icon-folder.gif   Conference Reports for NATAP  
  9th Conference on Retroviruses and Opportunistic Infections
Seattle, Washington, February, 2002
Back grey_arrow_rt.gif
HIV and Women at Retrovirus
Written for NATAP by Danielle Milano, MD, Boriken Family Health Services, NYC
  At the Retrovirus Conference in Seattle there was a combined total of 874 posters, lectures and oral presentations. Of those, 46 pertained to women (30), or perinatal transmission (16). Retrovirus is a basic science as well as clinical research conference, therefore many of the presentations have no practical applications yet. Perinatal transmission is covered in a separate report posted to the NATAP website.
A number of abstracts dealt with cervico-vaginal shedding and detection of HIV in cervical lavage. Quantifying the amount of HIV in cervical secretions is difficult. One abstract (782-W) compared the traditional method of cervico-vaginal lavage (CVL) with "sno-strips". Traditional CVL washes the cervical area with 10 ml of normal saline, and the fluid is collected. Sno-strips are small wicks that absorb a small amount of cervical secretions (8 mL). The sno-strip method was found to be more sensitive than CVL.
A small study out of Brown and Emory Universities looked at 97 women over a 2-year period to assess factors associated with HIV-1 shedding in the genital tract (Abs 784-W). They found that for each one log10 increase in plasma viral load, it increases the chances of finding HIV in the female genital tract by 4-fold. Neither the CD4 count nor the use of HAART had an impact. The researchers split the data into 3 groups: a) ARV naive starting HAART, b) ARV experienced starting HAART, c) not taking HIV medications. Even at baseline, the majority of women did not have detectible HIV in their genital tract by CVL (67%, 71%, 79%). The take-home message is that a detectible HIV plasma viral load is associated with finding HIV in the female genital tract. The CD4 did not have any effect on genital HIV shedding.
A study out of Senegal (Abs-785-W) showed slightly different results in that a lower CD-4 was associated with the presence of HIV in genital secretions, as was recent menses or amenorrhea.
Overriding the data from the two aforementioned studies was a "late-breaker" entitled "Relationships between the menstrual cycle and the daily pattern of HIV-1 RNA shedding in the Genital Tract of HIV-1 infected women" (LB-2). These researchers studied 17 women intensively over the period of 1 month. 89% of the samples showed the presence of HIV (unlike the Brown & Emory study in which only 1/4 to 1/3 of the samples has detectable HIV), and the amount of virus was quantified. They found that HIV-1 genital shedding is lowest just prior to the "LH Surge" (rapid increase in levels of luteinizing hormone just prior to ovulation) and increases with each day post surge. Although this was not an unexpected finding, prior to this study the data had been contradictory. It is known that pregnancy and hormonal contraception affect cervical shedding of HIV so it seemed logical that the hormone fluctuations associated with the menstrual cycle would do so also.
There was an abstract on the presence of cytomegalovirus (CMV) in vaginal secretions. In men, it is known that CMV can be found in the semen when HIV-VL is high in plasma. This was the first study that examined CMV in female genital secretions. This study (ABS 786-W) looked at 45 women. 7/45 had detectable CMV in genital secretions and there was an association with higher HIV viral loads. 2 of the 7 women were harboring 2 strains of CMV. The concern is that in men, infection with more than 1 strain of CMV is associated with an increased progression to AIDS. Presumably this is true in women also.
The prevalence of other viral infections in HIV infected women was examined in 871 HIV+ and 439 HIV- high-risk women enrolled in the HER study (HIV Epidemiology Research Study) (Abs 778). CMV, Hepatitis B, HCV, HHV-8 (KS virus), Herpes-2 (genital herpes) and HPV (Papilloma virus) were more prevalent in the HIV+ women. HSV-1 (oral herpes) had a similar prevalence in the 2 groups. The difference was most striking for HPV (63.8% in HIV+, 27.4% in HIV-).
Studies in HPV were not well represented at Retrovirus. One poster announced the discovery of 3 new HPV subtypes (Abs 607-W). Another looked at HPV viral load in the genital tract in women, some of whom started HAART (Oral Abs 119). Initiation of HAART had no impact on HPV genital viral load, even with an HIV-VL BLD! This study only went to 6 months post-HAART.
It is known that:
  • HIV- women are more likely to clear HPV infection than HIV+ women.

  • HIV+ women are more likely to have high-risk HPV subtypes (i.e., the types that progress to CIN).

  • HIV+ women, as CD4 counts drop, are more likely to change from low-risk to high-risk subtypes.

