HIV clinicians may have concerns about the potential safety of buprenorphine/naloxone (bup/nx) treatment in HIV-infected patients.1,2 Establishing the safety of bup/nx for the treatment of opioid dependence in this patient population is a necessary step in the expansion of bup/nx to the HIV primary care setting. Accordingly, providing empiric evidence for the safety of its use among HIV-infected patients and patients with comorbid HIV and hepatitis C infection (HIV/HCV) is important. This matter is particularly true with regard to hepatotoxicity and clinically significant pharmacodynamic medication interactions with antiretrovirals.
Buprenorphine, a thebaine derivative, is an effective treatment for opioid dependence. There are, however, concerns about potential hepatotoxicity, particularly in patients with viral hepatitis and when co-administered with hepatotoxic medications. Early in vitro studies suggest that with high hepatocyte concentrations, buprenorphine is a proton donor that can interfere with mitochondrial respiration, resulting in hepatocyte necrosis and hepatic enzyme elevations.3 Acute hepatic enzyme elevations have been seen in individuals with HCV who injected buprenorphine intravenously.4 A case series reports 7 patients with HCV who developed acute transaminitis, up to 39 times the normal range, although receiving buprenorphine sublingually.5 The acute hepatitis, manifested as jaundice and transaminitis, resolved rapidly and spontaneously despite continuation of buprenorphine. Of these 7 cases, 4 continued at the same dose and 3 reduced their buprenorphine dose by 50%. Another study of patients receiving 2, 4, or 8 mg of buprenorphine per day monitored transaminase levels before medication initiation and subsequently after a minimum of 40 days.6 Compared with those without viral hepatitis, those with hepatitis exhibited statistically significant (but not necessarily clinically meaningful) increases in alanine aminotransferase (ALT), with a median increase of 8.5 U/L and a median increase in aspartate aminotransferase (AST) of 9.5 U/L. Higher buprenorphine doses were associated with greater odds of an increase in AST. Notably, none of these aforementioned studies has been conducted in HIV-infected patients receiving bup/nx.
Compared with methadone, buprenorphine has been shown to have a relatively small number of pharmacokinetic or pharmacodynamic drug interactions with antiretrovirals.7-12 Similar to many antiretrovirals, buprenorphine is primarily metabolized via the hepatic cytochrome P450 3A4 system to its main active metabolite, norbuprenorphine through n-dealkylation. Theoretically a number of potential interactions between bup/nx and medications commonly prescribed in HIV-infected patients are possible. In vitro pharmacokinetic studies suggest ritonavir inhibits n-dealkylation of buprenorphine by P450 3A4.7 One in vivo study documented increased buprenorphine concentrations, yet no adverse subjective effects in patients administered ritonavir, lopinavir/ritonavir, or nelfinavir.8 A recent study on the interaction between lopinavir/ritonavir and bup/nx in 12 patients showed no change of buprenorphine area under the curve (AUC) levels but did show a minimal decrease of norbuprenorphine AUC levels, though this difference did not lead to clinical withdrawal or require a change in dose.9 In a case report of 3 patients receiving bup/nx, the combination of atazanavir/ritonavir was associated with symptoms of opioid excess.10 Subsequently, the combination of bup/nx with either darunavir/ritonavir (600 mg/100 mg),11 atazanavir (400mg), or atazanavir/ritonavir (300 mg/100 mg)12 demonstrated increased AUC for either buprenorphine or norbuprenorphine in 2 human in vivo pharmacokinetic studies. Only the combination of atazanavir/ritonavir, however, demonstrated clinically significant interactions. Three of 10 patients reported feeling drowsy or sleepy on a symptom checklist when atazanavir/ritonavir was added to a stable bup/nx dose.12 The authors cautioned that patients receiving atazanavir and atazanavir/ritonavir may require reduced doses of bup/nx. The prevalence and severity of these symptoms among patients engaged in routine clinical care, however, is not known. More importantly, pharmacokinetic and pharmacodynamic medication interactions identified between bup/nx and antiretroviral medications have not been associated with serious adverse events, to date, and have unknown clinical significance.
