8th Annual Retrovirus Conference
Strategic Treatment Interruptions in Primary and Chronic HIV infection
Reported by Christopher D. Pilcher, M.D. of the UNC Center for AIDS Research, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Investigators have envisioned three basic settings in which supervised (ìstrategicî) treatment interruption could benefit HIV infected patients: 1) primary HIV infection (PHI; individuals who have only just been infectedósee separate report), 2) successfully treated chronic infection, and 3) chronic infection in patients failing HAART with multi-drug resistant virus.› There are different rationales for STIs in these distinct situations.
First considering patients with primary HIV infection: theoretically, this tiny subgroup of people living with HIV is just the group one might imagine could benefit most radically from only a brief course of HAART, or multiple intermittent courses of HAART separated by STIs. The argument goes that people with PHI have basically intact immune systems, but immune systems under onslaught by the virus at the same time.›› In part because of HIVís attack, they often fail to achieve adequate immune control of the virus.› CD4 cells that should be able to fight HIV are almost universally killed during PHI since they are preferentially activated and thus very susceptible to killing by their intended target (the virus).› Theoretically, brief or intermittent HAART initiated during PHI could shield these newly growing CD4 cell populations from interference by HIV long enough to enable potent anti-HIV immune responses to develop in treated patients.› The idea is therefore to boost innate immunity and possibly make all patients with PHI into ìlong term slow progressorsî that can live long, healthy lives without needing to take antiretroviral drugs or getting sick.
STIs in PHI
Bruce Walker from Harvard presented updated data from the MGH/Brigham/Fenway trial on Tuesday in his invite ìState of the Art Lectureî (no abstract available).› Preliminary results from the trial have been reported this fall in an article by Eric Rosenberg in the journal Nature.› These highly selected patientsóall with acute PHI, with a median viral load of 10 million copies per mlóhave had some encouraging success in achieving at least temporary control of their viremia using treatment interruption.› The protocol calls for stopping treatment when viral loads are confirmed to be undetectable (<50 copies) and restarting it if, and only if, viral loads rebound to consistently above 5,000 copies for 3 weeks.› Of 14 patients starting the trial, 6 have never rebounded consistently above 5,000 and 8 have had to restart.› These 8 again went undetectable on the second course of HAART, and a second interruption saw only 1 subject going consistently > 5,000 copies and having to restart.›› Control was associated with dramatic increases in both CD4 and CD8 cell virus-specific immune responses following interruptions.›› This control has lasted for up to 450 days in the longest-observed subject, but with a couple of subjects resuming therapy outside the protocol andóominouslyówith two other subjects experiencing late high-level rebounds in viremia after a few months of suppression.› The results are dramatic in thatófor the first timeóthere is a suggestion that a treatment protocol can improve immune control in a group of patients and enable them to be free of drug treatment.› But there remain important concerns.› First, these types of interruptions cannot be expected to have anything like similar results in chronically infected people.› Second, even in this group with very early PHI, the proof of the benefit of STIs will be in whether they result in durable suppression of virus without HAART.› So the jury is out.
Marty Markowitz from the Aaron Diamond group [Abstract 288] used a different protocol, including patients much later in PHI (up to four months out) with lower entry viral loads (median 40,000 copies) and treated longer (2 to 5 yrs) on HAART.› Subjects stopped therapy but there were no provisions for restarting. ›Only one of 14 patients ever ended up controlling viremia according to their criterion of <1,000 copies and 7 of 14 elected to restart.› Sorting out the reasons for the differences between outcomes in the MGH and Aaron Diamond studies is going to be a struggle in the research community in the next few years.
STIs in Chronic Infection: discouraging so far
7 Days On-7 Days Off & Therapy Rest
In chronically infected patientsói.e., everyone except those with PHIóthere remain no convincing data to support using treatment interruptions for the purpose of inducing virus control.› There has been some hope that STIs could be useful in chronic infection, however, based on the theory that by exposing patientsí immune systems to brief surges of virus during brief STIs, one could stimulate ravaged immune responses to better control HIV off therapy (so-called ìauto-vaccinationî)
A muted note of encouragement came from a presentation of Ruiz and coworkers from Barcelona [Abstract 291], who demonstrated in a group of 12 chronically infected patients that 3 cycles of brief interruption could modestly boost CD8 immune responses to HIV and that these seemed to slow the ìdoubling timeî (growth rate) of HIV after interruptions.› This trial did not allow patients time to control their viremia during interruptions since the protocol called for treatment to be reinitiated within 30 days or whenever the virus went >3,000 copies per ml (virtually all of Walkerís patients had rebounded to >3,000 copies per ml during the first weeks of interruption, but many of these subsequently came down.› Even they would have been restarted according to this study design.)› A few STI studies did allow chronic patients time to control viremia, and in each case, all patients had virus appear during STIs and resuppress with reintroduction of therapy.› In the Swiss-Spanish trial [Abstract 357] 128 patients (some chronic and some with primary infection) underwent 4 cycles of interruption after being suppressed to <50 for at least 6 months.
