Two very large cohort studies published this month both agree: antiretroviral treatment should not be delayed after the CD4 count falls below 350. However, the two studies, previously presented at international conferences, show contradictory evidence on whether starting treatment before the CD4 count falls below 500 has an additional benefit.
Both studies were presented at the Sixteenth Conference on Retroviruses and Opportunistic Infections in February, and fuller details can be found in a news report published at the time.
The When to Start Consortium analysed 22,444 patients who started antiretroviral treatment after 1997 in Europe, North America and Argentina, and compared their risk of AIDS or death with a historical control group, who did not receive antiretroviral therapy, during the period between 1989 and 1997.
They carried out statistical adjustments in order to approximate the effects of a clinical trial, although they say their methodology cannot eliminate bias in the way that a randomisation in a clinical trial could do.
They found a significantly greater risk of AIDS or death in those who started treatment with a CD4 count below 350 cells/mm3, but no additional reduction in the risk of death if people started treatment at CD4 counts above this level.
The North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) analysed the effect of early versus deferred antiretroviral treatment on survival in 17,517 patients who began treatment between 1996 and 2005.
They found that the risk of death was 94% greater in those who deferred treatment to a CD4 count below 500 cells/mm3 compared to those who started treatment with a CD4 count above 500 (confidence interval 1.37 to 2.94, p3 (confidence interval 1.26 to 2.26, p
Editorials accompanying both publications agree that the only way the field of HIV treatment will reach a definitive conclusion on the question of when to start antiretroviral treatment will be by carrying out a large randomised trial of immediate versus deferred treatment.
In the New England Journal of Medicine Dr Paul Sax and Dr Lindsey Bingham argue that patients who started treatment early in the late 1990s were “the ideal patients: highly adherent, committed to doing whatever they could to prevent AIDS and willing to push through the sometimes punishing side effects and drug regimen burdens of the early therapies.” This may have biased the result of the NA-ACCORD study, and its effect could only be ruled out in a large randomised study.
They also note that almost 45% of patients in each stratum in the NA-ACCORD dataset either did not start treatment or did not experience a decline in their CD4 count, and so were not included in the analysis. Would antiretroviral treatment have benefited this group? We have no way of knowing, note Drs Sax and Bingham. In addition, we don’t know what effect earlier treatment has on the development of resistance or subsequent treatment options.
But they also point out that many patients in the study were taking what would now be regarded as sub-optimal treatment – either less potent or more toxic than today’s recommended regimens – so “the results…are all the more remarkable.”
In a commentary in The Lancet, leading South African researchers Robin Wood and Stephen Lawn of the Desmond Tutu HIV Centre at the University of Cape Town warn that results from studies in the developed world cannot be extrapolated to resource-limited settings.
“Mortality rates and the range of morbidity differ between cohorts in industrialised countries and resource-limited settings…Early mortality in patients receiving ART in sub-Saharan Africa is also substantially greater than in those treated in high-income countries,” they point out.
They also note that fewer antiretroviral drugs are available in resource-limited settings, changing the risk of toxicity and the cost-benefit ratio of earlier treatment.
They argue that any trial of when to start treatment should include developing-country sites in order to inform international guidelines.
A major study is planned by the US National Institute of Allergy and Infectious Disease.
The Strategic Timing of Antiretroviral Therapy study (START) will compare the effects of starting treatment at a CD4 count above 500 cells/mm3 with deferring treatment until the CD4 count falls below 350 cells/mm3.
The primary outcome being measured in the study is death, whether from AIDS or non-AIDS causes, and the development of AIDS. The secondary outcome is non-AIDS-defining illnesses and all-cause mortality.
The START study is due to begin recruiting treatment-naive patients within months into a first exploratory phase, in which investigators will attempt to recruit 900 patients in one year through 70 participating sites. If they are successful, they are likely to recommend to funders that the study be expanded to recruit a total of 4000 patients.
In the interim, European and North America guidelines currently recommend that treatment should start when the CD4 count falls to around 350 cells/mm3.
World Health Organization guidelines for resource-limited settings recommend that where CD4 counts are available, treatment should start before the CD4 count falls below 200 cells/mm3, with consideration given to treating patients with a CD4 count below 350 cells/mm3.
When to Start Consortium Timing of initiation of antiretroviral therapy in AIDS-free HIV-infected patients: a collaborative analysis of 18 cohort studies. The Lancet (early online publication, April 9, 2009)
Wood R, Lawn S Should the CD4 threshold for starting ART be raised? The Lancet (early online publication, April 9, 2009)
Kitahata MM et al., for the NA-ACCORD Investigators. Effect of early versus deferred antiretroviral therapy for HIV on survival. N Engl J Med 360 (early online publication, April 1 2009)
Sax P, Bingham L When to start antiretroviral therapy: ready when you are? N Engl J Med 360 (early online publication, April 1 2009)