Therefore the purpose of the current study was to describe the impact of bup/nx on hepatic enzymes (AST/ALT), particularly, in those with comorbid HIV/HCV, assess the need for bup/nx dose adjustments in patients receiving atazanavir and atazanavir/ritonavir, and conduct surveillance for pharmacodynamic medication interactions in HIV-infected opioid-dependent patients.
As described more fully in this supplement,13-15 from 2004 to 2009, the HIV/AIDS Bureau of the Health Resources and Services Administration funded, through its Special Projects of National Significance (SPNS), the development of demonstration programs that integrated HIV care and bup/nx treatment for opioid dependence at 10 sites across the United States. Health Resources and Services Administration also funded an Evaluation and Technical Assistance Center (the Center) to coordinate the multisite evaluation, provide clinical and evaluation support and technical assistance, and promote dissemination of findings. Data from 9 of the 10 sites were included in the evaluation. In brief, the current study reports observational data collected on 303 HIV-infected opioid-dependent patients who received at least 1 dose of bup/nx and were followed prospectively for a maximum of 12 months.
All sites enrolled patients who met the following criteria: 18 years or older, HIV-infected, DSM-IV criteria for opioid dependence, AST or ALT less than 5 times the upper limits of normal per the local reference laboratory, willing and able to participate for 1 year, able to provide informed consent, and able to speak English or Spanish. Patients were excluded if they met criteria for benzodiazepine dependence, alcohol dependence, were pregnant or trying to become pregnant, were acutely suicidal or had psychiatric conditions affecting their ability to provide informed consent (eg, dementia, delusions, active psychosis), or were deemed otherwise inappropriate for the study according to the clinical judgment of the prescribing physician.
Medical and Medication Data
Sites performed chart abstraction at baseline and on a quarterly basis. Chart abstractions collected data on AST and ALT values, HCV antibody status (baseline only), patient dose of bup/nx, and antiretroviral regimens. Sites were encouraged to measure AST and ALT at baseline before prescription of bup/nx and quarterly thereafter. Frequency of laboratory collection varied from site to site.
Surveillance for Potential Pharmacodynamic Interactions
Each site conducted independent surveillance for potential pharmacodynamic interactions. Staff and clinicians at each site were encouraged to report any potential pharmacodynamic interactions with bup/nx to the site Principal Investigator. Principal Investigators agreed to discuss any potential pharmacodynamic interactions with the Center. For surveillance, the Center provided the sites with a standardized form, based on forms used by the AIDS Clinical Trials Group. This form was distributed to each site before the study implementation and periodically during the course of the study. If potential pharmacodynamic interactions were noted, sites submitted the form to the Center. In addition, the Center coordinated monthly 1-hour long technical assistance conference calls to discuss issues related to clinical management and concerns regarding potential pharmacodynamic interactions.
To test for hepatotoxicity, we compared AST and ALT data at 2 time points using 2 separate methods. First AST and ALT median values were compared pre- and post-bup/nx prescription for patients who had data obtained from at least 2 time points. When more than 1 hepatic enzyme value was available for a patient, we used the last available value to maximize the duration of time that the patient was prescribed bup/nx. Second, we compared the medians and evaluated the ranges and outliers of the first and last AST and ALT values during bup/nx prescription. When more than 2 AST and ALT values were available for a patient, we used the first and last available values to maximize the duration of exposure to bup/nx. Finally, we examined cases of liver enzyme elevation which we defined as an increase of greater than 5 times baseline ALT or AST or greater than 3.5 times baseline if baseline was greater than 40 U/L.16
Buprenorphine/Naloxone Dose in Patients Receiving Atazanavir
We compared the mean bup/nx dose, by quarter, for those patients receiving atazanavir-containing regimens to the dose in those patients who were not receiving atazanavir-containing regimens.
All analyses were conducted using SPSS version 15.0. Analyses of baseline and quarterly measurements of AST and ALT values and bup/nx dose were analyzed using a Wilcoxon signed-rank test and Student t test, as appropriate. The P values ≤ 0.05 were considered statistically significant.
