SEATTLE – The general public and HIV researchers alike were excited by the report out of the Conference on Retroviruses and Opportunistic Infections (CROI) this week that a second patient, the London patient, has been in long-term remission from HIV after receiving a bone marrow transplant with HIV-resistant cells to treat cancer.

The consensus among researchers, though, was that the London patient should not yet be considered cured. To most medical experts, in analogy to the cancer field, it would take five years of undetectable virus in the absence of antiretroviral treatment to consider a patient cured.

Even if the London patient, who is currently roughly at the 18-month-mark, sustains remission for another 3.5 years, it is "really not clear... how to get to remission for most individuals," John Mellors, professor of medicine at the University of Pittsburgh, told reporters at a CROI press conference.

"There is no roadmap here, there is no known timeline," he said. "We have to work our way along."

A key part of that work comes in the form of analytical treatment interruption (ATI), the discontinuation of ART.

The most obvious goal of ATI is to see whether the immune system of a patient, or primate, can control the virus in the absence of treatment. For cure strategies, "stopping ART is the litmus test, and looking at a variety of tissues of any sign of relapse," Mellors said.

But ATI can be informative even if viral rebound occurs. At this point, studies that use ATI as a strategy basically expect rebound – though surprises are always possible. The VISCONTI study, the first study to demonstrate delayed rebound after early initiation of ART and subsequent ATI, found a higher proportion of long-term controllers than expected.

VISCONTI was "a bit of a game-changer," University of Oxford's John Frater told the audience at a session on "Insight from analytical treatment interruptions." Prior to VISCONTI, it was essentially a dogma of the field that "when you interrupt treatment, the virus comes back, and it comes back quickly... this was the first bit of evidence to suggest that in some people, that is not the case."

Even when rebound occurs, the nature and timing of that rebound can teach scientists much about the reservoir, including how time on ART and size of the viral reservoir affect the time to rebound.

One important question is the location of the reservoirs that fuel rebound even after the virus has been undetectable for years. At the session chaired by Frater, researchers shared their findings about two quite different anatomical sites of putative reservoirs: the brain and the male genital tract.

Richard Price, professor of neurology at the University of California at San Francisco, presented data on rebound virus in the central nervous system (CNS). The brain is a suspected reservoir site, but Price's analysis of virus in the cerebrospinal fluid (CSF) of patients who rebounded after either treatment interruption or treatment failure suggested that the virus, though it was detected in the CNS, originated from the blood.

"Rebound virus in the CSF is most easily accounted for as coming from CD4 T cells trafficked from blood," he told the audience, though he pointed out that there may be a smaller CNS reservoir that his team missed in the current study, because it was dwarfed by the amount of blood-derived HIV.

The brain may get its virus from the blood, but the male genital tract seems to have its own. That was the conclusion of data presented by Sara Gianella, assistant professor of medicine at the University of California at San Diego's Center for AIDS Research.

Gianella and her team looked for virus in the sperm of patients undergoing ATI for 16 weeks. Though rebound in semen was delayed compared to that in blood, and not sustained in most cases, sequencing showed that the molecular diversity of HIV was higher in semen than in blood, suggesting "a distinct evolutionary dynamic [and] unique viral populations in semen," she told the audience.