Antiretroviral Treatment in Developing Countries By Jean Nachega, M.D., M.P.H.

• HIV/AIDS Prevention vs Treatment
• HAART Pilot Programs in the Public Sector
• HAART Pilot Programs in the Private Sector
• Treatment Adherence
• Neonatal PEP with NVP and/or AZT for Prevention of Mother-to-Child-Transmission of HIV
• Simple and Affordable Diagnostic and Monitoring Lab Tests
• An AIDS Activist’s Voice
• Conclusion

HIV/AIDS Prevention vs Treatment

The debate over prevention versus treatment efforts took center stage at the XIV International AIDS Conference in Barcelona, Spain, July 7-12, 2002. In the background behind the debates were two controversial papers recently published by Marseille and colleagues [Lancet 2002;359:1851] and Creese and colleagues [Lancet 2002;359:1635], both of which used a narrow, cost-effectiveness analysis that suggested that prevention of HIV/AIDS should take priority over treatment in Africa. At the Barcelona conference, most attendees seemed to agree that this false dichotomy between prevention and treatment should be abandoned. Not only did the analyses by Marseille and Creese offer a static perspective on drug costs, which in reality are a rapidly moving target and may decline with time, but such analyses also fail to consider the impact of treatment on preventing transmission, or the positive impact treatment has on national economic development. Many delegates felt that the decision to treat is a humanitarian and ethical one that cannot be based solely on cost-effectiveness data. The prevailing sentiment, from eminent HIV/AIDS experts and activists, was that the lives of 28 million HIV infected Africans do matter and that debating prevention versus treatment misstates the problem. Although many African govern-ments have focused almost exclusively on prevention, the HIV infection rate in these countries has continued to rise, and growing numbers of experts are saying that HIV/AIDS can only be fought effectively with programs that link prevention and treatment.

HAART Pilot Programs in the Public Sector

AIDS researchers in developing countries have started to report data on the feasibility of providing HAART in resource-limited settings. In a plenary session, Paul Farmer of Harvard University reported his experience in rural Haiti with a small (n=60) directly observed therapy–HAART pilot project at Clinic Bon Sauveur, which is located in one of the poorest parts of the poorest country in the Western Hemisphere [Lancet 2001; 358:404]. The treatment was directly observed therapy, given once or twice per day by community health workers (accompagnateurs, many of whom are HIV infected themselves). While Farmer noted that it is too early to draw conclusions about mortality, so far, all patients enrolled have had a positive clinical response characterized by weight gain and abatement of AIDS-related symptoms, and the medications have been well tolerated. More objective data on immunologic, virologic and clinical responses are needed, but Farmer has shown that treatment and adherence are possible under very adverse conditions.

Another encouraging report about a pilot HAART project in a developing country came from South Africa. In May 2001, Médecins Sans Frontieres added HAART with generic drugs imported from Brazil to the bundle of services available in dedicated HIV clinics at three government-run primary health care centers in Khayelitsha, a poor township of Cape Town [Kasper T, et al. Abstract 1095]. Patients with CD4 counts <200 cells/mm³ or with WHO stage 3 or 4 disease were enrolled in the program. In a primary analysis of the first 85 patients, presented by Kasper, the median CD4 cell count was 48 cells/mm³, and the median HIV RNA viral load was 5.20 log₁₀ c/mL (3.54-6.83). Fifty-eight patients were treated with AZT/3TC/nevirapine (NVP), and 27 received AZT/3TC/efavirenz (EFV). Incidence of opportunistic infections (OIs) was signifi-cantly decreased following initiation of HAART (4.4 new OIs per patient-year, prior to initiation of HAART vs 1.2 new OIs per patient-year following initiation). For patients who had been on HAART for >6 months, HIV RNA levels were <125 c/mL in 24/26 (92%) analyzed, the mean CD4 cell increase was 128 cells/mm³ (n=28), and the mean weight gain 8 kg (n=29). Side effects were limited, with 45% of patients reporting at least one, 87% of which were grade 1. Lab abnormalities of grade 1 were detected in 54%, and 8 patients switched therapy for drug intolerance.

HAART pilot programs conducted by the U.S. Centers for Disease Control and Prevention (CDC) in Kampala, Uganda, Abidjan (Ivory Coast), and Kisumu (Kenya) were reviewed by Lackritz [Abstract 1515]. The preliminary results of the UNAIDS/Uganda Ministry of Health HIV Drug Access Initiative in collaboration with the CDC have just been published [Lancet 2002;360:34]. The clinical and laboratory data for 476 patients at three centers were assessed from August 1998 to July 2000. Of those, 399 started antiretroviral therapy: 204 (51%) received HAART, 189 (47%) received dual nucleoside therapy, and 6 received nucleoside analog monotherapy. The median baseline CD4 count was 73 cells/mm³, and the median viral load was 193,817 c/mL. A Cox proportional hazards model showed that a CD4 count less than 50 cells/mm³ was strongly associated with death (hazard ratio of 2.93 [1.51-5.58] p=0.001). In addition, among patients with a viral load >1,000 c/mL for more than 90 days after beginning therapy, phenotypic resistance to NRTIs was found in 47 (57%): 29 of 37 (78%) who received non-HAART regimens vs 18 of 45 (40%) who received HAART (p=.0005). An encouraging HAART pilot program was also reported by Laurent [Abstract 3279].