The baseline demographic and clinical characteristics of the 303 HIV-infected patients who received at least 1 dose of bup/nx are presented elsewhere.14,15 The majority of patients were male, black, high school graduates, unemployed, on antiretroviral medications, infected with HCV, and reported recent injection drug use and recent cocaine use. One hundred and eighty-one of the 303 patients (60%) were prescribed antiretroviral medications at baseline. Of these, 38 of 181 (21%) received regimens that included atazanavir, of whom 27 of 38 (71%) received the atazanavir/ritonavir combination.
Figure 1 shows the AST and ALT results for all patients with data available. For 141 patients who had hepatic enzyme values measured pre-bup/nx and post-bup/nx initiation, median AST (37.0 vs. 37.0 U/L, respective ranges 14-318 and 11-421 U/L) and ALT (33.0 vs. 33.0 U/L, respective ranges 7-240 and 6-383 U/L) values did not differ with prescription of bup/nx. There were 207 subjects with at least 2 sets of hepatic enzyme values assessed during bup/nx treatment, a median of 6 months apart (interquartile range, 4-8 months). Median AST (36.0 vs. 35.0 U/L, respective ranges 10-345 and 10-421 U/L) and ALT (29.0 vs. 31.0 U/L, respective ranges 5-383 and 6-383 U/L) did not change between the first and the last values in these patients. Table 1 shows the results by HCV antibody status. Among the 82 HCV antibody-positive patients with AST/ALT values measured pre-bup/nx and post-bup/nx prescription, the median AST and ALT values did not change over time. However, among 123 HCV antibody-positive subjects with at least 2 sets of hepatic enzyme values assessed during bup/nx prescription, the median AST demonstrated a small but statistically significant decrease from the first to the last value, although the ALT values did not change over time. Three patients met criteria for liver enzyme elevation. The first patient was HIV/HCV co-infected and had AST/ALT values of 65/75 U/L at baseline and 275/383 U/L during quarter 1. The second patient was also HIV/HCV co-infected who had baseline AST/ALT values of 37/19 U/L, which increased to 122/123 U/L in quarter 3 and decreased to 77/86 U/L in quarter 4. The last patient had HIV infection at baseline with AST/ALT values of 20/16 U/L and subsequently became HCV antibody positive with elevations in AST/ALT to 141/218 U/L in quarter 4. Additionally, in retrospect, 1 HIV/HCV-coinfected patient was entered into the study despite having AST/ALT values greater than 5 times normal at baseline, 318/230 U/L. In this case, surveillance of AST/ALT values during each quarter revealed persistent elevations through quarters 1, 2, and 3, without ever meeting criteria for liver enzyme elevation (302/187, 367/201, 421/282 U/L, respectively).
Buprenorphine/Naloxone Dose in Patients Receiving Atazanavir
Of those on highly active antiretroviral therapy, 38 patients were receiving atazanavir or atazanavir/ritonavir. Seventy-one percent (27 of 38) of those receiving atazanavir received ritonavir concurrently. There were no differences in mean bup/nx dose between those patients who received antiretroviral regimens containing atazanavir (n = 38) and those whose regimens did not contain atazanavir (n = 143) across the 4 quarters of follow-up (Table 2). In addition, no pharmacodynamic interactions between bup/nx and atazanavir-containing regimens were reported.
Potential Pharmacodynamic Medication Interactions
Neither of the two potential pharmacodynamic interactions that were reported to the Center was determined to represent a true medication interaction upon review by independent arbiters not involved in the conduct of the study.
Our findings from the largest clinical cohort of HIV-infected patients receiving bup/nx should provide some reassurance to HIV providers regarding the safety of prescribing bup/nx in their patients. In a large diverse cohort of patients receiving treatment at multiple sites, hepatotoxicity attributable to bup/nx was rare at most, and no clinically significant interactions with atazanavir or other antiretroviral medications were noted. Our data, collected prospectively by clinical staff and using standard and regularly scheduled assessments, represent the most complete, detailed information on the use of bup/nx among HIV-infected patients. These findings on safety augment those regarding drug abuse treatment effectiveness,17 HIV,18 and cost19 outcomes described in this supplement.
