HIV Infection in Women

     Reported for NATAP by Kathleen Squires, MD, Director of the University of Southern California HIV Clinic & Research

A search of the official CD-ROM of the program and the abstracts presented at the conference utilizing the words "woman" or "gender" reveals approximately 100 abstracts. In reality, the number of abstracts that actually cover subjects specific to the manifestations of, course of disease, and efficacy and/or toxicity of antiretroviral therapy in HIV-infected women is a much more modest twenty-six. If one adds associated conditions such as cervical dysplasia and genital HPV infection, the total rises to thirty-one. This relative paucity (scarcity) of data needs to be considered in light of the fact that women comprise the majority of affected individuals in Sub-Saharan Africa and up to 40% of recent infections in some sections of the United States. Despite clear evidence that a number of medical conditions-perhaps the best studied is heart disease-differ significantly in male and female patients, the assumption is that there is no substantial difference in HIV infection between the sexes. A number of observations from clinical trials or cohort studies do however suggest gender-based differences in a number of areas ranging from viral load to toxicities and side effects to utilization of health care resources. Prospective hypothesis-driven studies are needed to determine the significance of these findings. There were presentations that covered a number of topics:

Heterosexual transmission

Studies involving highly exposed but sero-negative women (Abstract 51) suggest that the major difference between them and women who do become infected through ongoing monogamous relationships is their ability to generate HIV-specific activity in immunologic assays. Review of factors that may affect heterosexual transmission in both discordant couple cohorts and mathematical models document that the level of viral load and history of or presence of an STD in the partner (Abstracts 221 - 222 - 223) are associated with an increased risk of transmission. These cohort studies have been instrumental in defining host genetic, virologic and immunologic correlates of transmission.

The identification of topical microbicides that allow women to control their contraceptive choices is an important area of research. Preliminary data was presented on PRO 2000, a gel antimicrobial compound that is active against HIV-1, HIV-2, herpes simplex viruses, chlamydia and gonococci. In a study utilizing female rhesus macaques, the majority of animals treated with 3 different concentrations of PRO 2000 remained infection free in contrast to all animals in the control group who were uniformly infected with SHIV within two weeks of challenge (Abstract 721).

(Editorial comment: In abstract 720, it was reported that ß- cyclodextrin (CD) is a cholesterol-removing compound that has effects on cell membranes, specifically on lipid rafts, the sites from which HIV-1buds from infected cells. The study reported that CD (3%) is non-toxic to vaginal epithilium of theHuPBL-scid mice used in this test tube experiment and inhibits the vaginal transmission of cell-associated HIV-1 in this mouse model. Authors concluded CD warrants further evaluation as a vaginal microbicide to prevent HIV-1 transmission).

Disease Progression and Gender Differences in Viral Load: hard-drug use, post-menopause & pregnancy affect on CD4

The use of illicit drugs has been implicated in accelerated HIV disease progression. Women participating in the Women and Infant Transmission Study (WITS) were asked about their use of drugs as part of the research protocol. CD4 cell counts and HIV viral loads in women who reported hard-drug use were compared to the levels in non hard-drug using women. The use of hard-drugs (heroin, methadone, cocaine, IVDU) did not appear to affect these immunological or virological parameters, suggesting that hard-drug use does not adversely affect disease progression (Abstract 204). One recurring question is whether the variations in estrogen and progesterone that occur in pregnancy or menopause impact HIV disease progression. CD4 cell counts in 487 women participating in the European Women Study were analyzed in relation to the woman's hormonal status. Postmenopausal and pregnant women had lower CD4 cell counts than pre-menopausal or non-pregnant women. These changes may reflect the changes in levels of reproductive hormones and may help to explain the gender difference in CD4 cell counts that has been reported (Abstract 205). Other studies, however, have documented that CD4 cell counts in pregnant women rapidly return to baseline after delivery; additional studies are needed to determine the implication of varying hormone levels on CD4 cell counts, viral load and long-term outcome. In the only session that offered oral presentations on HIV infection in women, Dr. Pitt reported on gender differences in immunologic and virologic markers in children born to HIV-infected women (Abstract 513). The import of this study lies in the finding that CD4 cells in HIV-infected and uninfected female babies were significantly higher than CD4 cells in both groups of male babies and HIV RNA levels were lower in infected female compared to the male babies (not statistically significant). These findings parallel those seen in HIV-infected adults and suggest that non-hormonal mechanisms need to be considered since children of this age have similar hormonal environments.

Differences in access to care, prescription of HAART-based regimens, and HIV progression by sex and injection drug use persist (Abstracts 488 and 494). Although the use of HAART-based regimens has increased in all patient groups over the past several years, the gains have not been equally distributed over all the groups. A common theme is that certain populations of patients-women and/or African-Americans and/or IDUs-are less likely to benefit from the improvement in long-term outcome. HIV-infected women are disproportionately represented in these patient populations.

(Editorial comments: abstract 488 reports that although women & IVDUs clinically benefit from HAART, women received less clinical benefit from HAART than men & IVDUs received less clinical benefit from HAART than non-IVDUs. We need programs that improve how the medical system address the issues of women & IVDUs).

