Patients starting antiretroviral treatment in Switzerland and South Africa do equally well, provided they start treatment soon enough, according to a study published in the July edition of PLoS Medicine. Despite having fewer antiretroviral drugs available to them, South African patients were just as likely as those in Switzerland to see falls in their viral load to below 500 copies/ml after a year of treatment, with similar proportions of patients in the two countries experiencing rebounds in their viral load.
But there were some differences between the two study populations. Patients in South Africa had a higher risk of death, mainly because they had more advanced HIV disease than patients in Switzerland at the time anti-HIV treatment was started. Swiss patients were more likely to change treatment because of side-effects than those in South Africa, reflecting different approaches to treatment and the greater availability of antiretroviral drugs in the country.
Effective triple-drug anti-HIV treatment became available in industrialised countries like Switzerland in 1996 and quickly brought about significant and sustained reductions in rates of HIV-related illness and death. Over 20 anti-HIV drugs are now available in richer countries, and doctors are able to use a range of tests to regularly monitor patients and select the best possible treatment to suit each individual.
A different approach to antiretroviral therapy has been adopted in resource-limited countries such as South Africa. Based upon the experience of treating tuberculosis, the World Health Organization (WHO) developed a “public health” approach to antiretroviral therapy. This recognises the weak healthcare systems in such countries, poor access to monitoring tests, and limited availability of drugs. Anti-HIV treatment is, therefore, provided in standardised first- and second-line regimes, with few tests used to monitor patients.
An international team of investigators compared the outcome of individualised HIV therapy in Switzerland to the public-health approach adopted in South Africa by comparing falls and rebound of viral load, as well as rates of treatment changes, and mortality.
The study involved 2,348 patients from the Khayelitsha and Gugulethu townships in South Africa, and 1,016 patients in Switzerland. All the patients started their first anti-HIV treatment combination after 2001 and were over 16 years of age.
The South African patients were younger (33 years vs. 38 years) and more likely to be female (77% vs. 33%). Importantly, the South African patients had more advanced HIV disease at baseline than those in Switzerland, as indicated by significantly lower median CD4 cell counts (80 cells/mm3 vs. 204 cells/mm3) and by a higher proportion of patients having an AIDS diagnosis (91% vs. 19%).
Although the proportion of patients starting anti-HIV treatment remained constant in Switzerland, it almost doubled every year in South Africa from 79 individuals in 2001 to 509 in 2003. Swiss patients had their viral load monitored a month after starting anti-HIV treatment, but the monitoring of viral load was less frequent in South Africa, with the first test after the initiation of therapy performed after three or four months.
Nearly all (95%) the patients in the Cape Town townships started antiretroviral therapy with one of four NNRTI-based regimens. The most commonly used regimens were d4T/3TC with efavirenz (39%) or nevirapine (33%). By contrast, patients in Switzerland started anti-HIV treatment with 36 different combinations, the most common being AZT/3TC with efavirenz (30%) or Kaletra (20%). The NRTI, d4T, was only used by 3% of Swiss patients.
Patients in both countries were equally likely to start treatment in accordance with the appropriate HIV treatment guidelines (100% in South Africa, 95% in Switzerland).
Treatment changes
Changes to first-line treatment were significantly more frequent in Switzerland (53%) than in South Africa (22%). In both countries, these changes mostly involved the replacement of a single drug (South Africa, 64%; Switzerland, 51%). The cumulative two year probability of changing treatment due to side-effects was 24% in Switzerland compared to 12% in South Africa. But the probability of changing treatment because of virological failure was similar in both settings (Switzerland, 5%; South Africa, 4%).
However, the South African patients who did switch treatment due to virological failure had significantly higher viral loads at the time of such a change (25,000 copies/ml vs. 2,500 copies/ml), which potentially increases the risk of the development of resistance.
In both South Africa and Switzerland, side-effects were the main reason for changing treatment during the first three months of treatment, but the rate of such changes was significantly higher in Switzerland than South Africa (53 per 100 person years vs. 21 per 100 person years). Of particular note, 13% of all treatment changes in South Africa were due to lactic acidosis (a recognised side-effect of d4T), whereas no Swiss patient changed therapy because of this toxicity. Swiss patients, however, were more likely to change treatment because of abdominal or gastrointestinal complaints than those in South Africa (12% vs. 7%).
Whereas pregnancy and diagnosis with tuberculosis were important “other” reasons for treatment change in South Africa, changes because of patient request were common in Switzerland.
Response to treatment
After a year of anti-HIV treatment, 96% of patients in both South Africa and Switzerland had a viral load below 500 copies/ml. Furthermore, similar proportions of patients in both countries (26% vs. 27%) experienced a rebound in their viral load to detectable levels during the two year period of the study.
CD4 cell counts increased amongst South African patients from a baseline median of 80 cells/mm3 to 372 cells/mm3 after two years. Amongst the Swiss patients, median CD4 cell counts increased from 204 cells/mm3 to 449 cells/mm3. In both South Africa and Switzerland, those patients who started treatment with lower CD4 cell counts tended to have lower CD4 cell counts throughout the period of the study.
But mortality was significantly higher amongst the South African patients. After six months, 9% of South African patients had died compared to 4% of patients in Switzerland. The investigators believe that this difference is at least partially explained by the more advanced stage of HIV disease at the time treatment was started in the South African patients. They also suggest invasive bacterial and fungal infections in the South African patients are also likely to be an important cause of mortality. However, they also note that many patients in Switzerland start treatment late and they suggest that late diagnosis and initiation of therapy are likely to be important causes of early mortality in Switzerland as well.
“A public-health approach to HAART provision using a limited repertoire of drugs and relatively few viral load measurements resulted in virologic treatment outcomes in townships in South Africa that were similar to outcomes in Switzerland,” conclude the investigators. They add, however, “our study shows that many patients would benefit from earlier initiation of therapy, particularly in South Africa.”
Keiser O. et al. Public-health and individual approaches to antiretroviral therapy: township South Africa and Switzerland compared. PLoS Medicine 5: e148, 2008.