Our findings of limited elevation of hepatic enzymes among patients prescribed bup/nx mirror those observed in a smaller cohort of HIV-uninfected subjects,6 and in a case series of 4 patients who underwent acute hepatitis C seroconversion.20 HIV-infected patients with opioid dependence often have a high prevalence of HCV co-infection, consume alcohol, and receive antiretroviral medications plus other medications with known hepatotoxicity. The findings of the current study in patients with a high prevalence of these specific risk factors for hepatotoxicity support the safety of bup/nx in similar patients.
Importantly, this study demonstrates that most patients with HIV and HCV co-infection do not have adverse hepatic effects when prescribed bup/nx. Our findings cannot, however, be generalized to patients who have substantial elevations in their hepatic enzymes before initiating bup/nx because subject enrollment in the current study was predominantly restricted to patients whose hepatic enzymes were less than 5 times the upper limits of normal at baseline per the local reference laboratory.
The findings from this current study regarding atazanavir contrast with those in the literature. In a previous anecdotal report, symptoms of opioid excess (eg, hypersomnolence) were reported in 3 HIV-infected outpatients receiving atazanavir/ritonavir.10 In one of the subjects opioid tolerance was diminished or absent due to recent incarceration, and in another, multiple medications, including bup/nx, were initiated simultaneously. Unfortunately, no objective assessments were conducted on these patients; and active surveillance for potential concomitant psychoactive substances was not reported. The authors recommended cautious use of bup/nx with atazanavir and ritonavir. They speculated that atazanavir or ritonavir inhibited the major phase I metabolic pathway (CYP3A4), or that atazanavir inhibited the major phase II pathway for bup/nx, resulting in increased levels and clinical symptoms of opiate excess. These 3 patients 3 patients required daily doses of 4-12 mg of bup/nx.
A subsequent pharmacokinetic study, conducted on an inpatient research ward in patients who were not infected with HIV, collected serial objective measures of adverse effects including cognitive impairment, urine toxicology analyses, and breathalyzer testing. These patients were stabilized on 16 mg of bup/nx and challenged with exposure to atazanavir (400 mg) or atazanavir/ritonavir (300/100 mg).12 The study revealed elevations in levels of buprenorphine and its primary metabolite, norbuprenorphine, throughout the dosing cycle in subjects exposed to both atazanavir-containing regimens. Three of 20 patients reported symptoms of opioid excess (eg, feeling drowsy or sleepy) on a symptom checklist, although these measures did not reach statistical significance and the 3 patients did not have higher buprenorphine metabolite concentrations than those without sedation. The authors concluded that the interactions were mild and may be idiosyncratic.12
In the current study, no subjective effects were reported in patients receiving bup/nx and the atazanavir-containing regimens. In addition, the mean doses of bup/nx were not lower among patients receiving regimens that contained atazanavir. There are several potential explanations for the discrepant findings. Intersubject metabolism of buprenorphine is known to vary considerably.21 Second, because buprenorphine has high affinity for the mu opioid receptor, plasma levels do not necessarily correlate directly with pharmacologic activity. Third, there is a ceiling effect to the opioid agonist activity of buprenorphine at the mu opioid receptor and buprenorphine doses up to 40 mg/70 kg (5-6 times daily maintenance doses) have been tolerated without subjective sedation.22,23 Last, elevations in norbuprenorphine are thought unlikely to result in significant subjective effects as norbuprenorphine does not cross the blood-brain barrier efficiently and is a weak mu agonist.21 Although the total number of HIV-infected patients prescribed atazanavir or atazanavir/ritonavir at baseline who went on to be prescribed bup/nx (n = 38) is small, the number in this study is larger than both previous reports. One would anticipate that reports of intolerance of atazanavir would be more likely in this larger group.
