Dear Editor:
Management of HIV is complicated by the presence of hepatitis C (HCV) infection.1 The recent addition of first-generation nonstructural protein 3/4A (NS3/4A) serine protease inhibitors, such as boceprevir and telaprevir, has significantly enhanced the probability of achieving a sustained virologic response (SVR).2 Recently, simeprevir and sofosbuvir have been incorporated into treatment guidelines,3 but availability and uptake in foreign markets is a long process. Therefore, patients with HIV/HCV infection in developing countries may still rely on treatment with first-generation NS3/4A protease inhibitors.
Despite the availability boceprevir and telaprevir, their uptake has been limited by concerns about cost/reimbursement, toxicity, and the risk of drug–drug interactions. Of these, the latter has been among the clinical concerns limiting the use of NS3/4A protease inhibitors among coinfected patients.
The objectives of this study were to (1) quantify the prevalence of clinically significant drug–drug interactions (CSDDI) when boceprevir-containing triple HCV therapy (BCTT) is added to a patient's medication profile, and (2) determine the predictors of contraindicated drug–drug interactions. This study focused on boceprevir because it has gained licensure in a significant number of developing countries where HCV coinfection is a concern, and is likely to be used.4
A cross-sectional study was performed among Veterans' Affairs Medical Center (VAMC) patients receiving care in the Upstate New York Veterans' Healthcare Administration (VISN-2) between January 1, 2000 and July 31, 2011. Inclusion criteria for this study were: (1) age ≥18 years, (2) documented HIV infection, and (3) laboratory-confirmed HCV infection. Patients with no medication history were excluded.
From the patients' medical records we extracted social history, demographics (age, sex, race, height, and weight), and clinical data (years since HCV and HIV diagnosis, CD4 cell counts, comorbidities, and medication lists). Further details on the variables that were collected are described elsewhere.5
The outcomes of interest in this study were the presence of (1) CSDDI, and (2) contraindicated drug–drug interactions. Drug–drug interactions were identified by Lexi-Interact drug interaction software. The rating system in Lexi-Interact was used to define which drug–drug interactions were considered (1) clinically significant and (2) contraindicated. Interactions rated by Lexi-interact with a level D or X severity rating were considered clinically significant; those rated only X were considered contraindicated drug–drug interactions.
To determine the effect of BCTT on the excess risk of CSDDI and contraindicated drug–drug interactions, boceprevir, pegylated interferon α, and ribavirin were added to the patients' medications lists in Lexi-Interact, and reanalyzed. Any additional D- or X-rated interactions were documented.
For the bivariate analyses, categorical variables were compared by χ2 or Fisher's exact tests, and continuous variables were compared by Student's t or Mann–Whitney U tests. Classification and regression tree (CART) analysis was used to identify break points in continuous variables. McNemar's test was used to compare the frequency of outcome before and after the addition of boceprevir-containing HCV therapy.
Variables in the bivariate analyses that were associated (p<0.2) with contraindicated drug–drug interactions were considered for inclusion in the explanatory multivariate log-binomial regression model with robust variance estimates.6 All calculations were computed using SPSS version 11.5 (SPSS Inc., Chicago, IL), SAS version 9.3 (Cary, NC), and CART software (Salford Systems, San Diego, CA).
The mean (standard deviation, SD) age of the 244 patients was 57.4 (6.1) years. The median [interquartile range (IQR)] durations of HIV and HCV infections were 19 (15–24) and 15 (12–19) years, respectively. The most common antiretroviral therapy (ART) regimen types were non-nucleoside reverse transcriptase inhibitor (NNRTI)-based (37.7%) and protease inhibitor (PI)-based (36.9%). The median (IQR) numbers of non-HIV medications and CSDDI per patient were 8 (5–11) and 3 (1–6), respectively.
The frequency of CSDDI (D- and X-rated interactions) significantly increased after the addition of BCTT from 85.7% (n=209) to 97.5% (n=238), p<0.001. Similarly, the prevalence of contraindicated drug–drug interactions (X-rated interactions) increased from 20.9% (n=51) to 73.9% (n=180), p<0.001, after the addition of BCTT. Clinically significant drug–drug interactions involving ART increased from 74.6% (n=182) to 84.8% (n=207), p<0.001 and contraindicated drug–drug interactions involving ART changed from 6.6% (n=16) to 77.9% (n=190), p<0.001, after the addition of BCTT. In a restricted analysis of patients only receiving ART (n=222) the addition of BCTT would have resulted in statistically (p<0.05) and clinically significant increases in contraindicated drug–drug interactions for mixed class ART regimens (18.9–67.6%), NNRTI (17.4–84.8%), and PI (25.6–83.3%).
