AZT, 3TC and efavirenz: a tolerable first-line regimen for South African men

The safety and tolerability of a first-line regimen of AZT, 3TC and efavirenz has been assessed in a workplace antiretroviral treatment programme in South Africa. The regimen was generally well-tolerated in the mostly male cohort, with short-term increases in nausea, central nervous system symptoms, and neutropenia, but few treatment-limiting events and no serious incidences of anaemia. The study was reported in the January 2^nd edition of AIDS.

In 2002, the World Health Organization (WHO) recommended a three-drug first-line antiretroviral regimen for developing countries consisting of lamivudine (3TC) plus either stavudine (d4T) or zidovudine (AZT), and either nevirapine or efavirenz . Stavudine was initally preferred because it was cheaper, was available in more fixed dose generic combinations and was expected to cause fewer severe toxicities than AZT.

However in 2006 these recommendations were revised to favour AZT or tenofovir over d4T due to accumulating evidence of d4T-associated toxicities, even though monitoring for AZT-associated anaemia places additional demands on resource-limited settings.

This study, conducted by an African/UK/US research team, investigated the safety and tolerability of a triple regimen of AZT, 3TC and efavirenz. Study participants were enrolled from an HIV workplace programme in the country of South Africa. Eligibility for the HAART programme required either: CD4 cell counts less than 250 cells/mm³, CD4 cell counts less than 350 cells/mm³ with WHO stage 3 clinical symptoms, or WHO stage 4 clinical symptoms. The first-line regimen was AZT, 3TC, and efavirenz. Participants were monitored every six months if not on treatment; after starting HAART, follow-up was conducted at two, six, and twelve weeks, and every three months thereafter.

During the study enrollment period, 938 people were started on antiretroviral therapy through the workplace programme. Of these, 85 were excluded due to lack of consent, lack of laboratory data or being on an alternate regimen. The remaining 853 were enrolled in the study; all started antiretroviral therapy between November 2002 and October 2005 and were followed up until November 2006. Nearly all participants (98%) were male. On starting antiretroviral therapy, the median age was 40 years, median weight 64 kg, median CD4 cell count 186 cells/mm³, and median viral load 50,000 copies/ml. Nearly all (93%) were already at WHO clinical stage 3 or 4 of HIV disease.

The study analysed the occurrences of specific adverse events commonly associated with AZT (anaemia, neutropenia, nausea and vomiting, lipodystrophy), efavirenz (hepatotoxicity, CNS symptoms, rash), and d4T (lactic acidosis, peripheral neuropathy). (The latter were analysed for comparison with other antiretroviral programmes, although this study itself did not include anyone on d4T.) Occurrence of CNS symptoms was based on self-report of dizziness, ataxia, insomnia, bizarre dreams, or hallucinations.

A total of 92 participants (11%) developed grade 3 or 4 laboratory toxicities while on antiretroviral therapy and 18 (2.1%) developed grade 3 or 4 clinical adverse events; 20 (2.3%) stopped or changed treatment due to adverse events. Laboratory adverse events were reported as follows:

[* observed incidents per 100 person-years, (95% confidence interval (CI)) ]

Eleven participants had low haemoglobin on beginning antiretroviral therapy. (Low hemoglobin was defined as

Clinical adverse events were reported as follows:

Adverse event n Incidence*

Rash, pre-antiretroviral therapy 227 62 (54.4 – 70.6)

Rash, on antiretroviral therapy 228 48 (42 – 55)

Peripheral neuropathy, pre-HIV therapy 61 8.2 (6.3 – 10.4)

Peripheral neuropathy, on HIV drugs 62 11 (8.5 – 14)

Nausea & vomiting, pre-antiretrovirals 41 10 (7.3 – 14)

Nausea & vomiting, on antiretrovirals 143 27 (22 – 31)

CNS symptoms, pre-antiretrovirals Not collected n/a

CNS symptoms, on antiretrovirals 187 37 (32 – 43)

[* observed incidents per 100 person-years, (95% confidence interval (CI)) ]

The researchers concluded that this regimen was "well tolerated with a minimal increase in clinical and laboratory events from … pre-HAART level[s]," and that most adverse events "occurred within the first 12 weeks of therapy and were effectively managed." Significantly, "despite a concern for potentially worsening anemia, we found the opposite," with "a sustained overall rise in haemoglobin after week six". This finding is in contrast to most findings in developed countries. High rates of neutropenia were seen, but the team "did not detect evidence of [resultant] severe infection;" one neutropenia-related death occurred in a patient with advanced HIV disease and disseminated tuberculosis. The researchers state that "further evaluations of … tolerability ... are important, especially among women and in regions with a high prevalence of endemic malaria."