Long-term outcomes of protease inhibitor-based therapy in antiretroviral treatment-naive HIV-infected injection drug users on methadone maintenance programmes

Abstract

The effectiveness of protease inhibitor (PI)-based highly active antiretroviral treatment (HAART) relies on a strong commitment to adhere to complex therapies with potentially adverse effects [1,2]. Although methadone maintenance programmes (MMP) reduce drug abuse-related complications and improve the compliance of HIV-positive intravenous drug users (IDU) with the monitoring and treatment of their infection [3,4], the belief that these individuals are poorly adherent, and pharmacokinetic concerns in patients on methadone have led to the suboptimal use of PI-based therapy in this population [5–8]. To date, there are few reports of outcomes in observational cohorts using HAART in IDU [9]. Between March 1996 and May 1998, we prospectively studied all antiretroviral treatment (ART)-naive patients (n = 208) starting therapy at a specialized oupatient clinic from a tertiary centre in Madrid, Spain. The observational nature of this study allowed us to evaluate the one year effectiveness of PI-based HAART in IDU on methadone (n = 54) in comparison with the overall population of non-methadone individuals (n = 154). Table 1 contains a summary of the univariate analysis of baseline features and outcomes according to methadone use in the cohort. The great majority of methadone patients (85%) were already on a MMP before starting ART (median time 5.5 months; range 1–44). At entry, the only significant difference between the groups was a longer time from AIDS diagnosis in methadone patients (27 versus 2 months, P = 0.0001), whereas there were no differences according to age, sex, or prescribed PI, indinavir being the most frequently used (83 versus 80% in MMP patients versus non-MMP patients, P = 0.61). After 12 months, both groups obtained similar clinical, virological and immunological outcomes, despite at least 30 days of continued drug interruption being more frequently reported by methadone patients (41 versus 18%, P = 0.001). During the follow-up period, four (7%) subjects from the MMP group had a recurrence of intravenous heroin abuse, whereas six (11%) reported other substance abuse, mostly cocaine (83%). Of note was the fact that regardless of the recurrence of drug abuse or the adherence level with ART, 75% of methadone subjects maintained their regular visits to the AIDS clinic, and after a median follow-up time of 546 days [95% confidence interval (CI) 517–574], the overall mortality rate was 0.9%, exclusively among non-methadone patients.Table 1: Univariate analysis of baseline features and outcomes according to methadone use (208 patients). Using multivariate analysis, methadone was not a significant factor predicting a worse long-term virological outcome. The lack of AIDS diagnosis at baseline [relative risk (RR) 0.35; 95% CI 0.14–0.89;P = 0.02] and adherence over 90% (RR 16.66; 95% CI 5.5–50;P = 0.0001) were the strongest predictors for long-term virological success. The results of our study support the usefulness of methadone for the adequate management of HIV infection in active IDU, including the use of the most optimal ART [1]. In contrast with previous reports [9,10], we did not find significant differences in the rates of virological or immunological success achieved by methadone users when compared with naive patients with similar baseline features starting a potent PI-based therapy. A remarkable finding was the significantly longer time from AIDS diagnosis in methadone patients in our cohort. This probably reflects a higher rate of progression of HIV infection in IDU before MMP enrolment, with a lifestyle that did not allow timely prophylaxis or treatment for opportunistic diseases [4], and also that physicians may be less likely to recommend ART for these patients because of the belief that they are less adherent to complex regimens and have a high rate of comorbidities, potential drug interactions or poorer tolerability [11–13]. Adherence has arisen as the ‘Achilles’ heel’ of potent ART [14]. In our cohort, most patients maintained an adequate adherence level regardless of methadone use, but a significantly greater proportion of methadone users reported drug interruptions of at least 30 consecutive days. Other drug abuse maintenance therapies, such as buprenorphine, have been able to increase the adherence to HAART among this population [15]. The association between poor adherence, virological failure and resistance has been clearly established [16], and therefore it was difficult to explain why it did not translate into a worse long-term prognosis in our methadone patients. Previous reports [17] suggested a lack of deleterious effect on reinitiated therapy in patients who temporarily discountinued ART. If a successful therapy is completely and abruptly interrupted, the absence of drug selection applied on the virus could avoid the emergence of resistant strains. However, an extended treatment interruption might erase the benefits of a long period under HAART and bring the patient back to the pre-treatment starting point. Active efforts should thus be made to improve adherence further in these patients, through a trusting and accessible relationship with experienced physicians in charge of HIV-infected IDU [18] along with adequate support services [19]. In summary, our results showed that in the population of IDU, current stable enrolment in a MMP could allow the adequate management of HIV infection, and that methadone patients are able to reach similar outcomes to the overall ART-naive population after the administration of potent PI-based therapy. However, their adherence to ART should be supported by strict monitoring and timely collaboration with primary care services. Acknowledgements The authors would like to thank Dolores Lopez and Miguel A. Huerta for their assistance and technical support. This article is dedicated to Drs Luis Buzón, Pedro Montilla, and Jose M. Hermida, for their personnal efforts in the implementation of methadone maintenance programmes for injection drug users and HIV-infected IDU in our community. Ana Moreno María J. Perez-Elías Jose L. Casado Vicente Muñoz Antonio Antela Fernando Dronda Enrique Navas Santiago Moreno