrid, Spain (abstract P9.6/03). Received: December 28, 2007 Revision accepted: June 24, 2008 Published online: May 26, 2009

244

Infection 37 2009 No. 3 URBAN & VOGEL

C. Torti et al. Atazanavir Hyperbilirubinemia and Hepatotoxicity

virale) Cohort and from the Italian ATV Expanded Access Program (EAP) database (Protocol AI424900) was conducted. The Italian MASTER Cohort is a longitudinal multicenter cohort consisting of a general HIV patient population in 13 referral centers throughout Italy (http://www.mastercohort.it). The distinguishing characteristic of this cohort is that data are compiled in a common electronic database (Health and Notes ver. 3.5; Healthware S.p.A., Naples, Italy) in use in the participating centers for clinical purposes. The resulting cohort is therefore an open cohort in which non-preselected patients are continuously being enrolled. Data are recorded over a standardized time-scale every 3 months, and merging and data cleaning are performed at a central level every 6 months. All sequential antiretroviral experienced patients included in the MASTER cohort who were started on any highly active antiretroviral therapy (HAART) that included ATV for the rst time between 2003 and 2006 were enrolled in our study. Measurements of bilirubinemia and alanine aminotranferase (ALT) levels within 90 days prior to starting the treatment were a prerequisite to participation. For the ATV EAP, patients satisfying the following criteria were enrolled between 2002 and 2005: (1) treatment failure dened as HIV antiretroviral resistance, metabolic abnormalities, or other intolerance or adherence problems; (2) inability to construct an alternative effective HAART regimen; (3) serum creatinine < 1.5 upper limit of normality (ULN), liver enzyme elevation < 5 ULN, total serum bilirubin < 1.5 ULN. Duplication of patient cases in the two databases was avoided. The follow-up was censored in the case of ATV discontinuation, loss of followup, or at month 12, whichever came rst. Patients were classied as HCV-positive or HCV-negative based on the results of the HCV antibody (Ab) test at inclusion.

or without grade III hyperbilirubinemia in the first three months on the ATV-containing regimens. For this comparison, stratification by HCV serostatus was applied in the Cox regression analysis. All data analyses were conducted using SAS statistical software ver. 8 (SAS Institute, Cary, NC). All p-values presented are two tailed, and p < 0.05 indicates conventional statistical significance.

End-points
Grade III or IV bilirubin alterations according to the AIDS Clinical Trial Group (ACTG) classification (i.e., any increase by > 2.5-fold ULN or by > fivefold ULN, respectively) were used as outcome measures. Any increase by > fivefold ULN in ALT levels was considered to be grade III hypertransaminasemia. The ULN was 1.2 mg/dl (20.5 lmol/l) and 40 IU/l for bilirubinemia and ALT, respectively.

Results Population Characteristics Among the 17,456 patients included in the MASTER database, 2,163 patients started a new ATV-containing HAART in the study time-frame, of whom 402 (18%) were not enrolled in the study because they lacked a bilirubin measurement either at baseline or after ATV initiation. Data on the remaining 1,761 patients from the MASTER database were merged with those of 643 patients from the ATV-EAP (see Table 1). The resulting cohort of 2,404 patients was followed for a median of 257 (interquartile range [IQR] 121331) days. The median number of bilirubin measurements was four (IQR 25). Most patients were males and former intravenous drug use (IVDU). HCV-Ab-positive patients accounted for 47.3% of the study population. ATV was mainly prescribed in combination with ritonavir as booster (75.8%). Patients in the MASTER cohort were more likely to have positive HCV-Ab and/or HBsAg and higher bilirubin levels and to use ritonavir boosting than those in the ATV EAP (p < 0.001). Incidence of Hyperbilirubinemia and Hypertransaminasemia Among the 2,404 patients studied, 1,072 (44.6%) experienced grade III hyperbilirubinemia. The proportion of patients with grade III hyperbilirubinemia was higher in the Italian MASTER cohort than in the ATV EAP database (48.4 vs 34.2%; chi-square p < 0.001). Grade IV hyperbilirubinemia was observed in 174 patients (7.2%). Again, more patients from the MASTER database experienced grade IV hyperbilirubinemia than patients from ATV EAP database (8.8% vs 2.9%; p < 0.001). Grade III ALT increase occurred in 155 patients (6.4%). Patients with positive HCV Ab were more likely to experience severe hypertransaminasemia than HCV Ab negative individuals (11.9% vs 1.7%; p < 0.001). Predictors of Hyperbilirubinemia
In the multivariable analysis, higher CD4+ T-cell counts at baseline, abnormal bilirubinemia at baseline, and the use of a ritonavir booster were associated with a higher risk of grade III hyperbilirubinemia. By contrast, female gender, clinical class C, and the use of NNRTI co-medication was protective (Table 2). HCV-Ab positivity, which was slightly associated with a risk of relevant hyperbilirubinemia in the univariate analysis (see Figure 1), was no longer associated with the outcome using

