Researchers from South Africa have published the first detailed journal report of clinical outcomes from the pioneering antiretroviral treatment programme delivered through primary care clinics in the Western Cape township of Khayelitsha. The findings appear in the April 9th edition of the journal AIDS.
Patient selection
MSF provided free antiretroviral treatment for 287 patients selected on the basis of clinical and social criteria. The clinical criteria were symptomatic HIV disease (WHO stage 3 or 4) and a CD4 count below 200 cells/mm3. The social criteria were geared to ensuring that patients had the capacity to adhere to treatment. They included:
- Evidence of residence in Khayelitsha
- Evidence of disclosure of HIV status to at least one other person
- Nomination of a `treatment assistant` to support pill taking
- Home visit to review family environment
- Attendance on time at three previous appointments over the previous four months
Social criteria were reviewed by a community selection committee in an anonymised fashion before patients were approved for therapy.
Baseline characteristics – advanced HIV disease
The median baseline CD4 cell count was 43 cells/mm3 (IQR 13-94) and 52% had already been diagnosed with an AIDS-defining illness. 55% began treatment with a CD4 cell count below 50 cells/mm3
Patients had high viral loads when they began treatment (a mean of 5.18 log10 copies/ml, or around 150,000 copies/ml). Almost two-thirds (60%) began treatment with zidovudine/lamivudine (AZT/3TC) and efavirenz, with the remainder receiving zidovudine/lamivudine and nevirapine. Stavudine (d4T) was not used in this cohort and this paper does not report on clinical experience in the use of the fixed dose combination Triomune (d4T/3TC/nevirapine).
Survival – extremely advanced HIV disease raises problem for providers
After a median follow-up period of 13.9 months, 38 patients died (71% after less than three months of HAART) and a Kaplan-Meier plot of survival at 24 months estimated that 86.3% of those on treatment would be alive. Individuals who started treatment with CD4 cell counts below 50 cells had a significantly lower survival estimate at 24 months (81.8% vs 91.4%), but it is important to note that the Kaplan-Meier estimates are based on 34 individuals followed to 24 months, 54 to 21 months, 74 to 18 months, 123 to 15 months and 168 to 12 months. 257 of 287 patients had received at least three months of antiretroviral treatment by the time this report was finalised for publication in February 2004.
The risk of death was significantly greater in patients with a CD4 count below 50 cells/mm3 (AJR 2.41) and a previous diagnosis of Kaposi’s sarcoma (4.82). Twenty of the 38 deaths occurred in the first month of treatment (median 21 days) due to very advanced HIV disease (median CD4 cell count: 7 cells/mm3). Later deaths were due to tuberculosis (three, after variable time courses), Kaposi’s sarcoma (three within three months of commencing ART) and CMV colitis (one case after three and half months, having failed to respond to ART). All deaths occurred in the first year of treatment.
The remaining deaths were attributed to poor adherence or a treatment interruption that led to disease progression. One death due to lung cancer and one due to a cerebrovascular insult (stroke) were not associated with HIV infection. No deaths were associated with adverse effects of antiretroviral treatment.
The authors note that survival estimates are broadly in line with cohort estimates for patients with very low CD4 counts recorded in the United States and Canada, and that analysis of the risk of death at a lower cut-off point might offer a more powerful discriminatory tool for assessing which patients are at very high risk of death despite receiving ART.
“The difference in survival in the Khayelitsha cohort between those starting treatment with CD4 lymphocyte counts of 50 [cells/mm3] or more compared to the group with CD4 lymphocyte counts below this level, suggests that one of the major challenges in contexts such as Khayelitsha is to ensure a fair balance between providing access to ART for those presenting very late, whilst preserving the opportunity for a better prognosis for those who have enrolled in the programme in a timely manner.”
Weight gain
Patients gained a median of 5kg after six months of treatment and 9kg after 12 months of treatment. Weight has been proposed as a useful marker for efficacy of treatment although it is unclear how sensitive this marker might be to differences in the extent of immune reconstitution, food intake or virologic suppression.
Viral suppression
Eighty-nine per cent of patients who completed six months of treatment had viral load below 400 copies/ml at this point and 84.2% had viral load below 400 copies/ml at month 12 (n=165). At month 24, the proportion with viral load below 400 copies/ml fell to 69.7% (n=34), of whom three were receiving a second-line ART regimen.
It is important to note that these on-treatment analyses are not comparable with reports from developed world clinical trials and that the US/Canadian cohorts used for survival comparisons did not report on rates of viral suppression over time in a comparable cohort.
CD4 cell gains
Patients who completed six months of treatment gained a median of 134 cells/mm3 and those who completed 24 months of treatment gained a median of 288 cells/mm3.
Treatment changes – nevirapine and AZT tolerability
Nevirapine was the drug most likely to be changed due to intolerance; 8.8% (n=10) of patients swapped to efavirenz within a median of 20 days whilst only 1.2% (n=2) of patients swapped efavirenz for nevirapine (usually in the first week of treatment). A further ten patients switched from nevirapine to efavirenz due to a diagnosis of TB after starting treatment whilst three women changed from efavirenz to nevirapine due to pregnancy.
Twelve patients switched from zidovudine (AZT) to stavudine (d4T) after a median of 53 days.
Twelve patients changed the entire regimen due to treatment failure by month 24.
No other data on adverse events are provided.
Conclusions
These are the first substantial data to be reported in a peer-reviewed journal from treatment programmes that have rewritten the rules for delivering antiretroviral therapy, and indicate some of the challenges facing treatment providers. They also show the extraordinary success that can be achieved with antiretroviral therapy in patients with very advanced HIV disease
“Careful preparation of patients is essential… together with treatment in a primary care setting,” the authors note.
However, these data should not be treated as the final word on the viability of the Khayelitsha approach. Of note, the median follow-up in this cohort amongst those who survived the first year of treatment is only 14.9 months, and rates of both longer-term virologic failure in those who have already started treatment and survival and virologic failure as the cohort grows will be important evidence in convincing doubters that low-tech treatment interventions can provide highly effective treatment in resource-limited settings.
Another notable caveat is that the majority of patients in this cohort were treated with efavirenz. Triomune began to be used only towards the end of the treatment period captured in these data, so evidence of its clinical utility cannot be derived from these published data.
Full text of the article
Full text of the article is available at Medscape's website
Case study of ARV therapy in Khayelitsha published by WHO
Case study from WHO Essential Drugs Monitor 2003 (includes information on adverse events)
Coetzee D et al. Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS 18: 887-895, 2004.