Antiretroviral Therapy (ART) : women will enroll in studies if properly designed

The final 48 week results of the Women's First Study was presented in poster format (Abstract 330). Sixty-eight women were randomized to receive BID (twice daily) or TID (3 times daily) regimens of nelfinavir/ saquinavir/ d4T/ 3TC. In the as-treated analysis, >80% of patients on both arms achieved HIV RNA <400 copies/mL and CD4 cell count increases 170 cells/cc3. There were a number of gender-specific analyses performed that have been presented at previous conferences. Although the common perception is that it is difficult to enroll and maintain women in clinical trials, this study clearly shows that HIV-infected women will participate in studies involving complicated regimens if the studies are designed to address issues pertinent to this patient population. Dr Cu-Uvin assayed serial plasma and genital tract specimens from HIV-infected women receiving HAART-based regimens to determine the effect of antiretroviral therapy in both compartments (Abstract 718). She demonstrated that 1) suppression of viral load in each compartment was better in ART-naive than NRTI-experienced patients, 2) there was concordance between plasma and genital tract viral loads in patients with VL<LOQ and 3) HIV RNA was detected more frequently in the plasma than genital tract of patients failing ART. Finally, two groups reported the genotypic analysis of plasma and genital tract RNA obtained from HIV-infected women failing ART (Abstracts 446 and 719). The drug-resistance mutations in cervico-vaginal secretions differed from those found in the plasma in both studies. Further studies are needed to define the implications of these findings for both vertical and horizontal transmission.

Complications of Disease and Therapy

Boehringer-Ingelheim recently revised the management guidelines in the nevirapine package insert to indicate that liver function tests should be monitored closely, especially during the first eight weeks of therapy, to assess for drug-related hepatotoxicity. These guidelines were based, in part, on the reports of severe liver toxicity seen in FTC-302, a randomized study comparing FTC and 3TC with a background regimen of d4T and either nevirapine and efavirenz (Abstract 19). Ten percent of the patients receiving nevirapine experienced grade 4 elevations in liver enzymes; the incidence of grade 4 elevations in women was twice that of males. Two patients developed liver failure and died; both were women. Dr. Moore and colleagues determined the incidence of pancreatitis in HIV-infected patients attending the Johns Hopkins HIV Clinic who were receiving NRTI's (Abstract 620). Of note, female sex was an independent risk factor for pancreatitis in addition to the use of hydroxyurea and the "d" drugs-ddI and d4T (Moore did not analyze data to see if pregnancy was a risk factor).

A randomized, placebo-controlled study conducted by the ACTG demonstrated that nandrolone decanoate increased weight and lean body mass in HIV-infected women with weight loss. Although hoarseness, hirsutism (excessive facial & body hair) and clitoral enlargement did occur in the group receiving nandrolone decanoate, these expected side effects were rare.

Genital Human Papilloma Virus Infection and Cervical Intraepithelial Neoplasia; consider routine biopsy

It is not clear if HAART can change the natural history of cervical intraepithelial neoplasia (CIN) in the same way that it has impacted most of the HIV-associated opportunistic infections. Dr. Heard and colleagues reported that biopsy-proven CIN either reverted to normal or regressed to a lower grade at a significantly higher rate in a group of HIV-infected women followed in France receiving HAART-based regimens (Abstract 518). The number of women who had high-grade CIN at baseline was much higher than has been reported in earlier studies. There are ongoing studies that should help delineate the impact of HAART on this commonly encountered infection. Data from the WIHS cohort indicate that the correlation between either cervical cytology or colposcopic impression and biopsy is poor and biopsy of visible lesions should be routinely considered for the optimal management of HIV-infected women with abnormal cervical smears (Abstract 722).

Perinatal Transmission: c-section

Perinatal transmission rates remain low in HIV-infected women who receive antiretroviral therapy during pregnancy. In the United States, the proportion of deliveries by caesarian section increased from 19% in 1994 to 44% in 2000 with the greatest increase in 1999 (Abstract 702). A meta-analysis of 1, 202 mothers with viral loads <1000 copies/mL at or near delivery derived from 7 prospective studies of perinatal transmission indicated that the transmission rate is only 1% in women receiving ART and 10% in untreated women (Abstract 517). Single-dose nevirapine prophylaxis was associated with the selection of mutations associated with nevirapine resistance in women and children [HIVNET-012] or mothers alone [PACTG 316] (Abstracts 470, 516, 712). In HIVNET-012, the resistance mutations seen in the children were different than those seen in the mother, suggesting that resistance to nevirapine developed independently in viral isolates in the children. Although the use of short course or single dose regimens is not standard of care for HIV-infected pregnant women in the industrialized world, long-term follow up studies are needed to delineate the impact of resistance on the use of antiretroviral therapy for management of the woman's infection or in subsequent pregnancies (Editorial comments: if a pregnant woman is on a 2nd or 3rd line therapy how does underlying resistance affect resistance in child, viral load of mother & its effect on transmission risk).