ORIGINAL RESEARCH. Introduction

Transcription

1 DOI: /j x HIV Medicine (2009), 10, ORIGINAL RESEARCH r 2009 British HIV Association Impact of a modified directly administered antiretroviral treatment intervention on virological outcome in HIV-infected patients treated in Burkina Faso and Mali CM Pirkle, 1 C Boileau, 2 V-K Nguyen, 3 N Machouf, 4 S Ag-Aboubacrine, 5 PA Niamba, 6 J Drabo, 7 S Koala, 8 C Tremblay 9 and S Rashed 10,11 1 Unité de Santé Internationale, Centre de Recherche du Centre Hospitalier de l Université de Montréal, Montréal, Canada, 2 Institute for Health and Social Policy, McGill University, Montréal, Canada, 3 Département de Médecine Sociale et Préventive, Université de Montréal, Montréal, Canada, 4 Clinique l Actuel, Montréal, Canada, 5 Unité de Médecine Interne, Hôpital National du Point G, Bamako, Mali, 6 Unité de Dermatologie, Centre Hospitalier Universitaire Yalgado-Ouédraogo, Ouagadougou, Burkina Faso, 7 Centre Hospitalier National Yalgado-Ouagadogou, Ouagadogou, Burkina Faso, 8 Projet Fonds Mondial de Lutte contre le SIDA, la Tuberculose et le Paludisme, Ouagadogou, Burkina Faso, 9 Département de Microbiologie et Immunologie, Centre de Recherche du Centre Hospitalier de l Université de Montréal, Montréal, Canada, 10 Unité de Pédiatrie, Hôpital Maisonneuve-Rosemont, Montréal, Canada and 11 SIDA-3 project, Mali Objective This study explores whether viral load measurements can be used in resource-limited settings to target those in need of adherence assistance. It was hypothesized that high plasma viral loads (pvls) ( 500 HIV-1 RNA copies/ml) were the result of poor antiretroviral therapy adherence and amenable to improvement with adherence assistance. Design A single-arm, multicentre pilot study was conducted from November 2003 to March 2004 on 606 treatment-experienced patients who had initiated an antiretroviral regimen in Mali and Burkina Faso 6 months before study enrolment. In these patients, those whose pvl was 500 copies/ml were offered 1 month of modified directly administered antiretroviral treatment (mdaart) with weekly follow-up visits from pharmacists or adherence counsellors. Methods An adherence questionnaire was given to all cohort patients and viral load was used to screen for patients with 500 copies/ml. mdaart participants included cohort patients with 500 copies/ml, who completed the adherence questionnaire. Genotypic analyses were conducted on samples taken prior to and after the intervention. The intervention was considered effective when there was a decrease of 1 log 10 in pvl. Results mdaart was effective in over one-third of the intervention participants, while in two-thirds no decrease in pvl was observed. The majority of mdaart participants had major resistance mutations. Conclusions pvl measurement was useful to identify patients who needed adherence assistance. However, because it was performed 6 months after starting treatment, mdaart came too late for most participants, as they had already developed important resistance mutations that might have been avoided with better laboratory monitoring. Keywords: adherence, antiretroviral therapy, drug resistance, HIV/AIDS, modified directly administered therapy, West Africa Accepted 12 August 2008 Correspondence: Catherine Pirkle, Unité de Santé Internationale, Université de Montréal, Édifice Saint-Urbain, 3875 rue Saint-Urbain, Montréal, QC, Canada H2W 1V1. Tel: ext 15927; fax: ; catherinepirkle@gmail.com Introduction Although the roll-out of antiretroviral treatment (ART) in sub-saharan Africa initially raised concerns about the 152

2 mdaart in West Africa 153 potential for antiretroviral anarchy [1], numerous successful ART programmes are now in operation across the continent with approximately a quarter of the 4.6 million people in need currently receiving treatment [2]. With initiatives such as the Global Fund and President s Emergency Plan for AIDS Relief (PEPFAR) and an estimated $8.3 billion investment in HIV treatment in resource-limited settings, critical debates have emerged over how best to deliver and scale up ART [3]. One such debate centres on the provision of laboratory technologies in resource-limited countries and around concerns that limited resources should be applied to prevention measures and the initiation of treatment, rather than the performance of expensive laboratory tests to monitor patients already receiving treatment [4]. As countries struggle to scale up ART delivery, there are concerns that adopting technologies such as viral load monitoring will over-tax already overdrawn laboratories and greatly limit the abilities of countries to expand treatment [5]. However, with the concerns around the rapid emergence of resistant viral strains, spelling doom for the individual, curtailing future treatment options, and leading to the transmission of the resistant virus [1, p. 410], these technologies may be essential to assuring ART sustainability. Meticulous adherence to treatment has been shown to be the most important factor in delaying the development