STIs: more questions than answers

At Day 3 of the 8th Retroviruses Conference being held in Chicago yesterday, delegates focussed their attention on ‘structured treatment interruptions’ (STIs); the name given to supervised breaks in anti-HIV therapy which became a hot topic in HIV research around a year ago.

At slide and poster presentation sessions, a number of groups presented new information on the use of STIs in many contexts. Broadly speaking, there are several reasons why breaks in treatment are under investigation. On the one hand, some hope that the short rebounds in viral load which occur when someone who is virologically suppressed on HAART stops treatment may act as a type of ‘auto-vaccination’, priming the immune system to better control HIV. On the other, treatment breaks result in less overall exposure to drugs with all the presumed advantages which that implies, such the possibility of less toxicity, of cost savings and improved quality of life.

Unfortunately, many of these studies are uncontrolled, and consider small numbers of patients and short periods of follow-up. The variance in treatment cycles, disease stage, and indeed their findings leave them looking something of a hotch potch at present, prompting Marty Markowitz, reporting data from David Ho’s Aaron Diamond Centre, to remind delegates that both STIs, and treatment during in primary infection must continue to be seen as legitimate avenues of research, rather than legitimised treatment protocols.

In this context, randomised, controlled trials are particularly welcome, and extended follow-up from one such study being conducted by the US National Institutes of Health was of interest. 70 people on HAART with viral load below 500 copies for at least three months (and below 50 copies at the time of screening), and CD4 counts above 300, were randomised to remain on continuous therapy or to undergo STIs. Cycles were of four weeks off HAART, followed by eight weeks on, for 22 months.

Data is available for 47 enrolled patients, 24 of whom were randomised to STIs, two of whom later chose to withdraw. Median CD4 nadir was between 150 and 693 cells in the control group, and 22-617 amongst the interrupters. Median CD4 at enrollment was 430-1194, and 356-1419 respectively. Pre-HAART viral load was 61-868,550 copies and 650-451,375 respectively.

Following several cycles, no clear pattern of viral load response was observed. Eight of 20 interrupters had at least one viral load result over 50 copies after 8 weeks of HAART prior to an interruptions, and nine of 24 had at least one value over 50 copies during 3-4 weeks of continuous HAART. In comparison, in those on continuous therapy, 15 of 23 had at least one viral load over 50 copies during visits in the study period.

Rebound was observed in all interrupters during periods off therapy, but 19 of 20 who have undergone 2-6 cycles returned below 500 copies on eight weeks of HAART. Though the interrupters took 30% less therapy, there was no difference in the number of participants whose viral load rose above 50 copies after 8 weeks of HAART across groups. Though CD4 counts fell in periods off treatment, cells were typically regained, returning to baseline levels following four weeks of treatment.

This cohort remains under investigation and further data on the longer-term effects of this strategy will be reported in due course.