COMMENTARY sation in therapy, the longest of whom is now controlling for 18 months. After retreatment and a second interruption, the other five were able to transiently control viremia, with two people persistently controlling. The interruptions in therapy were associated with strong enhancement in both the breadth and magnitude of cellular immune responses to HIV. Although these data are encouraging and highly statistically significant compared with historical controls untreated during acute infection21, they do not address whether a true clinical benefit is conferred, nor do they indicate how durable this control will be.

The above study21 does suggest that STI might be more beneficial than simple treatment cessation—a conclusion supported by other recent reports in which subsequent immune control occurred but was the exception in cases of treatment termination following HAART in early infection22, 23. Full comparison of these studies is difficult because the groups were different at the time of initiation of treatment; the average viral load in one study was of 10 million RNA copies per ml plasma21, and less than 100,000 copies in another study23. The cohorts with lower average viral loads were studied at a later phase of infection, and thus might have incurred more initial immune damage prior to therapy.