of drug resistance [4]. It is also widely recognized as a crucial determinant of therapeutic success, as proper adherence is strongly correlated with virological and clinical outcome [4,6]. However, knowledge of effective strategies to increase adherence is limited, particularly in these settings [7]. Modified directly administered antiretroviral therapy (mdaart), in which the administration of one to two doses per day is witnessed, has been proposed as a strategy to improve adherence and treatment outcomes, as well as a public health strategy to prevent the development of drug resistance [6,8,9]. We describe an intervention using plasma viral load (pvl) measurements to target those in need of adherence assistance in a cohort of patients from Burkina Faso and Mali in West Africa. This intervention explores whether viral load measurements, in conjunction with mdaart, can be used in resource-limited settings to help prevent unnecessary switches to expensive second-line treatments. We had initial evidence from a cross-sectional study suggesting that adherence was less than optimal in this cohort. The assumption was that high pvl (4500 HIV-1 RNA copies/ml) was attributable either to poor adherence or to treatment failure and, because of previous evidence [8], adherence was believed to be the underlying problem. Materials and methods Patients and design In a multicentred cohort of patients recruited from three community (n 5 218) and three hospital-based (n 5 388) ART delivery sites in Ouagadougou (Burkina Faso) and Bamako (Mali), we conducted a single-arm pilot study from November 2003 to March In this study, we offered mdaart to patients whose pvl was 500 copies/ml. Eligible patients consisted of treatment-experienced cohort participants who had initiated a highly active antiretroviral therapy regimen at least 6 months before study enrolment. Cohort participants whose pvl was 500 copies/ml were offered 1 month of mdaart with weekly follow-up visits with pharmacists or adherence counsellors. The intervention was carried out by an accompagnateur (a family member, a friend, or a health care professional) chosen by the patient. All mdaart providers were given instructions on how to monitor the doses using a follow-up chart. Nonmedical providers were given further information on treatment regimen. Data extraction Adherence questionnaire Trained interviewers collected data on demographic characteristics (age, gender, marital status, literacy, country of residence and distance from health centre), adherence behaviour (adherence in the last week, treatment interruption and knowledge of ART), treatment (currently prescribed and previous ART, and time on current ART) and clinical status [CD4 cell counts, body mass index (BMI: kg/m 2 ) and self-reported health] using a close-ended questionnaire and patient medical files. CD4 cell counts, viral load, and genotypic analysis All measurements were taken prior to and after the intervention. CD4 cells were quantified by flow cytometry using the Becton Dickinson FACSCaliburt system (Becton- Dickinson, Mountainview, CA, USA) in Ouagadougou and FACScan (BD Biosciences, San Jose, CA, USA) in Bamako. Viral load was measured in Montréal, Canada, using the COBAS Amplicor HIV-1 Monitor Test, version 1.5 (Roche Diagnostics, Branchburg, NJ, USA), according to the manufacturer s instructions. The methods used for the genotypic analysis have been described in detail elsewhere [9]. Study outcome The primary study outcome was the log 10 difference in pvl following the 1-month mdaart intervention. We considered

3 154 CM Pirkle et al. the intervention effective when there was a decrease of at least 1 log 10 in pvl following the intervention. Table 1. Baseline characteristics of the cohort patients vs. modified directly administered antiretroviral treatment (mdaart) participants Results Cohort patients (n 5 550) mdaart participants (n 5 56) P-value Patient characteristics Viral load testing was used to screen 606 patients who had received ART for 6 months at six sites in Burkina Faso and Mali. Eighty-five patients had pvl 4500 copies/ml (range copies/ml) and were considered eligible for the mdaart intervention. Fifty-six of these patients agreed to, and completed, both the intervention and adherence questionnaire. The overwhelming reason for nonparticipation in the intervention was an inability to find an accompagnateur. The 29 eligible nonparticipants differed significantly from their mdaart counterparts in terms of BMI, pvl and having ever interrupted treatment. Compared with eligible, but nonparticipating cohort patients, mdaart participants had a higher BMI (22.85 vs ; P ) and a lower pvl (4.18 vs. 4.46; P ), and were more likely to have interrupted treatment (79.4 vs. 20.6%; P ). Characteristics of the 56 mdaart participants, as well as patients from the remainder of the cohort, at initial pvl screening are presented in Table 1. Compared with the remainder of the cohort, there was a borderline-significantly higher proportion of men among the