Statistical Analysis
Cox proportional hazards regression models were conducted to assess possible associations of the following factors with either grade III or IV hyperbilirubinemia: gender, age, risk factor for HIV acquisition, clinical class C {i.e., occurrence of major opportunistic infections according to the Centers of Disease Control (CDC93) classification [7]), HCV-Ab and hepatitis B surface antigen (HBsAg) serostatus, CD4+ T-cell count and HIV-RNA at baseline (i.e., at initiation of ATV-containing regimens), baseline ALT, baseline bilirubinemia, aspartate aminotransferase (AST)/platelet ratio index (APRI)-score as a surrogate of the stage of liver disease [8], use of low-dose ritonavir, inclusion of other protease inhibitors (PIs) at active dosages in the regimens, and non-nucleoside reverse transcriptase inhibitors (NNRTI) or tenofovir (TDF) co-medications. The variables found to be signicant (p < 0.2) by the univariate analysis were entered into multivariable models. The study cohort was also used as a covariate in all multivariable models explored. To explore whether hyperbilirubinemia could predict the occurrence of severe hypertransaminasemia among the patient population, the risk of ALT increase in month 3 of the pharmacotherapeutic regimen was compared between patients with

Infection 37 2009 No. 3 URBAN & VOGEL

245

C. Torti et al. Atazanavir Hyperbilirubinemia and Hepatotoxicity

Table 1 Patient characteristics of the 2,404 study patients starting atazanavir-containing regimens.

Baseline characteristics Female gender (%) Age, mean years (SD) HIV risk factor (%) Intravenous drug use Homo/Bi-sexual intercourses Heterosexual intercourse Other/Not declared Clinical class C CDC93 (%) HCV-Ab (%) Negative Positive Unknown HBsAg (%) Negative Positive Unknown CD4+ T-cell/mm3, median (IQR) HIV-RNA <50 copies/ml (%) ALT level, median UI/ml (IQR) AST level, median UI/ml (IQR) Bilirubin level, median mg/dl (IQR) Bilirubin >1.2 mg/dl20.5 lmol/l (%) Platelets 103/mm3, median (IQR) APRI-score, median (IQR) Use of ritonavir boosting (%) Use of double PI (%) Use of NNRTI (%) Use of tenofovir (%)

MASTER database n = 1,761 32.5 42.3 (7.8) 43.5 11.9 34.8 9.8 26.9 42.8 41.7 15.5 79.3 7.4 13.3 341 (202492) 43.6 40 (2368) 33 (2451) 0.61 (0.470.88) 9.4 204 (155252) 0.36 (0.220.72) 85.5 2.9 2.7 66.8

Italian ATV-EAP database n = 643 34 43.1 (9.8) 35.6 23.5 31.1 9.8 39.5 59.7 35.6 4.7 90 6.7 3.3 273.5 (145432.5) 21.9 35 (2358) 32 (2251) 0.55 (0.40.77) 8.1 185 (140228) 0.4 (0.260.73) 49.1 5.1 5.3 63.4

