Background:  Use of interleukin-2 (IL-2) before ART interruption may lead to a higher CD4 count while off ART and prolong the time off ART in any CD4 count-guided interruption strategy.

Methods:  We performed a randomized trial in patients with viral load <50 copies/mL on stable ART with current CD4 count >300 /mm³, and nadir >100 mm³. Subjects were randomly allocated to arms A, B, or C. Arm A received continuous ART—if drugs needed to be switched, it was done without interruption. Arm B continued with ART for 9 weeks, then interrupted, and restarted with the same ART when the CD4 count fell <200 cells. Further treatment interruption was considered when viral load <50 copies for >12 weeks and when CD4 count was at or above the baseline level. For Arm C, 2 cycles of IL-2 (4.5 MU) were given 8 weeks apart, while still on ART. ART was stopped at week 9, and a new cycle of IL-2 was used alone whenever the CD4 count falls <300 cells. If the CD4 count fell below 200 cells, ART was restarted. Further treatment interruption after 2 cycles of IL-2 was considered when viral load <50 copies for >12 weeks and when CD4 count was below baseline level. The last visit was at week 105.

Results:  The study ran from March 2001 to July 2006, during which, we randomized 86 patients:  27 in arm A, 28 arm B, and 31 in arm C. Subjects were mostly white men who have sex with men (MSM) with a mean age of 40 years (sd 7.8), baseline median CD4 count of 754 /mm³ (range 240 to 1400), and median nadir CD4 count of 268 /mm³. New ART drugs were used in 60% in arm A, 57% in arm B, and 45% in arm C (p >0.05). By 96 weeks from interruption of ART (ie, week 105), there was a 66% probability of having restarted ART in the interrupt-only arm (arm B) compared with 34% in the interrupt-with-IL-2 arm (arm C) (p = 0.002; log rank test). At week 9, differences in median CD4 count from baseline were –10, –30, and +321, respectively (p <0.0001). In contrast, at week 105 they were –24, –212, and –268 (p = 0.0002) and the proportion on ART was arm A 96%, arm B 61%, and arm C 35% (11 of 31);  χ² p = 0.05 comparing arms B and C; 3 patients in arm C experienced a CD4 count <100 /mm³. There were no new AIDS events or deaths.

Conclusions:  Use of IL-2 before interruption of ART was associated with a prolonged period off ART compared with interruption without IL-2, although CD4 count outcomes were also slightly, but not significantly, poorer.