The bivariate analyses of demographic and clinical characteristics associated with a contraindicated drug–drug interaction after the addition of BCTT are described in Table 1. Drug classes with higher frequencies of contraindicated drug–drug interactions were nucleoside reverse transcriptase inhibitors (n=162, 90.0%), NNRTIs (n=89, 49.4%), PIs (n=93, 51.7%), and erectile dysfunction drugs (n=51, 28.3%). The results of the multivariate regression analyses are displayed in Table 2. The only variables that were independently associated with contraindicated drug–drug interactions were atazanavir, darunavir, lopinavir, and efavirenz, and use of eight or more medications (CART-derived break point).
Table 1.
Bivariate Analyses of Clinical Covariates Associated with Contraindicated Drug–Drug Interactions Involving Boceprevir-Containing Hepatitis C (HCV) Therapy
| Covariate | No contraindicated drug–drug interaction (n=64) | Contraindicated interaction present (n=180) | p Value |
|---|---|---|---|
| Demographic characteristics | |||
| Age, mean±SD | 58.1 (7.0) | 57.2 (5.8) | 0.32 |
| Age≥71 yearsa | 4 (6.3) | 0 (0) | 0.004 |
| Male sex | 62 (96.9) | 177 (98.3) | 0.61 |
| Race | 0.91 | ||
| African American | 44 (69.8) | 117 (65.0) | |
| Hispanic | 6 (9.5) | 18 (10.0) | |
| White | 12 (19.0) | 42 (23.3) | |
| Other | 1 (1.6) | 3 (1.7) | |
| History of HIV and HCV infections | |||
| Duration of HIV infection, median [interquartile range (IQR)] | 20 (15–25) | 19 (14–24) | 0.39 |
| Duration of HIV≥8 yearsa | 64 (100) | 165 (91.7) | 0.01 |
| Duration of HCV infection, median (IQR) | 14 (13–19) | 15 (12–19) | 0.88 |
| Most recent CD4 count, median (IQR) | 515 (328–690) | 395 (18–621) | 0.01 |
| Most recent log-transformed HIVRNA, mean±SD | 2.65 (1.14) | 2.46 (1.04) | 0.24 |
| Antiretroviral medication use | |||
| Receiving antiretroviral therapy | 43 (67.2) | 179 (99.4) | <0.001 |
| Antiretroviral regimen type | 0.02 | ||
| Non-nucleoside reverse transcriptase inhibitor | 14 (32.6) | 78 (43.6) | |
| Protease inhibitor | 15 (34.9) | 75 (41.9) | |
| Integrase strand transfer inhibitor | 2 (4.7) | 1 (0.6) | |
| Other/mixed class regimen | 12 (27.9) | 25 (14.0) | |
| Use of nucleoside reverse transcriptase inhibitors | 41 (64.1) | 162 (90.0) | <0.001 |
| Zidovudine | 17 (26.6) | 35 (19.4) | 0.23 |
| Lamivudine | 33 (51.6) | 66 (36.7) | 0.04 |
| Emtricitabine | 6 (9.4) | 78 (43.3) | <0.001 |
| Tenofovir | 12 (18.8) | 93 (51.7) | <0.001 |
| Abacavir | 5 (7.8) | 22 (12.2) | 0.33 |
| Didanosine | 2 (3.1) | 11 (6.1) | 0.52 |
| Stavudine | 12 (18.8) | 19 (10.6) | 0.09 |
| Use of non-nucleoside reverse transcriptase inhibitor | 17 (26.6) | 89 (49.4) | 0.002 |
| Efavirenz | 0 (0) | 83 (46.1) | <0.001 |
| Nevirapine | 16 (25.0) | 3 (1.7) | <0.001 |
| Rilpivirine | 1 (1.6) | 0 (0) | 0.26 |
| Etravirine | 0 (0) | 3 (1.7) | 0.57 |
| Use of protease inhibitor | 18 (28.1) | 93 (51.7) | 0.001 |
| Atazanavir | 0 (0) | 33 (18.3) | <0.001 |
| Darunavir | 0 (0) | 13 (7.2) | 0.02 |
| Fosamprenavir | 0 (0) | 5 (2.8) | 0.33 |
| Lopinavir | 0 (0) | 40 (22.2) | <0.001 |
| Saquinavir | 3 (4.7) | 2 (1.1) | 0.12 |
| Indinavir | 4 (6.3) | 0 (0) | 0.004 |
| Nelfinavir | 11 (17.2) | 3 (1.7) | <0.001 |
| Ritonavir | 2 (3.1) | 82 (45.6) | <0.001 |
