Can the SMART Study Data Be Used to Assess Risk Factors for Renal Disease?

IN RESPONSE:

We agree with Dr. Gupta that the renal outcomes before and after the protocol modification in SMART are interesting and require further study. The interaction P value corresponding to the treatment hazard ratio comparison before and after the protocol change was significant (P = 0.014); however, the number of participants with renal events, defined in SMART as death from renal disease or end-stage renal disease, was too small (a total of 18 across both treatment groups) to reliably study predictors. A careful study of risk factors for renal disease in the SMART study requires more events. Thus, we are exploring the possibility of using stored plasma samples to measure creatinine in SMART participants over the entire follow-up. With these data, an expanded renal outcome (for example, death due to renal failure, end-stage renal disease, or large decline in estimated glomerular filtration rate) would result in more events and would allow reliable assessment of risk factors for renal disease, as suggested by Dr. Gupta.

Our finding that treatment interruption increases risk for renal progression is supported by another recent investigation in SMART (1, 2). Stored samples were used to measure cystatin C, a marker of renal function, during the first year of the study but before the protocol change. Cystatin C levels increased significantly in the treatment interruption group compared with those in patients randomly assigned to receive continuous antiretroviral therapy (2).

Ultimately, we think that the risk and benefits of antiretroviral treatment are best assessed in a randomized trial of early therapy instead of a treatment interruption study, such as SMART. A trial called START (Strategic Timing of AntiRetroviral Therapy) is scheduled to begin next year and is designed to investigate the risks and benefits of early antiretroviral treatment on clinical outcomes, including renal disease, and other serious non-AIDS conditions, such as cardiovascular disease, liver disease, and malignant conditions.