Background: The HIV RNA viral load (VL) in vaginal secretions and semen is an independent predictor of HIV transmission. Blood VL is associated with semen VL, and local mucosal factors, such as semen cytomegalovirus (CMV) reactivation, may play an important role.
Methods: Twenty-one HIV-CMV–coinfected, antiretroviral-naive men received 900 mg of oral valganciclovir once daily for 2 weeks in an open-label study. Blood and semen were collected at baseline, after 2 weeks of valganciclovir, and 2 months after therapy completion. The primary end point was change in semen HIV levels at 2 weeks, and the secondary end points were change in semen HIV VL at 2 months and change in semen CMV levels.
Results: The HIV VLs fell significantly at 2 weeks in semen (median 3.44−3.02 log10 copies/mL, P = 0.02) and blood (median 3.61−3.10 log10 copies/mL, P < 0.01) and returned to baseline after therapy completion (median 3.24 and 3.71 log10 copies/mL in semen and blood, respectively). Semen CMV levels also fell on treatment (median 2.13−1.62 log10 copies/mL, P < 0.01) and continued to fall after therapy completion (median 0.91 log10 copies/mL at week 8, P < 0.001 vs. baseline). The reduced semen CMV VL was associated with decreased semen T-cell activation and enhanced CMV-specific T-cell responses in blood; changes in the semen HIV VL were not associated with immune parameters.
Conclusions: Although valganciclovir therapy was associated with reduced HIV and semen CMV levels, these results suggest that the reduced HIV VL was a direct drug effect rather than a CMV antiviral effect or CMV-associated immune alterations.
*Clinical Sciences Division, University of Toronto, Toronto, Ontario, Canada;
†Deptartment of Pathology and Molecular Medicine, Queens University Kingston, Ontario, Canada;
‡Maple Leaf Medical Clinic, Toronto, Canada;
§University Health Network, Toronto, Canada; and
¶Department of Immunology, University of Toronto, Toronto, Ontario, Canada.
Correspondence to: Lucy Y. Shin, MSc, Clinical Science Division, University of Toronto, Medical Sciences Building #6356, Toronto, Ontario, Canada M5S 1A8 (e-mail: firstname.lastname@example.org).
L.Y.S. and P.M.S. contributed equally to this work.
The authors have no conflicts of interest to disclose.
Supported by grants from the Canadian Institutes of Health Research (R.K., HET-85518 and MOP-115020) and the Ontario HIV Treatment Network (R.K.; salary support).
Received September 04, 2013
Accepted September 04, 2013