When they broke the data down, 9 of 54 patients completing 52 weeks on protocol were successfully controlled at <5,000 copies for at least 12 weeks off drug, but almost all of these actually had evidence of primary infection.› However, overall both in that study and in a study by Garcia (also from Barcelona) [Abstract 289], chronically infected patients undergoing a series of planned interruptions appeared to have lower viral loads after their last interruption than they did prior to initial therapy.› But these decreases in viral load were extremely modest and there may have been some sacrifice in terms of CD4 cell countsóbut we really donít know since these were single-arm studies with no way to know what would have happened if patients had continued on HAART.
In a poster [Abstract 364], Dybul and colleagues from Tony Fauciís group unveiled an ongoing protocol that will look at 5, 4 week STIs each followed by 8 weeks back on ART for a total duration of 22 months, with a planned enrollment of 70 patients.› Unfortunately, so far they have witnessed significant CD4 declines of 17% in the 14 patients completing the first cycle of interruption.› Like the Spanish group, they saw that virus always rebounded and always resuppressed.› But they did see one of the fifteen subjects tested develop new resistance in the virus after a few interruptions.› So, the safety of the approach is very questionable.› Over the course of the sessions, several observers commented that trials are needed comparing possible CD4 cell loss and viral load between infected patients continuing HAART and undergoing STI.
Possible improvements in immune control aside, there is another important reason to consider STIs in chronic infectionónamely, reducing exposure to ARV drugs in order to limit toxicity.› Again, the obvious questions that need to be answered for this type of approach to go forward concern safety: CD4 cell loss, increases in OIs, evolution of viral resistance and an increased risk of subsequent treatment failure could all result from repeated interruptions in treatment.› Fauci himself presented some more very preliminary data in the Symposium on reservoirs Wednesday afternoon, concerning their second STI protocol using a 7 days on- 7 days off treatment schedule in chronically infected patients with viral loads already suppressed to below 50 copies at entry.› Viral rebounds >500 did not occur in the first 11 patients treated for several months, except for one who took his own ìdrug holidayî and broke the protocol schedule.› There was no information provided on toxicities, CD4 changes or long-term suppression.› So, as before, data are incomplete and preliminary but promise important results so stay tuned.››
(Editorial note: although STIs in chronic infection have not shown the hoped for immune response, an STI may still have an appeal. If we can determine how to eliminate risks, interrupting therapy may offer a period of grace from the rigorous demands of taking therapy. But this approach has yet to be studied. It's possible that such interruptions may limit the development of toxicities and side effects (lipodystrophy, elevated lipids-glucose, etc), but this has yet to be seen & also needs further study).
STIs in Persons with Extensive Drug Resistance & Detectable Viral Load
The last situation in which investigators are thinking about STIs in chronic patients is for patients who have multidrug-resistant virus on failing regimens (detectable viral load).› The thinking goes like this: when treatment is stopped in these patients, we know from previous studies that their resistant viruses are almost always replaced with old, drug sensitive viruses after a few months.› These viruses--which had been chased into viral ìreservoirsî of latently infected T cells by drugs in the blood streamóare free to well back up and take over again when drugs are no longer present.› Since these ìwild-typeî (non-mutant) viruses are more ìfitî than the relatively wimpy drug-resistant strains, they outcompete the drug resistant mutants for T cells and the resistant strains recede into the background where they now persist in lower numbers.
Theoretically one could hope that if this process went on for several months, the resistant viruses might become so low in number that they might be more easily controlled by a new regimen despite their being resistant.› Unfortunately, many studies, such as those from Steve Deeks and Veronica Miller at previous conferences, have shown that the reemergence of ìwild-typeî virus in the blood is almost universally accompanied by dramatic loss of CD4 cells.› Deeks is now able to provide the first information on outcomes of STI followed by a new salvage regimen and the results are alternately encouraging and frightening.› In terms of virologic outcomes (viral loads), patients have actually done pretty well over the short term.› 22 patients (treated on a regimen that had been failing continuously at levels >3,000 copies for an average of 3 yrs) underwent an 18 week interruption and 18 of 22 reverted to wild-type virus during that period.› When therapy was reintroduced using RTV+ another PI + 2 NRTI +/- NNRTI or T-20, 14/22 went to <50 copies/ml by 24 weeks.
Critically, 11 of those 14 were completely naÔve to one class of drugs in their salvage regimen (either NNRTI or T-20).› Clearly, the strategy worked much better for patients that had at least the possibility of one active drug in their regimen.› Overall, viral loads dropped 3 logs from their STI set point with therapy introduction and the CD4 losses that were observed during STI were recovered to baseline levels in 18/22 patients.› The most troubling part of the presentation, though, was that 5 subjects had significant AIDS defining illnesses or died during the time of CD4 loss associated with the STI.›› Clearly, an STI in this setting is frought with danger and the potential clinical benefits of this type of strategy remain to be proven.(Editorial note: During the question & answer session following Deeks' presentation it was suggested that the 11 patients described above may have remained sensitive to other drugs used in the salvage regimen & resistance testing was not used to try to answer this question).