Emerging data from a study that came to a grinding halt more than three years ago may reopen a door that most people assumed was firmly closed: the possibility of HIV treatment interruptions. The latest report has at least one activist questioning whether the death knell of treatment interruption research was sounded too soon.

The debate began with the premature January 2006 termination of a 6,000 person international clinical trial examining the benefits and safety of intermittent antiretroviral (ARV) therapy. That study, called the Strategies for Management of Anti-Retroviral Therapy (SMART) trial, hypothesized that people would fare better with periods off medications that have a long list of side effects. It halted enrollment and ended the treatment interruption arm early because a safety board monitoring the study found that people who interrupted their ARV treatment did significantly worse than people who stayed on continuous therapy. Specifically, people who stopped treatment when their CD4 cells climbed above 350 and then resumed it when their CD4 counts fell below 250 were more likely to develop a serious illness or die than people who stayed on treatment without interruption.

Since the SMART study’s surprisingly early conclusion, researchers and care providers have largely soured on the idea of people with HIV going off treatment for any length of time. However, a letter published last month in the journal Lancet Infectious Diseases (LID)—based on an October 2008 revelation by the SMART researchers in the online journal PLoS Medicine—suggests that treatment interruptions might be possible for some people living with HIV.

According to the PLoS article, SMART participants outside the United States appeared to do much better when they interrupted therapy than those enrolled at U.S. sites. Unfortunately, no possible explanations were provided for this curious discrepancy. Considering that such explanations could potentially help researchers ferret out the factors associated with a more favorable response to an on-again, off-again treatment strategy, two London-based researchers—Justin Stebbing, PhD, and Angus Dalgleish, MD—submitted a letter to LID with thoughts of their own, essentially publicizing a key finding that was going unnoticed and unacknowledged.

Mike Barr, a longtime AIDS activist currently living in Los Angeles, has been trying to challenge the stranglehold that SMART seems to have placed on people’s imaginations since the study results were first made public. Barr welcomes the LID letter and hopes it will reopen the door to an alternative treatment strategy allowing people to avoid continuous lifelong ARV therapy and the potential burden, side effects and high cost it can bring.

Explaining the Controversy

When Lewis Kuller, MD, and his colleagues from the SMART study published their article in PloS Medicine, they likely had no idea they’d spark such a controversy. In their analysis, which looked at markers of inflammation in SMART study participants, they included data for the first time showing that there were 79 deaths among U.S. participants compared with only six in non-U.S. participants. This is the information that caught the attention of Stebbing, of Chelsea and Westminster Hospital, and Dalgleish, of St. Georges University.

In their LID letter, Stebbing and Dalgleish parse the data further and end up with the conclusion that “patients on [ARVs] in the USA had [a significantly] increased risk of dying when compared with non-U.S. patients, irrespective of whether the non-U.S. patients were receiving [treatment] or not.”

Stebbing and Dalgleish also ask a number of questions not covered in Kuller’s article. Were there differences in the level of care provided to U.S. and non-U.S. study participants? Were people in the U.S. more likely to also be infected with hepatitis C? Were their immune systems more depressed than non-U.S. patients? Were there other cofactors not accounted for, and why didn’t the SMART researchers attempt to explain those differences?

James Neaton, PhD, of the University of Minnesota and one of the SMART study’s chief statisticians, confirms that the SMART team has sent a letter to LID answering Stebbing’s and Dalgleish’s questions. As that letter has not yet been published, however, Neaton could not disclose what it says. When asked if researchers had prematurely pulled away from treatment interruption research, as Barr contends, he responds by saying that the Department of Health and Human Services HIV treatment guidelines “have a very balanced statement about interruptions.”

Though the guidelines don’t forbid further treatment interruption research, they do state that it should never be conducted outside a clinical trial. They go on to recommend that people who are relatively new to HIV treatment not attempt to go off therapy for longer periods, even in a study, unless they currently have a CD4 count above 350. Lastly, they acknowledge that some small studies have shown interruptions to be safe, but that further research would be needed to confirm those results. Neaton agrees with the guideline’s recommendations, and further explains that any future studies should also look carefully at the risks for other illnesses in prospective study participants and that very close and careful monitoring should occur during the study.

Barr doesn’t feel, however, that the guidelines handle the issue of treatment interruptions in a balanced manner. He feels that the risks are too broadly extrapolated to all patients and that studies with negative outcomes are given much more precedence and attention than studies with positive outcomes. He feels we should be “tailoring our therapeutic approach to the specific situation of each HIV infected individual.”

Even if this exchange reinvigorates interest in treatment interruption research, Barr fears it will still be an uphill battle. He remains frustrated that SMART has already had such a profound effect on an intermittent therapy research agenda and that other studies—like one published in AIDS in April—showing treatment interruption to be safe have been largely ignored by scientists and care providers. He is particularly piqued that experts have applied the SMART study to every person with HIV, without questioning whether intermittent therapy or treatment interruptions may be a feasible possibility for those meeting certain criteria, such as higher CD4 counts, lower viral loads, the absence of other health problems or even smoking history.

“I guess you could say that SMART told you the worst way, and in the worst group of patients, to try treatment interruptions,” Barr says. “Using those ridiculously low triggers, like 250 and 350 CD4s, and then doing treatment interruptions in people with all of these [other health] conditions…the conclusion was kind of a no-brainer.”

Whether this exchange of research letters will spark new life into treatment interruption research remains to be seen, but at least it has renewed dialogue about the strategy. This is a welcome development to Barr, who concludes, “It’s really, really hard to get any press these days for a message that isn’t early treatment all the time.”