Antiretroviral treatment regardless of CD4 count: the universal answer to a contextual question.

Serge P Eholié,Raoul Moh,Anani Badje,Xavier Anglaret,Gérard M Kouame,Christine Danel,Jean-Baptiste N’Takpe

doi:10.1186/s12981-016-0111-1

Abstract

After a period where it was recommended to start antiretroviral therapy (ART) early, the CD4 threshold for treating asymptomatic adults dropped to 200/mm3 at the beginning of the 2000s. This was mostly due to a great prudence with regards to drug toxicity. The ART-start CD4 threshold in most international guidelines was then raised to 350/mm3 in 2006–2009 and to 500/mm3 in 2009–2013. Between 2012 and 2015, international guidelines went the last step further and recommended treating all HIV-infected adults regardless of their CD4 count. This ultimate step was justified by the results of three randomized controlled trials, HPTN 052, Temprano ANRS 12136 and START. These three trials assessed the benefits and risks of starting ART immediately upon inclusion (“early ART”) versus deferring ART until the current starting criteria were met (“deferred ART”). Taken together, they recruited 8427 HIV-infected adults in 37 countries. The primary outcome was severe morbidity, a composite outcome that included all-cause deaths, AIDS diseases, and non-AIDS cancers in the three trials. The trial results were mutually consistent and reinforcing. The overall risk of severe morbidity was significantly 44–57 % lower in patients randomized to early ART as compared to deferred ART. Early ART also decreased the risk of AIDS, tuberculosis, invasive bacterial diseases and Kaposi’s sarcoma considered separately. The incidence of severe morbidity was 3.2 and 3.5 times as high in HPTN052 and Temprano as in START, respectively. This difference is mostly due to the geographical context of morbidity. The evidence is now strong that initiating ART at high CD4 counts entails individual benefits worldwide, and that this is all the more true in low resource contexts where tuberculosis and other bacterial diseases are highly prevalent. These benefits in addition to population benefits consisting of preventing HIV transmission demonstrated in HPTN052, justify the recommendation that HIV-infected persons should initiate ART regardless of CD4 count. This recommendation faces many challenges, including the fact that switching from “treat at 500 CD4/mm3” to “treat everyone” not only requires more tests and more drugs, but also more people to support patients and help them remain in care.

Highlights

The optimal threshold for initiating antiretroviral therapy (ART): a 20‐year quest 1981–1996: The no‐treatment era During the first 15 years of the HIV pandemic, no treatment could sustainably control the replication of the virus and decrease mortality in HIV-infected people
This difference is likely explained by the context, the population, and the differences in the documentation of tuberculosis discussed above
In terms of relative risks, the overall risk of severe morbidity was 44–57 % lower in patients randomized to the early ART strategy as compared to those randomized to the deferred ART strategy in Temprano and START, respectively

Summary

Background

The optimal threshold for initiating ART: a 20‐year quest 1981–1996: The no‐treatment era During the first 15 years of the HIV pandemic, no treatment could sustainably control the replication of the virus and decrease mortality in HIV-infected people. RCT randomized controlled trial; IPT isoniazid prophylaxis for tuberculosis a ART-start CD4 threshold for asymptomatic patients randomized to the deferred ART strategy b For ethical reasons, the Temprano investigators considered that the WHO revised guidelines had to be followed as soon as they were released. In terms of absolute risks, the incidence of severe morbidity in the deferred ART strategy was 3.2 and 3.5 as high in HPTN052 and Temprano compared to START, respectively This difference is likely explained by the context (more patients in low resource settings in HPTN052 and Temprano), the population (higher baseline CD4 count in SMART), and the differences in the documentation of tuberculosis discussed above. In the three trials, ART appeared to be well tolerated in both strategies (data not shown)

Findings

Discussion

Conclusions