mdaart participants (44.6 vs. 32.0%; P ); for all other sociodemographic variables, there were no significant differences between groups. In accordance with their virological status, mdaart participants were more immunologically suppressed (CD4o200 cells/ml) and less likely to report good health. Concerning adherence-related behaviours, mdaart participants were more likely to report treatment interruptions and less likely to have good ART knowledge. Finally, mdaart participants had been on treatment for longer and were more likely to have been on a protease inhibitor (PI)-containing or other type of regimen compared to non-mdarrt participants. There was a borderline-significant difference in the proportions of mdaart participants and cohort patients on second-line treatments (21.8 vs. 12.8%, respectively; P ). Impact of mdaart on virological outcomes >Genotyping was performed on 34 of the 56 patients participating in mdaart. The remaining 22 samples could not be sequenced because of sample degradation. Figure 1 presents the effects of mdaart on pvl in genotyped and nongenotyped participants, according to resistance status. Major nucleoside reverse transcriptase inhibitor (NRTI) resistance mutations were M184V (85.7%), Thymidine Sociodemographics Country [n (%)] Burkina Faso 296 (53.8) 27 (48.2) Mali 254 (46.2) 29 (51.8) Gender [n (%)] Male 176 (32.0) 25 (44.6) Female 374 (68.0) 31 (55.4) Age (years) [mean (SD)] 38.0 (7.2) 38.0 (7.9) 0.96 Marital status [n (%)] Married 243 (44.2) 29 (51.8) Single 112 (20.4) 14 (25.0) Divorced/widow 195 (35.5) 13 (23.2) French literacy [n (%)]* 349 (63.8) 38 (67.9) Distance from health centre [n (%)] o1 h 421 (76.5) 38 (67.9) h 129 (23.5) 18 (32.1) Clinical characteristics Perceived good health [n (%)] 318 (57.9) 19 (33.9) o0.001 BMI [mean (SD)] w 23.0 (4.56) (4.58) Immunological suppression [n (%)] z 122 (24.3) 22 (41.5) Log 10 pvl (copies/ml) [mean (SD)] 1.88 (0.65) 4.18 (0.64) o0.001 Adherence [n (%)] Ever interrupted treatment 108 (19.6) 27 (48.2) o0.001 Perfect last week adherence 373 (66.5) 31 (55.4) Good knowledge of ART 145 (26.4) 9 (16.1) 0.05 Treatment Time on ART (months) [mean (SD)] 19.8 (10.7) 23.7 (15.1) Second-line treatment [n (%)] 70 (12.8) 12 (21.8) Prescribed treatment [n (%)] NNRTI-containing 459 (85.0) 40 (71.4) o0.001 PI-containing 79 (14.6) 13 (23.2) Other 2 (0.4) 3 (5.4) *Knows how to read and write in French. w Body mass index was calculated as weight (kg) divided by the square of height (m) (kg/m 2 ). z CD4o200 cell/ml. Based on nonmissing data only. ART, antiretroviral therapy; BMI, body mass index; NNRTI, nonnucleoside reverse transcriptase inhibitor; PI, protease inhibitor; pvl, plasma viral load; SD, standard deviation Fig. 1. The impact of modified directly administered antiretroviral treatment (mdaart) on plasma viral load according to resistance status.

4 mdaart in West Africa 155 Analog Mutation (TAM)-2 (42.9%) and TAM-1 (28.6%). Major nonnucleoside reverse transcriptase inhibitor (NNRTI) resistance mutations were K103N (57.1%), Y181C (32.1%) and G190A/S (22.2%). See Sylla et al. [9] for a more detailed description of resistance mutations detected in the cohort (including mdaart participants and nonparticipants). Viral load decreased by at least 1 log 10 in one-third of mdaart participants (n 5 20; 35.7%), while in two-thirds no decrease was observed (n 5 36; 64.3%). However, among those not genotyped, the intervention was effective in over 70% of participants, probably reflecting their lower average initial pvl (pvl 3.79 vs log 10 copies/ml; see Fig. 1). Of the 34 genotyped participants, 30 had major resistance mutations. In the majority of resistant participants, there was no diminution in pvl post-mdaart (n 5 27; 90%); in four participants, viral load increased by more than 1 log 10. Among genotyped participants, 78.6% had both NRTI and NNRTI mutations and 76.7% had three or more mutations. Of the three resistant participants who reduced their pvl by 1, none had dual NRTI/NNRTI mutations and all had fewer than three mutations (results not shown). These patients had the following major resistance mutations: one patient with 184V one with K103N and p225h, and one with K103N only. Except for an ART regimen containing nevirapine (NVP) (P ), no demographic or clinical variables were associated with a 1 log 10 decrease in pvl post-mdaart. In those taking NVP, 13% decreased their pvl by 1 log 10 compared with 44% in those on a regimen not containing NVP; this association was not significant when sex and CD4 cell count were taken into consideration. Discussion An important consideration in the management of patients on ART will be the level and quality of laboratory monitoring necessary to retain effective and sustainable treatment options, particularly with increasing concerns over widespread drug resistance [7]. Laboratory services are one of the most neglected areas of health provision in sub- Saharan Africa, engendering a reliance on clinical symptoms to determine diagnosis and treatment [4]. Recent evidence