| Integrase inhibitor | 2 (3.1) | 18 (10.0) | 0.11 |
| Raltegravir | 2 (3.1) | 18 (10.0) | 0.11 |
| Enfuvirtide | 0 (0) | 2 (1.1) | 1.00 |
| Non-HIV medication usage | |||
| Total number of medications, median (IQR) | 10 (6–13) | 12 (9–15) | 0.01 |
| Use of≥8 medicationsa | 39 (60.9) | 152 (84.4) | <0.001 |
| Number of non-HIV medications, median (IQR) | 8 (5–10) | 8 (6–12) | 0.17 |
| Antipsychotics | 16 (25.0) | 34 (18.9) | 0.30 |
| Antibiotics | 4 (6.3) | 7 (3.9) | 0.49 |
| Antidepressants | 17 (26.6) | 62 (34.4) | 0.25 |
| Azole antifungals | 2 (3.1) | 7 (3.9) | 1.00 |
| Statins | 3 (4.7) | 22 (12.2) | 0.10 |
| Calcium channel blockers | 6 (9.4) | 35 (19.4) | 0.08 |
| Central nervous system depressants | 31 (48.4) | 95 (52.8) | 0.55 |
| Erectile dysfunction drugs | 4 (6.3) | 51 (28.3) | <0.001 |
| Antiarrhythmic drugs | 3 (4.7) | 3 (1.7) | 0.19 |
| Corticosteroids | 11 (17.2) | 33 (18.3) | 0.84 |
| Comorbidities | |||
| Median (IQR) number of comorbidities | 8 (6–11) | 10 (7–13) | 0.01 |
| ≥11 comorbidities | 20 (31.3) | 87 (48.3) | 0.02 |
| Alcoholism | 29 (45.3) | 83 (46.1) | 0.91 |
| Anxiety | 3 (4.7) | 20 (11.1) | 0.21 |
| Bipolar/mood/personality disorder | 11 (17.2) | 23 (12.8) | 0.38 |
| Cardiovascular disease | 6 (9.4) | 22 (12.2) | 0.54 |
| Chronic kidney disease | 5 (7.8) | 12 (6.7) | 0.76 |
| Chronic obstructive pulmonary disease | 8 (12.5) | 19 (10.6) | 0.67 |
| Depression | 24 (37.5) | 83 (46.1) | 0.23 |
| Diabetes | 7 (10.9) | 31 (17.2) | 0.23 |
| Dyslipidemia | 3 (4.7) | 41 (22.8) | 0.001 |
| Epilepsy/seizures | 3 (4.7) | 9 (5.0) | 1.00 |
| Gastroesophageal reflux disease | 7 (10.9) | 22 (12.2) | 0.79 |
| Hepatitis B infection | 3 (4.7) | 20 (11.1) | 0.21 |
| History of cancer | 9 (14.1) | 20 (11.1) | 0.53 |
| History of substance abuse | 47 (73.4) | 124 (68.9) | 0.50 |
| Hypertension | 28 (43.8) | 77 (42.8) | 0.89 |
| Insomnia | 5 (7.8) | 18 (10.0) | 0.61 |
| Neuropathy | 8 (12.5) | 23 (12.8) | 0.95 |
| Osteoporosis/osteoarthritis/osteogenesis | 10 (15.6) | 20 (11.1) | 0.35 |
| Post-traumatic stress disorder | 12 (18.8) | 31 (17.2) | 0.78 |
| Prostate disease | 5 (7.8) | 24 (13.3) | 0.24 |
| Schizophrenia/psychosis | 7 (10.9) | 20 (11.1) | 0.97 |
All data presented as n (%), mean (standard deviation), or median (interquartile range), unless noted otherwise.
Classification and regression tree (CART)-derived break point
Table 2.
Clinical Covariates Independently Associated with the Presence of Contraindicated Drug–Drug Interactions Involving Boceprevir-Based Hepatitis C (HCV) Therapy
| Covariate | Adjusted prevalence ratio | 95% Confidence interval | p Value |
|---|---|---|---|
| Atazanavir | 3.77 | 2.73–5.19 | <0.001 |
| Darunavir | 3.27 | 2.17–4.95 | <0.001 |
| Lopinavir | 3.17 | 2.20–4.56 | <0.001 |
| Efavirenz | 3.73 | 2.69–5.17 | <0.001 |
| Use of ≥8 medications | 1.23 | 1.01–1.50 | 0.04 |
We observed a number of interesting findings in this study. First, the HIV/HCV coinfected population was highly vulnerable to drug–drug interactions when boceprevir-containing HCV therapy was added to the medication regimen. Second, the two most important drivers of contraindicated drug–drug interactions were polypharmacy (use of at least eight medications), and four specific ART medications (efavirenz, lopinavir, atazanavir, and darunavir).