Overall population n = 2,404 32.9 42.5 (8.4) 41.4 15 33.8 9.8 30.2 47.3 40.1 12.6 82.2 7.2 10.6 322 (190479) 38.6 38 (2365) 33 (2351) 0.6 (0.450.84) 9.1 198 (150247) 0.37 (0.230.72) 75.8 3.5 3.4 65.9

ATZ: Atazanavir; MASTER: Management Standardizzato di Terapia Antiretrovirale; EAP: Expanded Access Program; ALT: alanine amino transferase; AST: aspartate amino transferase; APRI: AST/platelet ratio index; ATV-EAP: atazanavir expanded access program; CDC, Centers for Disease Control and Prevention; HBsAg: hepatitis B surface antigen; HCV-Ab: hepatitis C virus antibodies; NNRTI: non-nucleoside reverse transcriptase inhibitor; PI: protease inhibitor

the multivariable analysis. When grade IV hyperbilirubinemia was considered to be a dependent variable, abnormal bilirubinemia at baseline (hazard ratio [HR] 3.55, 95% condence interval [CI] 2.485.01; p < 0.001) and the use of a ritonavir booster (HR 1.66, 95%CI 1.06 2.59; p = 0.026) were independent risk factors, while the use of NNRTI was protective, although not reaching statistical signicance (HR 0.30, 95%CI 0.071.21; p = 0.090). A cohort effect was found because patients included in the Italian ATV EAP had a lower risk of grade III hyperbilirubinemia (HR 0.39, 95%CI 0.24 0.65; p < 0.001). Moreover, a paradoxically protective effect of HCV-Ab positivity on the risk of severe hyperbilirubinemia was demonstrated (HR 0.65, 95%CI 057 0.92; p = 0.013).

these patients, 39 suffered from a grade III ALT increase, accounting for an incidence of 0.12 (95%CI 0.08 0.16) per patient-year of follow-up. The results were not statistically different among the 1,264 patients who did not display grade III hyperbilirubinemia in the first 3 months. In fact, 59 of these latter patients suffered from a severe ALT increase, accounting for an incidence of 0.10 (95%CI: 0.080.13) per patient-year of follow-up. Even after stratification by HCV-Ab serostatus, the occurrence of grade III hyperbilirubinemia was not associated with a higher risk of severe hepatotoxicity (HR 1.00, 95%CI 0.641.57; p = 0.997).

Discussion
ATV efficacy and tolerability profiles have been studied in both antiretroviral-naive and experienced patients [14, 6, 911]. To the best of our knowledge, the data presented herein were obtained from the largest cohort used to date to estimate the incidence of hyperbilirubinemia during treatment with ATV and to explore its possible predictors. Grade III hyperbilirubinemia occurred in almost

Risk of Severe Hepatotoxicity by Occurrence of Previous Relevant Hyperbilirubinemia Among 1,961 patients followed up for 3 months, 697 (35.5%) developed grade III hyperbilirubinemia in the first 3 months of the ATV-containing HAART. Among

246

Infection 37 2009 No. 3 URBAN & VOGEL

C. Torti et al. Atazanavir Hyperbilirubinemia and Hepatotoxicity

Table 2 Uni- and multivariable Cox regression analyses assessing predictors of grade III hyper-bilirubinemia (i.e., >3 mg/dl/51.3 lmol/l)

Predictors Database Master Italian ATV-EAP Female gender Age (per 1 year increase) HIV risk factor Intravenous drug use Homo/Bi-sexual intercourses Heterosexual intercourses Other/Not declared Clinical class C CDC93 HCV-Ab Negative Positive Unknown HBsAg Negative Positive Unknown CD4+ T-cell (per 50 cells/mm3 increase) HIV-RNA <50 copies/ml Bilirubin >1.2 mg/dl APRI-score (per unit increase) Use of ritonavir boosting Use of double PI Use of NNRTI Use of tenofovir

