Would early treatment for everyone be better? Study poses the question

Gus Cairns
Published: 21 December 2011

A US study has found that patients infected with HIV in the previous six months who were given 36 weeks (8.3 months) of antiretroviral therapy (ART) were only a third as likely to need to go back on therapy over the next two years than a comparator group of patients, who had not been given early therapy, were to need to start it.

The study

The ACTG A5217 or Setpoint Study randomised 130 people who had been infected with HIV more than one month but less than six months before diagnosis either to take 36 weeks of ART or defer treatment until treatment criteria were met. They excluded patients who already needed therapy due to low CD4 count, high viral load or AIDS-defining illness.

The initial aim of the study had been to see if a short period of early therapy produced a permanent reduction in the average viral load patients have off-treatment during chronic infection – their so-called viral ‘set point’, hence the name of the study.

This set point was due to be measured at week 72 of the study: that is, 72 weeks after diagnosis for both groups and 36 weeks after stopping therapy for those who took the short course.

Anticipating, however, that having had 36 weeks on therapy might produce better outcomes than having had none, a comparison was also due to be made between the viral load at 72 weeks in patients who had taken short-course therapy and at 36 weeks on patients who had not taken it, in other words after 36 weeks of uncontrolled HIV viremia (detectable viral load in the blood) in both groups.

However the study was stopped early in August 2009 after 130 of a planned 150 subjects had been recruited because twice as people than was anticipated, of those who had not had initial ART, progressed to the point when they needed to take it.

This meant that the primary endpoint – viral load set point – could not be determined as there were not enough patients left who had never taken therapy to provide a valid comparison.

Outcomes

There were however very significant differences between arms in terms of the proportion of patients who needed to start or re-start therapy.  The criteria for starting or restarting ART were:

  • two consecutive CD4 counts below 350 cells/mm3 or one below 200 cells/mm3 or with a CD4 percentage below 14%;

  • a viral load over 700,000 more than a month into the study or one over 200,000 more than three months into the study;

  • any of the CDC/WHO list of HIV-related symptoms. 

Including all 130 participants, regardless of the amount of time they had been in the study, 36% (23 out of 64) of those who had not taken 36 weeks of’ ART progressed to needing therapy versus 11% (seven out of 66) of those who had taken it.

The most common reason for initiating therapy was low CD4 count (six on initial treatment and 13 on no initial treatment, including one with a CD4 count below 200 cells/mm3 and two with a CD4 percentage below 14%).

The only other person who took initial ART who re-initiated ART did so because of chronic fatigue.

Of the other ten patients who did not take initial therapy and who had to start treatment, five did so because of a high viral load, one had both a high viral load and a low CD4 count, and four had HIV-related symptoms (two low platelets, one oral hairy leukoplakia and one persistent herpes symptoms). There were two deaths in the group not given initial therapy but neither appeared medically related to HIV (one was a suicide).

However these 130 patients included some who had only been in the study for a short time. The primary analysis was therefore of 79 participants who had been in the study for more than 72 weeks when it was stopped, half of whom had taken initial 36-week ART, .     

Amongst this group only 10% of those given initial ART  needed to re-start it, but 50% of those who had not taken initial ART needed to start it.

When patients who had not taken initial ART were compared 36 weeks after study entry with patients who had taken initial ART 72 weeks after study entry (i.e. 36 weeks after stopping ART), 27.5% of those not given initial ART needed to start therapy versus 11% who had been given initial ART.

In other words, if patients took initial therapy then stopped it, they were only just over a third as likely to progress to needing ART in the next 36 weeks than patients who had never taken therapy were in the initial 36 weeks.

Discussion and conclusions

Making a direct comparison in treatment outcomes between the two arms of this study is difficult. On the one hand, comparing both groups at week 72 of the study mainly proves that a short course of initial therapy puts off the time you need to start it – not a surprising result. On the other hand, comparing the arms at week 36 off-therapy, while including all the patients who did not take initial ART, only includes those patients who did take ART who were in the study for at least 72 weeks before it was closed.

It also includes no comparator arm of patients who remained on ART, so no comparison can be made between deferring lifetime treatment and initiating it.

The main surprise in the study was the speed at which people off therapy progressed to needing it. One model based on the CASCADE cohort (Lodi) has predicted that 27% of patients will develop a CD4 count below 350 cells/mm3 within two years of seroconversion; in contrast, this study found that half of patients not given initial therapy did.

The A5217 study may overestimate the speed of progression because it mainly recruited patients who had seroconversion symptoms; or the CASCADE study could underestimate progression because it excluded people who had taken therapy in early infection and then stopped, thus excluding the fastest progressors.

However, as the authors say, “If immediate therapy is not begun, progression to meeting standard criteria for ART initiation may occur more rapidly than expected...[this] contributes to the growing body of evidence favouring earlier treatment.”

In an accompanying editorial, Tossonian and Conway point out that the results of ACTG A5217 are compatible with the results from the SPARTAC study (Fidler) which found that initial therapy also delayed subsequent therapy but found that just under half of patients needed to start or restart therapy within four years.

They comment that “it is not often possible or medically indicated” for patients to start taking therapy at diagnosis, but that studies like A5217 and SPARTAC show that the time to starting ART may be considerably shorter, especially with changed CD4 criteria, than patients often suppose; doctors now have the evidence to suggest they may have a year to 18 months on average before they need to start ART, even if diagnosed early.       

References

Hogan CM et al. The Setpoint Study (ACTG A5217): effect of immediate versus deferred antiretroviral therapy on virologic set point in recently HIV-1-infected individuals. Journal of Infectious Diseases, advance online publication. DOI:10.1093/infdis/jir699. 2011.

Tossonian H and Conway B. Recent HIV-1 infection: to treat or not to treat, that is the question.  Journal of Infectious Diseases, advance online publication. DOI:10.1093/infdis/jir702. 2011.

Lodi S et al. Proportion of individuals likely to need treatment for CD4 thresholds <200, <350, and <500 cells/mm3. Seventeenth Retrovirus Conference, San Francisco. Abstract 525. 2010.

Fidler S et al. The effect of short-course antiretroviral therapy in primary infection: final results from an international randomised controlled trial; SPARTAC. Sixth IAS Conference, Rome. Abstract WELB06. 2011.