Compartmentalization of HIV-1 within the female genital tract: implications for assessing antiretroviral interventions.

Abstract

Compartmentalization of HIV-1 between the blood and genital tract has been suggested for both men and women.1,2 The source of genital virus has been ascribed to infected cells in the genital-associated lymphoid tissue of the vagina and cervix submucosa although there may be a small contribution from the blood compartment.3 Nevertheless, the opportunity for local factors, such as immune activation and hormonal regulation, to affect viral shedding from the genital tract is expected but requires further study. We assessed short-term shedding and natural variability of HIV-1 RNA level from the blood plasma and genital tract of 55 HIV-1-infected women (median CD4 cell count of 209/μL, range 42-624) over 8 weeks.4 Specimens were collected weekly from the endocervical canal using Sno-strip™ wicks and cytobrushes and from the ectocervix with cervicovaginal lavage. The majority of subjects were either Black or Hispanic (85%), and heterosexual contact was a risk factor in 60%. Three-quarters of the women received combination antiretroviral therapy that contained a protease inhibitor. Short-term, serial, quantitative measurements of genital tract HIV-1 showed that the accurate detection of virus depended on the use of multiple sampling methods. For example, repeated sampling of genital fluids increased viral RNA detection from approximately 60%, reported in other studies, to 90% of HIV-1-infected women in our study.3 Moreover, approximately one-half of women shed HIV-1 continuously from the genital tract when sampled weekly for two months. Regression models were developed to predict the blood and genital HIV-1 RNA levels as a function of the menstrual cycle. Significant mean differences in genital tract HIV-1 RNA levels from menses were seen for all genital sampling methods; however, in contrast to some other studies, we found no differences in peripheral blood.5 The general pattern of HIV-1 shedding showed the lowest levels of viral RNA in the follicular phase immediately following menses with a gradual increase throughout the luteal phase until the next menses. Factors associated with HIV-1 genital tract viral shedding depended on the genital subcompartment sampled and include antiretroviral therapy, increased plasma viral RNA and the presence of other genital tract infections. The effect of menses on viral RNA shedding was most pronounced for specimens collected by cervicovaginal lavage but less so from the endocervical canal. For example, cervicovaginal lavage specimens collected during menses were more likely to have detectable viral RNA compared to specimens collected in the week immediately following menses (shedding odds ratio (OR) 0.23; 95% CI, 0.12-0.45; P < 0.001). In contrast, endocervical canal fluid viral RNA levels were influenced less by menses but trended towards having more viral shedding in the week prior to menses (shedding OR 1.94; 95% CI 0.96-3.91). Importantly, endocervical canal fluid viral RNA levels were significantly higher compared to either menses (0.2 log10 RNA copies/mL; P = 0.03) or blood plasma (0.3 log10 RNA copies/mL; P = 0.03). This association between the progesterone responsive luteal phase and increased viral RNA shedding may represent either endocervical canal specific viral production or, possibly, cell-free or cell-associated viral enrichment. These data further support viral compartmentalization between the blood and genital tract. We also found that there was greater natural short-term variation of HIV-1 RNA in the genital tract (13-fold) compared to blood (7-fold), which has important implications for monitoring genital tract-associated HIV-1 level [1]. For women, the variation in genital viral RNA was less than that reported for men for which the variability of seminal plasma viral RNA is approximately 38-fold [1]. This gender-specific viral RNA variation is likely due to differences between the genital microenvironments of men and women. However, in contrast to the male genital tract for which the basis of compartmentalization is largely defined by anatomical, immunological and microbiological barriers, the female genital-tract compartmentalization has a hormonal and microbiological basis. In conclusion, the precise measurement of HIV-1 in the female genital subcompartments (i.e., endocervical canal and ectocervicovaginal fluid) may shed further light on the genital source of HIV-1 and allow for assessment of the local hormonal, immunological and inflammatory effects on HIV-1 shedding. A better understanding of the genital viral shedding patterns is necessary for studying the effects of systemic and topical antiretroviral therapies to prevent both vertical and horizontal transmission of HIV-1.