The Brazilian Ministry of Health has made PI and NNRTI-based HAART universally available to patients in Brazil since 1996. Teixeira presented a 6-year progress report, noting that by December 2001, 113,000 patients were included in the program [Abstract 1098]. The results are comparable to what has been experienced in the United States and Western Europe: mortality has been reduced by 60% to 80%, with a notable reduction in the number of OIs. It was estimated that 358,000 hospital admissions were avoided because of this program in 1997-2001, representing a savings of $1 billion. It is worth knowing that today in Brazil, 63% of the antiretrovirals used are generics, and their prices have fallen by 82% over 5 years. In addition, the Ministry of Health negotiated a 60% cost reduction of imported drugs.

The HAART pilot programs described above demonstrate that HIV infected patients in developing countries can be managed successfully with HAART. When it comes to expanding or scaling up HAART access in countries with limited resources, however, much will depend on the local infrastructure and available resources. In addition, as recently reported by the WHO, there is an urgent need for the development and implementation of simplified, standardized treatment and monitoring algorithms that will facilitate such program expansion.

HAART Pilot Programs in the Private Sector

The HIV/AIDS epidemic is having a catastrophic socioeconomic impact on the private sector and hence on economic productivity in sub-Saharan Africa. A growing number of private companies in the region are moving on an action plan to provide HIV/AIDS services that include HAART treatment to their employees. Several studies were presented in Barcelona highlighting this trend. Eholie and colleagues reported on a study from Cote d’Ivoire that was conducted in a private company with 3,500 employees [Abstract 1096]. He and his colleagues evaluated social costs (absenteeism, employee replacement) and economic costs (low productivity, cost of care, replacements and funerals) attributed to HIV infection among personnel over two periods of time, 1998-99 and 1999-2000. During the first period, they found that the main problems were absenteeism, employees’ redeployment to other positions, and the social impact, which included fear and funerals. The economic effects included decreased productivity and an increased number of medical consultations and hospitalizations. During the first period the average cost for treatment of OIs was about U.S. $26,154 per patient and for funeral expenses approximately $107,692. Following the creation of a solidarity fund for HAART and aggressive promotion of voluntary counseling and testing services, the authors have been able to document significant decreases in healthcare costs and absenteeism.

In a taped plenary speech, South African AIDS activist Zachie Achmat sharply criticized the South African mining giant Anglo-American and challenged the company to create a program for providing AIDS treatment to its employees. It is estimated that 23% of the 90,000 people employed by Anglo in southern Africa (approximately 18,000 workers) are infected with HIV. In response, on August 6 the company announced that it would cover AIDS treatment of all its HIV-infected employees who do not have private medical coverage for HIV/AIDS.

Treatment Adherence

Preliminary data on HAART adherence in developing countries were presented at the XIV AIDS Conference in Barcelona. Pinheiro reported on a cross-sectional study of adherence to HAART in Southern Brazil [Abstract 5846]. Adherence was assessed by the self-report inventory method. There was a total of 195 patients, with 57% reporting 95% adherence in the previous two days; HIV RNA was <500 c/mL in 67.5% of this adherent group. Using a univariate analysis, the authors found that the odds of adherence decreased with frequency of dosing (OR=4.4 95% CI 0.2-0.94) and perception of negative affect and physical condition (3.50, 1.90-6.55). Adherence increased with self-efficacy (3.5, 1.90-6.55) and education (2.28, 1.12-4.66).

In a study conducted in Botswana (n=112) that used patient self-report and clinician assessment, Weiser and colleagues found that 54% of patients were adherent by self-report, while 53% were adherent by clinician assessment [Abstract 5851]. Agreement between both methods was only 68%, however. The main barriers to adherence were financial constraints related to the cost of antiretrovirals (44%), stigma (15%), migration (10%), side effects (9%), and lack of food (7%).

Another HAART adherence study came from the Joint Clinical Research Center (JCRC) in Kampala, Uganda, where Kityo and colleagues conducted a retrospective chart review of 577 patients on antiretroviral therapy from January 1998 to June 2001 [Abstract 5848]. Adherence was assessed via patient self-report, and patients were then characterized as “adherent” or “non-adherent.” An overall baseline adherence rate of 66% was reported.