Unadjusted HR 1 0.61 0.84 0.997 1 0.89 0.87 0.96 0.77 1 1.14 1.22 1 1.29 1.27 1.03 1.33 2.29 1.02 2.07 0.78 0.35 1.1 95%CI 0.530.71 0.740.96 0.991.004 0.741.06 0.761.01 0.781.18 0.670.89 0.991.29 1.011.48 1.041.59 1.051.53 1.011.04 1.181.51 1.932.72 0.991.05 1.762.44 0.541.11 0.220.56 0.971.25 p

Adjusted HR 1 0.86 0.86 95%CI 0.711.03 0.750.99 N.E. N.E. p

<0.001 0.012 0.351

0.097 0.031

0.242 <0.001

0.85 1 1.02 1.07 1 1.20 1.18 1.01 1.13 2.26 1.90 0.35

0.730.98 0.901.17 0.801.44 0.961.50 0.881.60 1.0011.03 0.991.30 1.882.70 N.E. 1.582.29 Removeda 0.210.59 Removeda

0.030

0.044

0.863

0.006 <0.001 <0.001 <0.001 0.256 <0.001 0.168 <0.001 0.142

0.176 0.033 0.083 <0.001 <0.001 <0.001

HR: Hazard ratio; CI: condence interval; N.E.: not entered a The use of double PI and of tenofovir were removed from the nal multivariable model, which produced the most clinically and statistically signicant result

half of our patients, thus conrming that this condition is a common side effect of ATV. However, grade IV hyperbilirubinemia occurred in only 7.2% of the study population. Bilirubinemia > 1.2 mg/dl (20.5 lmol/l) at baseline appeared to be very important because hyperbilirubinemia risk increased by more than twofold in our patient population independently of other factors. The concomitant use of ritonavir appeared to increase the risk of hyperbilirubinemia, while the concomitant use of NNRTI appeared to decrease it. Such an effect was probably due to the pharmacokinetic interaction between ATV and these drugs because ATV plasma concentrations are increased by ritonavir [12] and decreased by NNRTI [13]. Indeed, higher ATV plasma concentrations have been correlated with hyperbilirubinemia [14]. The correlation between a lower risk of hyperbilirubinemia and the prescription of NNRTI can also be explained by the induction of the metabolism of bilirubin induced by NNRTI (specically efavirenz) [15]. Baseline immunological and clinical conditions influenced the risk of grade III hyperbilirubinemia. Higher CD4+ T-cell counts at baseline predicted a higher risk of

hyperbilirubinemia. Consistent with this observation, patients with advanced clinical conditions at baseline (i.e., previous AIDS defining events) displayed a lower risk than the others. Our results suggest that specific counseling and discussion regarding the possibility of developing hyperbilirubinemia or possibly jaundice should be specifically targeted to patients with a high baseline CD4+ T-cell count. Interestingly, HCV-Ab positivity did not appear to increase the risk of hyperbilirubinemia. Although HCV seropositivity is only a surrogate for the presence of chronic hepatitis C, this finding is inconsistent with the hypothesis that liver damage impairs drug clearance, thus increasing plasma concentrations, and, therefore, the risk of hyperbilirubinemia. A cohort effect was found because the patients included in the MASTER cohort had a higher risk of hyperbilirubinemia than those included in the ATV-EAP due to their intrinsic differences at baseline. However, our results were confirmed in models conducted in the two cohorts taken separately. The risk of severe hypertransaminasemia in our cohort was strictly correlated with HCV-Ab serostatus.

Infection 37 2009 No. 3 URBAN & VOGEL

247

C. Torti et al. Atazanavir Hyperbilirubinemia and Hepatotoxicity

Proportion of patients free-of-event

Figure 1. Time to development of grade III (i.e., >3 mgdl51.3 lmol/l) or grade IV (i.e., >6 mg/dl102.6 lmol/l) hyperbilirubinemia.