Regarding polypharmacy, treatment of HIV infection alone requires the use of three or more medications and several others to manage other comorbidities in an aging population.7 With respect to the four specific ART medications likely to increase the risk of contraindicated drug interactions, three are considered the preferred ART drugs (atazanavir, darunavir, and efavirenz).8 More interactions may exist with these agents in the general population because their use is ubiquitous in clinical practice. Other preferred ART medications may be considered in patients who are likely candidates for HCV treatment.3
Some considerations should be noted. First, the HCV guidelines have recently been updated, and no longer advocate the use of boceprevir or telaprevir.3 The issue of drug interactions is still pertinent to simeprevir, and to a lesser extent, sofosbuvir (package inserts). However, boceprevir and telaprevir are still available, and will likely to be relegated to resource-limited areas where prompt therapy is needed. Second, patients in the study did not receive BCTT, and the exposure of interest was theoretical. The theoretical design allowed us to complete the study more quickly than if we had waited for the same number of coinfected patients to initiate BCTT in clinical practice. A high frequency of CSDDI may be expected for simeprevir, given the similarities in metabolic pathways (package insert). Third, we recognize that boceprevir is only indicated for patients with genotype 1 HCV infection. We did not use genotype as an entry criterion in this study because it was largely unavailable (documented in less than one third of the medical charts). However, the majority of patients would be expected to have genotype 1 infection.9 Additionally, the information bias that non-genotype 1 patients would introduce would be negligible, because there is no inherent reason to anticipate genotypic heterogeneity of medication use patterns. Fourth, these findings may underrepresent the magnitude of the issue, as adherence to screening guidelines is imperfect, and more patients may be affected.10 Finally, BCTT involves the use of pegylated interferon and ribavirin, and these medications can have their own drug–drug interactions, and this could upwardly drive up the number of CSDDI. This was not the case in this study. In a post-hoc analysis, all patients who had an interaction involving pegylated interferon or ribavirin also had interactions involving boceprevir.
Clinicians have a particularly important role in the evaluation of therapeutic agents for HCV against their country's unique backdrop of disease epidemiology, reimbursement, and potential drug interactions. The addition of BCTT in patients with HIV/HCV coinfection was associated with a high prevalence of drug–drug interactions, and was predicted by polypharmacy and specific ART medications. Future research should investigate the population-based risk of drug–drug interactions with emerging HCV agents, and whether these findings would be more pronounced in the developing world, where BCTT is most likely to be utilized.
Author Disclosure Statement
No competing financial interests exist.
References
- 1.Rotman Y, Liang TJ. Coinfection with hepatitis C virus and human immunodeficiency virus: Virological, immunological, and clinical outcomes. J Virol 2009;83:7366–7374 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mandorfer M, Payer BA, Niederecker A, et al. Therapeutic potential of and treatment with boceprevir/telaprevir-based triple-therapy in HIV/chronic hepatitis C co-infected patients in a real-world setting. AIDS Patient Care STDs 2014;28:221–227 [DOI] [PubMed] [Google Scholar]
- 3.American Association for the Study of Liver Diseases, Infectious Diseases Society of America and International AIDS Society–USA. Recommendations for testing, managing, and treating hepatitis C. http://www.hcvguidelines.org (Last accessed May15, 2014)
- 4.Sievert W, Altraif I, Razavi HA, et al. A systematic review of hepatitis C virus epidemiology in Asia, Australia and Egypt. Liver Int 2011;31Suppl 2:61–80 [DOI] [PubMed] [Google Scholar]
- 5.Patel N, Veve M, Bliss S, et al. Prevalence and predictors of important telaprevir drug interactions among patients coinfected with hepatitis C and human immunodeficiency virus. J Pharm Technol 2014:1–9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McNutt LA, Wu C, Xue X, Hafner JP. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol 2003;157:940–943 [DOI] [PubMed] [Google Scholar]
- 7.Tseng A, Szadkowski L, Walmsley S, Salit I, Raboud J. Association of age with polypharmacy and risk of drug interactions with antiretroviral medications in HIV-positive patients. Ann Pharmacother 2013;47:1429–1439 [DOI] [PubMed] [Google Scholar]
- 8.Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services. Available at http://aidsinfo.nih.gov/contentfiles/lvguidelines/AdultandAdolescentGL.pdf (Last accessed May15, 2014)
- 9.Nainan OV, Alter MJ, Kruszon-Moran D, et al. Hepatitis C virus genotypes and viral concentrations in participants of a general population survey in the United States. Gastroenterology 2006;131:478–484 [DOI] [PubMed] [Google Scholar]
- 10.Jonckheere S, Vincent A, Belkhir L, Wilmes D, Vandercam B, Yombi JC. Adherence to screening guidelines for hepatitis C among HIV-positive patients. AIDS Patient Care STDs 2013;27:317–319 [DOI] [PMC free article] [PubMed] [Google Scholar]