There are limitations to our findings. As pointed out earlier, our findings do not generalize to the entire population of HIV-infected opioid-dependent patients who might receive bup/nx because some potential patients were excluded. Nonetheless, we believe the patient population in the current study was diverse and representative of the majority of patients for whom this medication would be considered. We were not able to confirm self-reported adherence to bup/nx, antiretroviral medications, or other medications. Our data were collected as part of routine clinical care and through structured assessments and chart review. It is possible that patients overreported their medication adherence. This was an observational study without a comparison group. In addition, the absence of hepatotoxicity or pharmacodynamic interactions does not mean that these do not exist. We do not know if the elevated AST and ALT values that we observed were related to the administration of bup/nx or to other heaptotoxic medications that were not documented or collected during the study. Furthermore, we are not able to comment on the contribution alcohol consumption could have played in these patients. Our assessment of HCV disease was limited to antibody status and was not confirmed by viral RNA. Finally, not all patients had AST or ALT values drawn on a routine basis. Our findings would be subject to surveillance bias if there were a systematic cause for missing data.24 The decreasing number of patients over 4 quarters introduces the possibility of retention bias. These biases could lead us to incorrectly determine that bup/nx was safe in this patient population, especially if adverse events were the cause of missing data. Because there were no reports of patients discontinuing bup/nx due to hepatotoxicity or medication interactions, we do not believe the missing data is due to concerns related to medication safety.
Our findings have implications for clinical care and research. HIV providers should feel more comfortable offering bup/nx to HIV-infected opioid-dependent patients who meet the eligibility criteria that were used in the current study. Patients with a wider range of comorbid conditions and hepatic dysfunction may also benefit from this treatment without adverse effects, although increased clinical monitoring and/or additional services (eg, psychiatric care and counseling) may be required. Serial assessments of hepatic enzymes may be prudent, especially in patients with HCV co-infection, alcohol consumption, and/or receiving potentially hepatotoxic medications in addition to bup/nx, at least until stable doses are achieved and lack of enzyme elevations is documented. The use of atazanavir-containing regimens should continue to be considered in patients receiving bup/nx, especially if there is a compelling clinical indication for their use. Our current findings highlight the fact that many patients tolerate the co-administration of atazanavir and bup/nx without adverse subjective or clinical effects. Finally, hepatic safety and lack of pharmacodynamic interactions in this, the largest cohort of HIV-infected patients who has received bup/nx and undergone prospective surveillance, provides a measure of reassurance that this medication can be used to help expand access to opioid agonist treatment for this important patient population.
1. Netherland J, Botsko M, Egan JE, et al. Factors affecting willingness to provide buprenorphine treatment. J Subst Abuse Treat
2. Sullivan LE, Barry D, Moore BA, et al. A trial of integrated buprenorphine/naloxone and HIV clinical care. Clin Infect Dis
. 2006;43(Suppl 4):S184-S190.
3. Berson A, Fau D, Fornacciari R, et al. Mechanisms for experimental buprenorphine hepatotoxicity: major role of mitochondrial dysfunction versus metabolic activation. J Hepatol
4. Peyrière H, Tatem L, Bories C, et al. Hepatitis after intravenous injection of sublingual buprenorphine in acute hepatitis C carriers: report of two cases of disappearance of viral replication after acute hepatitis. Ann Pharmacother
5. Herve S, Riachi G, Noblet C, et al. Acute hepatitis due to buprenorphine administration. Eur J Gastroenterol Hepatol
6. Petry NM, Bickel WK, Piasecki D, et al. Elevated liver enzyme levels in opioid-dependent patients with hepatitis treated with buprenorphine. Am J Addict
7. Iribarne C, Berthou F, Carlhant D, et al. Inhibition of methadone and buprenorphine N-dealkylations by three HIV-1 protease inhibitors. Drug Metab Dispos
8. McCance-Katz EF, Moody DE, Smith PF, et al. Interactions between buprenorphine and antiretrovirals. II. The protease inhibitors nelfinavir, lopinavir/ritonavir, and ritonavir. Clin Infect Dis
. 2006;43(Suppl 4):S235-S246.
