Back HIV/AIDS HIV/AIDS Topics HIV Treatment Cascade of Care CROI 2015: Treatment Cascades and Viral Load Surveys Inform ART as Prevention in Africa

CROI 2015: Treatment Cascades and Viral Load Surveys Inform ART as Prevention in Africa


Reaching ambitious HIV prevention targets in South Africa will require intensified efforts to engage and retain men and young people in care, in order to increase the proportion of people on HIV treatment with suppressed viral load, according to a national study presented at the Conference on Retroviruses and Opportunistic Infections (CROI 2015) last week in Seattle. Another study, conducted in 3 countries in sub-Saharan Africa, showed that to maximize the preventive effect of antiretroviral therapy (ART), efforts to expand treatment coverage need to focus on those with the highest viral load off treatment -- mainly people who are already eligible for treatment under current guidelines.

[Produced in collaboration with Aidsmap]

South Africa already has the largest antiretroviral treatment program in the world. Over the past decade the country has scaled up from around 50,000 people on treatment in 2004 to more than 2.1 million people on treatment by the end of 2012, and has built an extensive program for prevention of mother-to-child HIV transmission.

However, as well as the largest treatment program in the world, South Africa also has the largest number of new infections. 400,000 people in South Africa are estimated to have been infected with HIV in 2012 alone. Reducing the rate of new infections is critical to ensure the long-term sustainability of the existing treatment program, as well as for the prevention of future illness and death.

Epidemiological modeling has projected that South Africa -- and neighboring countries -- could achieve big reductions in HIV incidence if the number of people with fully suppressed viral load on treatment can be increased. Until now, South Africa lacked a robust estimate of how many people have suppressed viral load, and which groups of people are more likely to be linked to care and retained on treatment.

Simbarashe Takuva of the National Institute for Communicable Diseases presented results of a study designed to show how South Africa's "treatment cascade" is performing, and where improvement is needed in order to maximize the preventive benefit of antiretroviral treatment.

In the absence of an integrated national surveillance system, the research group, supported by epidemiologists from Public Health England, combined population data from the National Household Survey and Vital Registration Data from 2012 to estimate the size of the population of people living with HIV. They also used laboratory data from the National Health Laboratory Services to estimate the numbers engaged in care (using CD4 count measurements as a proxy) and on treatment with viral suppression (using viral load measurements as a proxy).

The study found that almost half of all South Africans with HIV are not linked to care, but it was unable to show what proportion of diagnosed people are not linked to care. However, among those linked to care and on treatment, 75% are virally suppressed.

Treatment cascade, South Africa, 2012



Living with HIV

6.42 million


3.87 million

2.53 million

Linked to care

3.3 million





2.185 million




Virally suppressed

1.6 million




The study found a major disparity by gender, and also found substantial disparities by age group. Young people aged 15-24 were much less likely to be on treatment and virally suppressed than people aged 25-49, despite similar rates of linkage to care. Although this may be partially explained by more recent HIV acquisition and hence higher CD4 cell counts in this group, the treatment gap remains very large. Young people aged 15-24 were 24% less likely to be virally suppressed than their older counterparts.

Treatment cascade by age group, South Africa, 2012





50 and over

Living with HIV



4.7 million


Linked to care










Virally suppressed





Viral suppression ratio





*p<0.001 vs reference (50 and over age group)

The findings showed that more focus is needed on improving engagement with care, especially among men and young people, said Takuva, together with improving opportunities to test for HIV.

Population Viral Load

A 3-country survey of viral load measurements in people not taking antiretroviral treatment found that any attempt to maximize the preventive benefit of treatment probably needs to focus first on getting those individuals already eligible for treatment onto therapy, rather than seeking to expand treatment eligibility criteria to reflect results of the Temprano study, also presented at CROI.

As presented by David Maman of Médecins sans Frontières (MSF), the study -- carried out in 3 districts where MSF provides HIV treatment in Kenya, Malawi, and South Africa -- looked at a range of correlates of high viral load in people not taking ART in each setting. (A companion study on viral load in breastfeeding women was presented earlier in the week).

Data on viral load were gathered after a household survey tested a random sample of households for HIV and then provided CD4 cell and viral load testing for those who tested positive. People who tested positive were asked whether they were on treatment. In total, 19,005 people were tested across the 3 districts, of whom 3938 were HIV positive and had a viral load test.

HIV prevalence ranged from 17% in Malawi to 25% in South Africa. 60% of people tested were eligible for treatment under national guidelines in Kenya, compared to 69% in South Africa and 80% in Malawi. This difference was driven chiefly by national policies on antiretroviral treatment for pregnant and breastfeeding women. Whereas 62% of people with HIV had a viral load below 1000 copies/mL in Malawi, only 40% had a viral load this low in Kenya.

When evaluating median viral load by CD4 cell strata, the study found that although the median viral loads at all strata represented a risk for HIV transmission, those untreated people with CD4 cell counts below 500 cells/mm3 were far more likely to have viral loads above 100,000 copies/mL, the level associated with a very high risk of onward transmission. Whereas less than half of those with CD4 counts above 750 cells/mm3 had a viral load above 10,000 copies/mL and less than 15% had a viral load above 100,000 copies/mL, just over half of those with CD4 counts below 350 cells/mm3 had viral loads above 100,000 copies/mL. The median viral load in this CD4 cell stratum was 108,407 copies/mL. Mean viral load was significantly lower among people with CD4 counts over 500 cells/mm3 at each site. The study also found that untreated men had significantly higher viral load than women.



S Takuva, A Brown, W Macleod, et al. Disparities in Engagement Within HIV Care in South Africa. 2015 Conference on Retroviruses and Opportunistic Infections. Seattle, February 23-24, 2015. Abstract 154.

D Maman, H Huerga, G Van Cutsem, et al. Population Viral Load in Three High HIV Prevalence Settings in Sub-Saharan Africa. 2015 Conference on Retroviruses and Opportunistic Infections. Seattle, February 23-24, 2015. Abstract 153.