1.00

0.50

0.75

GradeIV Grade IV hyper-bilirubinemia

0.00

0.25

Grade hyper-bilirubinemia

0.5

Time from entry (years)

HCV-Ab negative HCV-Ab positive

Moreover, ATV hyperbilirubinemia was not correlated or predictive for hypertransaminasemia. This result suggests that, in most cases, the two phenomena arise through different mechanisms and that, generally, hyperbilirubinemia is an innocent phenomenon as far as the risk of subsequent liver enzyme elevation is concerned. Our study has both strengths and limitations. To the best of our knowledge, this is the largest cohort of patients treated with ATV reported to date, thereby enhancing the precision of our estimates. However, although data were prospectively recorded, our study was observational in nature. Therefore, biases cannot be excluded. Specifically, the rarity of grade IV hyperbilirubinemia could be due to the fact that patients undergoing low-level hyperbilirubinemia could have prematurely discontinued ATV. For the same reason, the chance to observe ALT increase after bilirubin elevation could have been low. Moreover, the reason for switching to ATV as well as adherence and ATV pharmacokinetics were not recorded in the database, thus limiting our inferences to a significant extent. HCV coinfection was defined based on the presence of Ab reactivity. As patients may control or eradicate HCV, Ab detection does not necessarily indicate that coinfection with HCV is present. Lastly, although we did not study the possible effects of hyperbilirubinemia on patient quality of life or on ATV interruption, current clinical practice and results from trials indicate that the latter event occurs rarely [16].

G. Carosi (Chair), C. Torti, G. Lapadula, M. Puoti, E. QuirosRoldan, G. Paraninfo, S. Casari, G. Cristini, F. Castelnuovo, I. El Hamad (Brescia); A. Antinori, G. Antonucci, A. Ammassari (INMI L. Spallanzani, Rome); G. Angarano, A. Saracino (Fog` Cattolica del Sacro gia); R. Cauda, A. De Luca (Universita Cuore, Rome); A. DArminio Monforte, P. Cicconi (Milan); F. Mazzotta, S. Lo Caputo, N. Marino (Florence); L. Minoli, R. Maserati, S. Novati, C. Tinelli (Pavia); F. Ghinelli, L. Sighinolfi (Ferrara); G. Pastore, N. Ladisa, (Bari); T. Quirino (Busto Arsizio); F. Suter, F. Maggiolo (Bergamo); G. Carnevale, A. Pan (Cremona); A. Gori (Monza). The following individuals are participants in the Italian Expanded Access Program (AI424900) Database (cities in which the clinical centers are based are given in parenthesis): A. Lazzarin, A. Castagna, M. Guffanti (San Raffaele Hospital, Milan); G. Di Perri, S. Bonora, L. Trentin (Turin), G.Rizzardini, A. Capetti, P. Meraviglia (Sacco Hospital, Milan), R. Esposito (Modena), M. Moroni, T. Bini (Sacco Hospital, Milan), F. Aiuti, (La Sapienza University, Rome), Caramello (Turin), G. Filice (Pavia), G. Penco (Genoa), Montoni (Ancona), Narciso (INMI L. Spallanzani, Rome), C. Viscoli, A. Di Biagio (Genoa).

References
1. Johnson M, Grinsztejn B, Rodriguez C, et al. Atazanavir plus ritonavir or saquinavir, and lopinavir/ritonavir in patients experiencing multiple virological failures. AIDS 2005; 19: 685694. 2. Molina JM, Andrade-Villanueva J, Echevarria J et al.: Efcacy and safety of once-daily atazanavir/ritonavir compared to twicedaily lopinavir/ritonavir, each in combination with tenofovir and emtricitabinein in ARV-naive HIV-1-infected subjects: The CASTLE study, 48-week results (abstract no. 37). Boston, MA: 15th Conference on Retroviruses and Opportunistic Infections, 2008. 3. Sanne I, Piliero P, Squires K, Thiry A, Schnittman S: Results of a phase 2 clinical trial at 48 weeks (AI424007): a dose-ranging,

Acknowledgments
The following individuals are participants in the MASTER Cohort (cities in which the clinical centers are based are given in parenthesis):

248

Infection 37 2009 No. 3 URBAN & VOGEL

C. Torti et al. Atazanavir Hyperbilirubinemia and Hepatotoxicity

4.