suggests that an exclusive focus on clinical symptoms has led to inappropriate treatment, increased morbidity and unnecessary loss of life [10 12]. In the case of ART, it may also lead to expensive and unnecessary changes to second-line treatments and/or the prolongation of ineffective first-line therapies. The goal of this intervention was to employ viral load monitoring and mdaart to weed out those who needed adherence assistance from those who needed second-line treatments. The majority of cohort participants (74.9%) were able to bring their pvl below 50 copies/ml [9]; nonetheless, previous evidence suggested that poor adherence was a problem in the cohort [8] and largely responsible for the high viral loads observed in mdaart participants, and we thus expected the intervention to be effective in most participants [8]. Our results, however, painted a more complex picture. As the majority of participants harboured major resistance mutations, our results suggest that, while adherence may have been the underlying problem, our intervention came too late for most participants. Genotyping revealed that resistance was already a problem upon mdaart initiation [9]. Certain baseline characteristics (mdaart participants had been on ART for significantly longer than their cohort counterparts and were significantly more likely to be taking an unboosted PI or other regimen compared to non mdarrt participants) hinted that resistance was the underlying reason for high pvl, as these factors have been associated with drug resistance in the literature [13]. Better laboratory monitoring may have prevented the high rates of resistance found in our study by detecting those with elevated pvls early enough for adherence assistance to be beneficial. One-third of mdaart participants did decrease their pvl by more than 1 log 10. In these participants, the mdaart intervention may have prevented unnecessary treatment switches. As second-line treatments cost fourtimes more than first-line treatments in Burkina Faso and Mali, this represents a significant saving in terms of potential treatment costs. However, as most samples in these patients were not genotyped, we do not know whether these patients were harbouring important resistance mutations that would eventually compromise their treatment regimen. Conclusion Our results underscore the importance of improving the laboratory capacities of resource-limited countries. As this intervention demonstrated, adherence assistance came too late for most participants, as they had already developed important resistance mutations that might have been avoided with more regular laboratory monitoring. The emergence of resistance, which could compromise future treatment, justifies investment in technologies such as viral load monitoring, to say nothing of the inherent benefits associated with building local laboratory capacity. Acknowledgement This study was supported by grants from the Canadian Institutes for Health Research (CIHR/IRSC).

5 156 CM Pirkle et al. References 1 Harries AD, Nyangulu DS, Hargreaves NJ, Kaluwa O, Salaniponi FM. Preventing antiretroviral anarchy in sub-saharan Africa. Lancet 2001; 358: UNAIDS. Fact sheet: Sub-Saharan Africa unaids.org/pub/globalreport/2006/ fs_subsaharan Africa_en.pdf (accessed 30 March 2008). 3 Calmy A, Ford N, Hirschel B et al. HIV viral load monitoring in resource-limited regions: optional or necessary? Clin Infect Dis 2007; 44: Koenig SP, Kuritzkes DR, Hirsch MS et al. Monitoring HIV treatment in developing countries. BMJ 2006; 332: Harries AD, Schouten EJ, Libamba E. Scaling up antiretroviral treatment in resource-poor settings. Lancet 2006; 367: Kagay CR, Porco TC, Liechty CA et al. Modeling the impact of modified directly observed antiretroviral therapy on HIV suppression and resistance, disease progression, and death. Clin Infect Dis 2004; 38 (Suppl. 5): S414 S Jaffar S, Govender T, Garrib A et al. Antiretroviral treatment in resource-poor settings: public health research priorities. Trop Med Int Health 2005; 10: Aboubacrine SA, Niamba P, Boileau C et al. Inadequate adherence to antiretroviral treatment and prevention in hospital and community sites in Burkina Faso and Mali: a study by the ATARAO Group. Int J STD AIDS 2007; 18: Sylla M, Chamberland A, Boileau C et al. Characterization of drug resistance in a cohort of subjects on antiretroviral therapy infected with HIV-1 CRF02_AG and AGK subtypes in Mali and Burkina Faso. Antivir Ther; 13: Bates I, Maitland K. Are laboratory services coming of age in sub-saharan Africa? Clin Infect Dis 2006; 42: Petti CA, Polage CR, Quinn TC et al. Laboratory medicine in Africa: a barrier to effective health care. Clin Infect Dis 2006; 42: Coutinho A, Mermin J, Ekwaru J et al. Utility of routine viral load, CD4 cell count, and clinical monitoring among HIVinfected adults in Uganda: a randomized trial. CROI Boston, USA; February Bangsberg DR, Charlebois ED, Grant RM et al. High levels of adherence do not prevent accumulation of HIV drug resistance mutations. AIDS 2003; 17: