Plasma and cervicovaginal secretion (CVS) samples were collected from 19 human immunodeficiency virus type 1-infected women on lopinavir-or indinavir-containing regimens. Lopinavir and indinavir were detectable in 29 and 93% of CVS samples, respectively, a finding that may be ascribed to these drugs' differences in protein binding and pK a . The relationship between lopinavir and indinavir pharmacodynamics and viral evolution in the female genital tract should be assessed over time.Differential viral load suppression and rate of emergence of drug resistance-associated viral mutations have been observed between blood and genital compartments, a fact which has possibly been related to inadequate penetration of antiretroviral drugs into sanctuaries (3,4,9,11,12,14,15). Numerous studies have evaluated the penetration of antiretrovirals into the male genital tract (7,8,16,17). In contrast, data on the diffusion of antiretrovirals in the female genital tract are scarce, but preliminary results evidenced major differences between drugs in terms of detection in cervicovaginal secretions (CVS) (8,14). Protease inhibitors (PIs) exhibit variable protein binding capacities and physicochemical properties, which might influence PI diffusion into the female genital tract. We thus decided to compare the pharmacokinetics of indinavir alone or combined with ritonavir and that of lopinavir-ritonavir in the plasma and genital secretions of human immunodeficiency virus type 1 (HIV-1)-infected women.HIV-1-infected women who were on a lopinavir-or indinavir-containing regimen for a minimum of 4 weeks were recruited between October 2001 and July 2002. They had no symptoms of genital infection and no genital bleeding during the 5 days preceding the visit and were asked to avoid sexual intercourse and intravaginal medications within 2 days before sample processing. To assess whether the patients were adherent to their treatment, they were asked to fill out a simple self-administered questionnaire regarding their drug intake during the 4 days preceding the study visit. After the introduction of a speculum, genital secretions were collected by a lavage performed with 3 ml of phosphate-buffered saline, pH 7.2, containing a 10 mM concentration of lithium chloride. Within 15 min after collection of the genital secretion, a blood sample was taken for the measurement of plasma lopinavir or indinavir concentration. None of the women had taken drugs susceptible to interaction with the metabolism of either indinavir or lopinavir. For each woman, samples were precisely collected just before and 3 to 4 h after intake of the last drug for measurements of 12-h trough and peak plasma concentrations simultaneously with that of drug levels in genital secretions. Both peak and trough concentrations were measured at two different visits separated by 1 to 3 months. The study protocol was approved by the Ethics Committee of the Pitié-Salpêtrière Hospital, Paris, France, and signed informed consent was obtained from all women. Measurements of antiretroviral d...