5.

6.

7.

8.

9.

safety, and efcacy comparative trial of atazanavir at three doses in combination with didanosine and stavudine in antiretroviral-naive subjects. J Acquir Immune Dec Syndr 2003; 32: 1829. Squires K, Lazzarin A, Gatell JM, et al. Comparison of once-daily atazanavir with efavirenz, each in combination with xed-dose zidovudine and lamivudine, as initial therapy for patients infected with HIV. J Acquir Immune Dec Syndr 2004; 36: 10111019. Ward D, Curtin J, Richardson J et al: Hyperbilirubinemia among Patients in the HIV Outpatient Study (HOPS) receiving atazanavir (abstract TuPe2 1B02). Rio de Janeiro: 3rd International AIDS Conference on HIV Pathogenesis and Treatment, 2005. Wood R, Phanuphak P, Cahn P, et al. Long-term efcacy and safety of atazanavir with stavudine and lamivudine in patients previously treated with nelnavir or atazanavir. J Acquir Immune Dec Syndr 2004; 36: 684692. Anonymous: 1993 Revised classication system for HIV infection and expanded surveillance case denition for AIDS among adolescents and adults. MMWR Recomm Rep 1992; 41: 119. Al-Mohri H, Cooper C, Murphy T, Klein MB: Validation of a simple model for predicting liver brosis in HIV/hepatitis C virus-coinfected patients. HIV Med 2005; 6: 375378. Gatell J, Salmon-Ceron D, Lazzarin A, et al. Efcacy and safety of atazanavir-based highly active antiretroviral therapy in patients with virologic suppression switched from a stable, boosted or unboosted protease inhibitor treatment regimen: the SWAN

Study (AI424097) 48-week results. Clin Infect Dis 2007; 44: 14841492. 10. Havlir DV, OMarro SD: Atazanavir: new option for treatment of HIV infection. Clin Infect Dis 2004; 38: 15991604. 11. Murphy RL, Sanne I, Cahn P, et al. Dose-ranging, randomized, clinical trial of atazanavir with lamivudine and stavudine in antiretroviral-naive subjects: 48-week results. AIDS 2003; 17: 26032614. 12. Colombo S, Buclin T, Cavassini M, et al. Population pharmacokinetics of atazanavir in patients with human immunodeciency virus infection. Antimicrob Agents Chemother 2006; 50: 38013808. 13. Dailly E, Tribut O, Tattevin P, et al. Inuence of tenofovir, nevirapine and efavirenz on ritonavir-boosted atazanavir pharmacokinetics in HIV-infected patients. Eur J Clin Pharmacol 2006; 62: 523526. 14. Smith DE, Jeganathan S, Ray J: Atazanavir plasma concentrations vary signicantly between patients and correlate with increased serum bilirubin concentrations. HIV Clin Trials 2006; 7: 3438. 15. Rotger M, Taffe P, Bleiber G, et al. Gilbert syndrome and the development of antiretroviral therapy-associated hyperbilirubinemia. J Infect Dis 2005; 192: 13811386. 16. Johnson M, Grinsztejn B, Rodriguez C, et al. 96-week comparison of once-daily atazanavir/ritonavir and twice-daily lopinavir/ ritonavir in patients with multiple virologic failures. AIDS 2006; 20: 711718.

Infection 37 2009 No. 3 URBAN & VOGEL

249

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.