- Home>
- Specialties>
- HIV/AIDS>
- Meeting Report
Further Exploring the Findings from SMART
Of the many SMART substudies described at the 2007 IAS Conference, the most intriguing was a report about patients who were treatment naive or off therapy at baseline.
Previous reports from the SMART study demonstrated that interrupting antiretroviral therapy (ART) at CD4 counts >350 cells/mm3 substantially increases the risk for HIV progression and for non-AIDS complications (ACC Nov 29 2006). Now, as described at the 4th IAS Conference on HIV Pathogenesis, Treatment and Prevention, investigators have conducted a variety of substudies in SMART to further evaluate the consequences of both delayed treatment initiation and treatment interruption.
Consequences of Delayed ART Initiation
Although the vast majority of SMART participants were receiving antiretrovirals at study enrollment, a small number of patients (approximately 500) were treatment naive or had not received ART in the 6 months prior to enrollment [Abstract WEPEB018]. These patients, who were randomized either to start ART immediately or to defer treatment until CD4 counts fell below 250 cells/mm3, provided intriguing information about whether therapy should be started at higher CD4-cell counts than is currently recommended. Those who delayed starting therapy were found to have a significantly higher risk for opportunistic disease and non-AIDS illness than did those who received immediate treatment. Overall, across all CD4-cell–count strata, deferred therapy was associated with a 5.7% increase in absolute risk for opportunistic disease and serious non-AIDS events.
When added to the main findings of SMART and other studies, these results encourage consideration of treatment initiation at higher CD4-cell counts than those recommended by current guidelines. The consensus at the meeting was that the issue of when to start therapy remains critical and that the time is ripe for another large randomized trial to provide definitive answers. Such a trial, the START (Strategic Timing of Antiretroviral Therapy) study, is currently in development.
Consequences of Treatment Interruption
Bacterial Pneumonia
Bacterial pneumonia was found to be the most common clinical diagnosis among all SMART participants, with 115 cases during an average follow-up of 16 months [Abstract MOPEB100]. The rate of this condition was significantly higher in the treatment-interruption arm than in the continuous-therapy arm. Independent predictors included treatment interruption, current cigarette use, and prior recurrent bacterial pneumonia. On further analysis, lower CD4-cell counts and higher viral loads almost entirely explained the higher rate of pneumonia seen in the treatment-interruption group. Among all SMART enrollees, the risk for bacterial pneumonia decreased by 8% with every additional increase of 100 CD4 cells/mm3. Even in the highest CD4 strata (>500 cells/mm3), the risk for bacterial pneumonia was more than twice as high in the treatment-interruption group as in the continuous-therapy group. These results further support the importance of continuing ART and quitting smoking to reduce the occurrence of bacterial pneumonia in HIV-infected patients.
Electrocardiographic Abnormalities
In a substudy conducted among approximately 2200 SMART participants who had repeat electrocardiograms [Abstract WEPEB077], those who interrupted treatment were more likely than those who did not to have irregular Q waves (hazard ratio, 1.48; 95% CI, 1.06–2.07), repolarization abnormalities (manifested by a wider QRS/T spatial angle), and higher resting heart rates. These results are consistent with and help explain the higher incidence of cardiovascular disease previously reported in the treatment-interruption arm of the study.
Importance of Underlying Hepatic Disease
New information also emerged regarding the importance of underlying hepatic disease. Of the nearly 5500 participants in SMART, 922 (17%) were coinfected with hepatitis B virus (HBV), hepatitis C virus (HCV), or both [Abstract TUAB203]. The relative effect of treatment interruption on risk for opportunistic infection or death was the same in these patients as in monoinfected patients. However, about half of all non-AIDS–related deaths in SMART occurred among coinfected patients. The overall risk for non-AIDS–related death was 3.8 times higher in coinfected patients than in monoinfected patients, although hepatitis itself did not explain this difference. Based on these findings, investigators deemed treatment interruption to be "particularly unsafe" for patients coinfected with HBV or HCV.
Published in AIDS Clinical Care September 10, 2007