Background:  Acute HIV-1 infection is characterized by unchecked rounds of viral replication with resultant massive CD4⁺ T cell depletion and immune activation. Cyclosporine A (CsA) suppresses the cellular immune response and has been reported in uncontrolled studies to result in superior T cell levels in patients given ART and CsA during acute and early HIV infection. We hypothesized that immune modulation during acute and early infection in conjunction with ART would have virologic and immunologic benefit.

Methods:  A multicenter, 48-week open label randomized phase II study with a 2:1 randomization to 4-weeks of CsA in combination with zidovudine (ZDV)/lamivudine (3TC)/abacavir (ABC) and lopinavir (LPV)/ritonavir (RIT) (Arm A) vs ZDV/3TC/ABC/LPV/RIT (Arm B). CsA dosing was based on ideal body weight using a liquid formulation dosed at 0.3 mg/kg orally twice daily with levels monitored weekly targeting serum concentrations of 250 to 450 ng/mL. Inclusion criteria included patients with a plasma log plasma viral load >4.7 and a Western blot with 5 bands or less within 28 days of study entry. Antiretroviral regimens could be modified as per the individual investigator however all patients had to remain on Kaletra-based ART for the first 4 weeks of therapy. Analyses were as-randomized, with completers only included in analyses, and the primary endpoint was the level of proviral DNA in peripheral blood mononuclear lymphocytes (PBMC) at week 48. A sample size of 45 subjects provided 85% power to detect 0.6 log difference  between the groups. Secondary endpoints included change in CD4⁺ T cell count from baseline.

Results:  We randomized 54 patients to either Arm A (n = 36) or Arm B (n = 18); 13 discontinued treatment, leaving 41 subjects available for analysis (39 completed week 48, 2 have completed week 44; Arm A, n = 28; Arm B, n = 13). Baseline CD4⁺ T cell count and log plasma viral load were 407 copies/mL and 5.0 for Arm A and 490 copies/mL and 4.9 for Arm B (p = 0.2 and 0.9 respectively). Proviral DNA levels (log copy /10⁶ CD4) at weeks 12, 24, and 48 in 21 subjects (Arm A, n = 14; Arm B, n = 7) were 2.5, 2.2, and 2.0 in Arm A and 2.2, 2.0, and 2.0 in Arm B (p = 0.3, 0.3, 0.9). The mean time to HIV-1 RNA below 50 copies/mL was 15 weeks in Arm A and 16 weeks in Arm B (p = 0.9). CD4⁺ T cell counts at weeks 12, 24, and 48 were 576, 629 and 662 copies/mL in Arm A (n = 28) and 616, 643, and 773 copies/mL in Arm B (n = 18) (p = 0.7, 0.8, 0.9). The change from baseline in CD4⁺ T cell count was 301 copies/mL in Arm A and 287 copies/mL in Arm B (p = 0.9).

Conclusions:  We conclude that 4 weeks of CsA dosed orally and targeted to 250 to 450 ng/mL in conjunction with ART fails to add apparent immunological or virological benefit in patients treated during acute and early infection.