Therefore, the hope was that HAART and subsequent immune reconstitution would clear the HPV infection. This study, at 6 months in length, may be too short to see an effect. Separately, an abstract was presented on anal HPV in men which showed pretty much the same result. (Abs 606-W). This was a cross-sectional study in which the patients had been on ARV for close to 5 years. Immune restoration had no impact on the presence of HPV in the anal area, and no impact on the prevalence of anal SIL. This is bad news.
A large study out of UC-San Diego (Abs 605-W) showed that men, whether or not they are a MSM, are at higher risk for SIL than women. Interestingly, this study found that patients with higher CD-4 cell counts were less likely to have SIL. This does not jive with the abstract summarized above. It is difficult to compare the 2 studies head-to-head. The patient populations were different. My interpretation is as follows. As CD4 drops, HPV switch to high-risk types, the immune system cannot fight the infection and carcinoma-in-situ develops. Just because CD4 recovers does not mean that the carcinoma goes away. So the first study looked at a group of patients that had low CD4's in the past but were 5 years out on HAART, the CD4's rebounded but the CIN did not reverse. The second study had a group of participants that were earlier in the HIV disease spectrum, and their CD4's did not drop yet.
As an aside, a concern of mine had been whether we were missing ano-rectal SIL in our HIV+ women. We have been performing rectal pap smears on our male patients who either are a MSM or have evidence of rectal condyloma. We have not yet begun to do so in women. In this study out of San Diego, the prevalence of ano-rectal SIL in women was low (5%), compared to 18% in MSM and 29% in non-MSM. Thus, all men warrant a rectal pap smear. Considering a prevalence of 5% in women, rectal pap smears should also be recommended.
HCV in genital secretions of women was examined in a small study (Abs 648-M). It is known that HCV can be detected in semen although at low levels. "HCV-RNA was detected in plasma and CVL of 5/9 women." (It was not clear if the 4 women without HCV in CVL actually had HCV viremia or if they were HCV resolves). Nevertheless, the researchers found sequence diversity (genetic diversity) between the HCV in the CVL and, in the plasma in 4/5 of the women with HCV in the genital secretions. This suggests an "HCV compartment" or the concept of the female genital tract as a separate immunologic compartment. The concern is that separate compartments act as reservoirs for HIV, can archive virus with resistance mutations, and can effectively isolate the virus so that it evolves differently. This leads to greater genetic diversity in the virus. Genetic diversity is associated with a faster disease progression.
This brings us to the abstract entitled "Genetic diversity and resistance mutations of HIV-1 in blood and genital secretions of women (Abs 787-W). These researchers from U-SC looked at genetic variation (sequence diversity) in the HIV-1 pol gene in plasma and CVL in 30 women. Diversity between HIV in the 2 compartments was 2.3%, which is not different than intra-compartment diversity (i.e. plasma-plasma; or CVL-CVL). Drug resistance mutations were found in 25/30. In 19, the mutations were identical in the plasma and CVL. Five showed mutations only in plasma and 1 showed a mutation in CVL only. These are encouraging findings.
The effect of gender on HIV-VL and disease progression was not well represented at the conference. One abstract reviewed 13 longitudinal and 9 cross-sectional studies (Abs 775-W). At similar stages of disease, women have 0.13 to 0.345 log10 (cross-sectional) or 0.33 to 0.78 log10 lower HIV-RNA. Much more interesting was a study out of Cote d'Ivoire. 104 seroconverters were followed over approximately 30 months: 77 men and 21 women. At entry, CD4's were similar (541 in men, 493 in women). Median initial HIV-VL was higher in men (39,810 vs. 6,310 copies/mL). However, if you looked at the patients who progressed to stage B or C disease, i.e. symptomatic HIV or AIDS, the HIV viral loads were not different between men and women (50,119 vs. 63,096 copies/mL). Thus, my interpretation is that a high viral load in a woman early in the course of HIV infection is a marker for someone who will progress quickly.
Effect of gender on the virologic and CD4 response to HAART was examined in a study out of Columbia University. 229 patients, 35% women, were included in this retrospective chart review. Meantime to start of ARV was longer in women (355 days vs. 184 days). CD4 counts were equivalent at the start of ARV (152 vs. 127 cells/mm3 which was not a statistically significant difference). 92.5% of the women and 80.5% of the men achieved an HIV-VL BLD. Thus, women were more likely than men to achieve an HIV-VL BLD (Odds ratio = 3.0). The proportion of women maintaining undetectable on HAART was 79% at 1 year and 70% at 3 years were similar for men. The CD4 increase was about the same for men and women per year (about 120). The reported HIV risk factors for women were heterosexual contact (84%) and IVDU (13%). The average time from the first clinic visit until starting ART was longer for women than men, 355 days vs. 184 for men. IVDU was a risk factor for not achieving undetectable viral load (odds ratio= 3.8) (abstract 777W).
Lastly, on a practical note, a study presented on the poster session by JH Arsten, MD of Montefiore/Albert Einstein looked at bone mineral density in older HIV+ women on Protease Inhibitors (PI's) (Abs 717-T). Decreased bone density is a normal sequelae of aging and may also be a side effect of PI's. Studies have also found associations with NRTIs. It is uncertain what may be causing bone loss in HIV. This study is small. The researchers looked at 40 older women (>40 years old or >35 years old if menses were irregular). Bone densitometry scans of the lumbar spine, hip, and total body were performed in 40 peri- and post- menopausal women using a Hologic DEXA scanner. Mean body weight was high (72.5 kg = 159 lb) and the women were mostly African-American or Latina. 19 women were HIV-infected and 21 women were HIV-uninfected. All HIV-infected had been on ARV's at some point and half had been on PI's. The women who had been on PI's were much more likely to be osteopenic (56% [n=11] vs. 13% [n=8]); p=0.05) and had lower t-scores, controlling for age and body weight. HIV-infection was not found to be associated with bone loss. This study found an association but did not break out the results by potential cofactors such as diet, lifestyle, weight, smokers, exposure to drugs that may be associated with bone loss, nor other potential cofactors. One of the problems with osteopenia/osteoporosis in HIV is that we don't know the mechanism of the bone loss. This knowledge might help in prevention or therapy. Other studies have not found any association between protease inhibitors and osteopenia. HIV itself or immune restoration from HAART may be associated with bone loss, as well as many risk factors including body mass. Study has found an association between having less fat or body mass and proclivity for bone loss. As in lipodystrophy it is difficult to design a study that will yield conclusive cause and effect. It is prudent to suggest Calcium supplements and consider early bone density testing.