9. Bruce RD, Altice FL, Moody DE, et al. Pharmacokinetic interactions between burprenorphine/naloxone and once daily lopinavir/ritonavir. J Acquir Immune Defic Syndr
10. Bruce RD, Altice FL, Gourevitch MN, et al. Pharmacokinetic drug interactions between opioid agonist therapy and antiretroviral medications: implications and management for clinical practice. J Acquir Immune Defic Syndr
11. Sekar V, et al. Pharmacokinetic Interaction Between Darunavir in Combination With Low-Dose Ritonavir and Buprenorphine/Naloxone
. Yardley, PA: Tibotec Inc. Presented at the 49th ICAAC; September 12-15, 2009; San Francisco, CA.
12. McCance-Katz EF, Moody DE, Morse GD, et al. Interaction between buprenorphine and atazanavir or atazanavir/ritonavir. Drug Alcohol Depend
13. Cheever LW, Kresina TF, Cajina A, et al. A model federal collaborative to increase patient access to buprenorphine treatment in HIV primary care. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S3-S6.
14. Weiss L, Egan JE, Botsko M, et al. A multi-site evaluation of integrated buprenorphine/naloxone and HIV treatment: overview of the BHIVES collaborative. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S7-S13.
15. Chaudhry AA, Botsko M, Weiss L, et al. Participant characteristics and HIV risk behaviors among individuals entering integrated buprenorphine/naloxone (bup/nlx) and HIV care. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S14-S21.
16. Cicconi P, Cozzi-Lepri A, Phillips A, et al. Is the increased risk of liver enzyme elevation in patients co-infected with HIV and hepatitis virus greater in those taking antiretroviral therapy? AIDS
17. Fiellin DA, Sullivan LE, Egan JE, et al. Drug treatment outcomes among HIV-infected, opioid-dependent patients receiving buprenorphine/naloxone. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S33-S38.
18. Altice FL, Lucas GM, Lum P, et al. HIV treatment outcomes among HIV-anfected, opioid-dependent patients receiving buprenorphine/naloxone treatment within HIV clinical care settings: results from a multisite study. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S22-S32.
19. Schackman BS, Leff J, Botsko M, et al. The cost of integrated HIV care and buprenorphine/naloxone treatment: results of a cross-site evaluation. J Acquir Immune Defic Syndr
. 2011;56(Suppl 1):S76-S82.
20. Bruce RD, Altice FL. Case series on the safe use of buprenorphine/naloxone in individuals with acute hepatitis C infection and abnormal hepatic liver transaminases. Am J Drug Alcohol Abuse
21. Chiang CN, Hawks RL. Pharmacokinetics of the combination tablet of buprenorphine and naloxone. Drug Alcohol Depend
22. Gross A, Jacobs EA, Petry NM, et al. Limits to buprenorphine dosing: a comparison between quintuple and sextuple the maintenance dose every 5 days. Drug Alcohol Depend
23. Petry NM, Bickel WK, Badger GJ. Examining the limits of the buprenorphine interdosing interval: daily, every-third-day and every-fifth-day dosing regimens. Addiction
24. Feinstein AR. The Architecture of Clinical Research
. Philadelphia, PA: W.B. Saunders Co; 1985.
APPENDIX I: BHIVES COLLABORATIVE
The CORE Center (Chicago, IL), El Rio Santa Cruz Neighborhood Health Center (Tucson, AZ), Johns Hopkins University (Baltimore, MD), Miriam Hospital (Providence, RI), Montefiore Medical Center (Bronx, NY), OASIS (Oakland, CA), Oregon Health Sciences University (Portland, OR), University of California San Francisco Positive Health Program at San Francisco General Hospital (San Francisco, CA), University of Miami Medical School (Miami, FL), Yale University School of Medicine (New Haven, CT) and The New York Academy of Medicine, (New York, NY).
Keywords:© 2011 Lippincott Williams & Wilkins, Inc.
buprenorphine; hepatitis; hepatotoxicity; HIV; HIV protease inhibitors; liver; opioid-related disorders