CROI: Risk of treatment interruption persists after restarting HAART

This article is more than 16 years old. Click here for more recent articles on this topic

The detrimental effects of interrupting antiretroviral therapy are reduced, but not entirely reversed, after restarting treatment, researchers with the SMART study reported on Monday at the Fifteenth Conference on Retroviruses and Opportunistic Infections in Boston.

SMART was a large international treatment strategy trial comparing structured treatment interruption versus continuous antiretroviral therapy. A total of 5,472 participants with a CD4 count above 350 cells/mm3 at baseline were randomly assigned to two strategy arms. In the drug conservation (DC) arm, patients stayed off therapy while CD4 count was above 350 cells/mm3 and resumed when it fell to 250 cells/mm3. Those in the viral suppression (VS) arm received continuous therapy throughout the study.

About three-quarters of the SMART participants were men, about 30% were black, and the median age was 43 years. At baseline, participants in general had well-controlled HIV disease; about 70% had an undetectable viral load (3, although many had had much lower CD4 counts in the past.


hazard ratio

Comparing one group with another, expresses differences in the risk of something happening. A hazard ratio above 1 means the risk is higher in the group of interest; a hazard ratio below 1 means the risk is lower. Similar to ‘relative risk’.


Expresses the risk that, during one very short moment in time, a person will experience an event, given that they have not already done so.

treatment interruption

Taking a planned break from HIV treatment, sometimes known as a ‘drugs holiday’. As this has been shown to lead to worse outcomes, treatment interruptions are not recommended. 

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.

detectable viral load

When viral load is detectable, this indicates that HIV is replicating in the body. If the person is taking HIV treatment but their viral load is detectable, the treatment is not working properly. There may still be a risk of HIV transmission to sexual partners.

Enrolment in SMART was halted in January 2006 after interim data showed that people in the DC arm had a higher risk of illness and death than those in the VS arm. At the 2006 Retrovirus conference, researchers reported that patients in the DC had a higher rate of opportunistic disease or death due to any cause, as well as an unexpected increase in serious cardiovascular, kidney, and liver problems.

At this year’s meeting, Wafaa El-Sadr of Harlem Hospital Center in New York reported final results from SMART, looking at outcomes after the trial was modified on January 11, 2006, with all DC arm participants encouraged to switch to continuous therapy. Follow-up continued through July 11, 2007.

At the time of study modification, 36% of participants in the DC arm and 94% in the VS arm were on treatment. Those in the DC arm had a higher median CD4 count (425 vs 625 cells/mm3) and were much less likely to have undetectable viral load (35% vs 82%). At the end of follow-up, 84% and 95%, respectively, were on treatment.

Before study modification, patients in the DC arm spent 34% of total follow-up time on treatment, compared with 94% in the VS arm. After study modification, between January 2006 and July 2007, DC arm participants spent 71% of total time on HAART, compared with 91% in the VS arm.

Before January 2006, there were 172 total instances of opportunistic disease or death due to any cause. Rates were 3.4 per person 100 person-years in the DC arm compared with 1.4 in the VS arm, for a hazard ratio of 2.5. In addition, there were 104 serious cardiovascular, kidney, or liver events. Again, the risk was higher in the DC arm, with a hazard ratio of 1.7.

Between January 2006 and July 2007, there were 131 total occurrences of opportunistic disease or death: 2.1 vs 1.4 events per 100 person-years in the DC and VS arms, respectively. The hazard ratio during this period was 1.4—significantly down from the pre-modification 2.5, but still representing a higher risk in the treatment interruption arm compared with the continuous therapy group.

Hazard ratios also dropped when looking at opportunistic diseases, all-cause death, and cardiovascular, kidney, or liver events separately, but the decrease was greatest for opportunistic illnesses. Study participants who experienced opportunistic disease or serious cardiovascular, kidney, or liver events before January 2006 still had nearly a 6-fold increased risk of death after study modification.

While there was a significant decrease the risk of opportunistic disease or death after the DC group was encouraged to switch to continuous therapy, there remained a “residual excess risk” at the end of follow-up for patients who had interrupted treatment. The researchers attributed this persistent excess risk to a lower average CD4 cell count and more patients with detectable viral load (in part because some patients did not resume continuous therapy as recommended). Importantly, CD4 cell recovery was slow after resuming treatment, even amongst individuals who experienced a rapid reduction in HIV RNA.

The researchers concluded that their findings support a recommendation not to interrupt antiretroviral therapy based on CD4 cell threshold, as was done in the SMART study, since “Antiretroviral therapy interruption is associated with long-term consequences beyond the period of treatment interruption.”


El-Sadr W et al. Re-initiation of ART in the CD4-guided ART interruption group in the SMART study lowers risk of opportunistic disease or death. Fifteenth Conference on Retroviruses and Opportunistic Infections, Boston. Abstract 36, 2008.