IAS: ART interruption particularly unsafe for HIV-positive patients coinfected with hepatitis B or C

Intermittent interruption of antiretroviral therapy appeared particularly unsafe for HIV-positive people coinfected with hepatitis B or C in the SMART study, researchers reported on Tuesday at the Fourth International AIDS Society Conference on HIV Treatment, Pathogenesis and Prevention in Sydney.

Coinfected patients accounted for nearly half of all deaths due to non-opportunistic disease, even though they made up just 17% of the total study population.

In the SMART study, more than 5000 participants enrolled with a CD4 cell count above 350 cells/mm³ and were randomly assigned to one of two antiretroviral treatment strategies: CD4-guided intermittent therapy (drug conservation arm) or continuous therapy (viral suppression arm).

In the intermittent therapy arm, participants stayed off treatment as long as their CD4 cell count remained above 350 cells/mm³, and resumed therapy when it fell below 250 cells/mm³. The investigators’ hypothesis was that this strategy would allow patients to maintain CD4 cell counts at a safe level while minimising drug side-effects.

The study was terminated prematurely in January 2006 after an interim analysis revealed that participants who interrupted treatment had a significantly higher risk of death and opportunistic disease, as well as a slightly elevated risk of cardiovascular disease and liver and kidney problems (El Sadr 2006).

Ellen Tedaldi and colleagues with the SMART Hepatitis Group wanted to examine whether hepatitis B or C coinfection increased the risk of opportunistic or non-opportunistic disease or death in patients who interrupted treatment.

Out of the total 5472 SMART participants, 110 (2%) were coinfected with hepatitis B, 798 (15%) with hepatitis B, and 14 (0.25%) with both hepatitis B and C.

Patients were classified as having hepatitis B if they had been hepatitis B surface antigen-positive for at least six months before joining the study, and as having hepatitis C if they were hepatitis C antibody-positive at baseline. Coinfected patients were evenly distributed into the drug conservation and viral suppression arms.

Coinfected individuals and those with HIV alone had similar relative increases in the risk of opportunistic or non-opportunistic disease or death in the intermittent treatment arm compared with the continuous therapy group (hazard ratios of 2.58 and 2.57, respectively, for opportunistic disease or death; 1.78 and 1.69, respectively, for non-opportunistic disease).

Overall, there were few deaths due to opportunistic disease in either coinfected patients or those with HIV alone, and rates were comparable in both groups (0.14 vs 0.8 per 100 person-years, respectively).

However, the coinfected patients had nearly a four-fold greater risk of death due to non-opportunistic disease compared with the rest of the study participants who did not have viral hepatitis (2.52 vs 0.69 per 100 person-years, respectively).

These findings still held when the researchers considered only patients with hepatitis C.

The main causes of non-opportunistic death among hepatitis B or C coinfected patients were substance abuse and non-AIDS defining cancers. Coinfected patients were also more likely to die of liver or kidney disease, although patients without hepatitis B or C were slightly more likely to die of cardiovascular disease. Notably, the coinfected participants were significantly more likely to have an unknown cause of death.

The researchers concluded that interruption of antiretroviral therapy is particularly unsafe for subgroups such as patients coinfected with hepatitis B or C who already have a greater underlying risk of death due to non-opportunistic causes.

In a related study, researchers looked only at SMART participants coinfected with HIV and hepatitis B. They found that these patients were significantly more likely to restart HAART after treatment interruption than those without hepatitis B, mainly due to a faster decline in CD4 cell count.

Greg Dore, presenting, noted that it is difficult to draw clear conclusions about this group because the number of hepatitis B coinfected people was small. This is likely due to the fact that the SMART investigators discouraged participation by patients who needed antiretroviral drugs to control hepatitis B as well as HIV.

Several drugs used in HAART regimens, notably 3TC (lamivudine, Epivir), emtricitabine (Emtriva) and tenofovir (Viread), are active against both HIV and hepatitis B, so coinfected patients who interrupt treatment face a double risk of progression of both diseases. Researchers are currently analysing data to see whether hepatitis B viral load rose among patients in the intermittent therapy group.