Leandre and colleagues reported that their experience in rural Haiti showed DOT to be an effective way to maximize adherence to HAART and prevent the emergence of drug resistance [Abstract 3246]. This study was modeled after their experience with community-based treatment of multidrug-resistant TB in Haiti, Peru, and Russia, employing DOT twice daily. At the Clinic Bon Sauveur in Haiti, HIV-infected patients who begin HAART receive daily visits from a community health worker who observes only the first dose of the treatment. Of 40 patients tested, 88% had virologic suppression, and of the five without virologic suppression, only three had significant drug resistance.

Neonatal PEP with NVP and/or AZT for Prevention of Mother-to-Child-Transmission of HIV

In developing countries, a significant proportion of mothers do not have access to antenatal services until delivery. Therefore, studies evaluating the benefit of post-exposure prophylaxis (PEP) to prevent maternal-to-child transmission (MTCT) have been undertaken. Two studies with slightly different designs and conflicting results were presented in Barcelona. Taha reported preliminary results from a randomized clinical trial from Malawi that included 1059 babies randomized to receive either NVP/AZT (n=531) or NVP alone (528) as post-exposure prophylaxis [Abstract 1427]. HIV PCR results were available for 809 at 6 weeks (408 in the NVP/AZT arm and 401 in the NVP arm). At 6 weeks, 14.7% were HIV positive in the NVP/AZT arm compared with 22.7% in the NVP arm (p=0.004). Among babies who where negative at birth but positive at 6 weeks, 7.2% were in the NVP/AZT arm, and 12.1% in the NVP arm. Serious adverse events were rare (2% in each arm). The authors concluded that compared with NVP alone, post-exposure prophylaxis with NVP/AZT was significantly more efficacious, reducing MTCT by 35%.

In another study conducted at Chris Hani Baragwanath Hospital in Soweto, South Africa, Gray and colleagues conducted a randomized controlled trial evaluating a single dose of NVP vs 6 weeks of AZT for PEP in an MTCT setting [Abstract LB13]. They found that babies who were given a single dose of NVP within 24 hours after birth were no more likely to become infected with HIV than babies given AZT for the first six weeks of life. Approximately 6% of infants were infected at birth; an additional 7% of NVP-treated children became infected at 6 weeks, vs 11% of AZT-treated children. Other factors associated with transmission were low maternal CD4 count, high maternal viral load, and breastfeeding.

Simple and Affordable Diagnostic and Monitoring Lab Tests

Several papers were presented that examined efforts to find alternative, affordable, and feasible laboratory tests for the diagnosis and monitoring of HIV/AIDS. Of interest is the Dynabeads assay, an alternative method for providing CD4 cell counts using anti-CD4 monoclonal antibody-coated magnetic beads. Diagbouga reported on an international multicenter study conducted in six countries in West Africa that validates the Dynabeads method and then compared this assay with flow cytometry [Abstract 1342]. The correlation coefficient between the two techniques was 0.89, and the ability to consistently measure the CD4 count at clinically relevant thresholds was close to 95%. The cost per assay was estimated to be less than U.S. $10.00. Low-tech assays like this will make monitoring therapy much more feasible in Africa and other developing regions.

Rodriguez presented preliminary studies in progress suggesting that microchip-based assays for HIV antibodies, p24 Ag, RNA, and CD4 counts could be feasible and affordable in resource-poor settings [Abstract 1343].

An AIDS Activist’s Voice

As at previous World AIDS Conferences, the Barcelona conference provided an opportunity for AIDS activists to raise their voices against the inequalities in access to care between the South and the North. Zachie Achmat, the South African AIDS activist who in 1998 co-founded South Africa’s Treatment Action Campaign (TAC), delivered a dramatic and touching pre-taped speech at the plenary session, as an acute lung infection prevented him from attending the conference in Barcelona. Achmat called on drug companies to waive patent restrictions and open the doors to a competitive market in generic drugs in the developing world as a way to deal with the epidemic in a sustainable manner. He noted that pilot HAART projects using generic drugs that combine prevention and treatment, such as the one conducted by Médecins Sans Frontieres in Kyaletisha township [Abstracts 1095 and 3685], have proved that HAART can be feasible in limited-resource settings and should be expanded. Achmat lives with HIV and has free access to antiretrovirals, and yet he declared he would not take the lifesaving drugs until the South African Government agreed to an action plan to provide HAART in the public hospitals in South Africa.

Conclusion

The XIV International AIDS Conference will be remembered as the meeting that emphasized the need for both prevention and treatment of HIV/AIDS. It will also be remembered for the presentation of the first reports of pilot HAART projects that demonstrated the feasibility and effectiveness of such interventions in developing countries.