Important note: Information in this article was accurate in September 2001. The state of the art may have changed since the publication date. IAVI Report - July / September 2001
Anne-christine d'Adesky
In many countries around the world, the rate of HIV infection in women is rising faster than in any another group. In sub-Saharan Africa, where over 70% of the world's HIV-positive people live, women made up about 55% of those living with the virus at the end of 1999, according to UNAIDS, and young women (ages 15-24) in the hardest-hit countries were up to three times more likely to be infected than males of the same age. In the US, women accounted for 23% of all new AIDS cases in 1999, compared to only 7% in 1986.
This vulnerability is deeply rooted in social and behavioral issues like poverty, gender power dynamics, and in many regions of the world, a decreasing age of sexual debut. But are there also biological factors at work? Do gender differences affect HIV transmission, as an intriguing new study suggests? If so, how? What about differences in immune responses? And what does this mean for vaccine development and other preventions?
For the past few years, such questions have sparked a growing debate among some HIV clinicians and researchers, as new data emerges to suggest that HIV can affect women and men differently. Several studies show that women's initial viral load levels are up to 50% lower than men's, though both progress to AIDS at the same rate—an important finding when it comes to antiretroviral therapy and clinical management of HIV disease. It also raises questions for vaccine developers, who increasingly view viral load as an indicator of efficacy for candidate vaccines that do not prevent infection but block progression to AIDS.
Turning to immunology, there are some known (but mostly ill-understood) gender differences which could influence responses to HIV. For starters, the female immune system has evolved to perform a delicate balancing act: it must protect the highly exposed genital tract from all manner of infections, but at the same time refrain from attacking immunologically foreign sperm; nor does it attack a fetus. It's also long been observed that women are at much greater risk than men for developing autoimmune diseases, although only recently has the field begun to tackle why this is so.
For HIV, several studies have linked local immune responses in the genital tract to protection in women—for example, in highly exposed, persistently seronegative sex workers (see article, "Closing in on Immune Protection in the Women of Pumwani"). It is also becoming clear that HIV in the genital compartment evolves independently from that in blood, with local immune dynamics influencing the process, and therefore that studies of HIV in blood alone may not give a complete picture.
But even acknowledging such differences, there has been no evidence that they can actually translate into different outcomes for any vaccine—that is, until last year's reports of data from two Phase III trials of an experimental vaccine against herpes simplex virus-2 (HSV-2) (see article, "Possibly, a Vaccine Against Herpes"). The startling results: signs of efficacy in women—the first apparent success for a candidate HSV-2 vaccine—but none at all in men.
New data on gender differences and viral load come from a recently published 5-1/2 year study of 317 heterosexual HIV-serodiscordant couples in Zambia (AIDS Res Hum Retroviruses 2001 Jul 1;17(10):901-10). In the study, a team from the University of Alabama at Birmingham (UAB) led by Susan Allen and Grace Aldrovandi used epidemiological linkage to confirm "true transmission pairs" and concluded that, as plasma viral loads increased, individuals were more likely to pass on HIV to their uninfected partners.
But surprisingly, the link appeared much stronger in women than in men, in whom the association between high viral load and transmission was deemed only "weakly predictive." For instance, women with viral loads above 100,000 were nearly six times more likely to pass on the virus than women with less than 10,000 copies, while for men the differential was less than two. The gender difference was especially pronounced at low viral loads—the most important range in terms of assessing how vaccines that reduce viral load will affect transmission rates. While only 9% of women with viral loads less than 10,000 passed on the virus, 24% of the men in this range transmitted (although the numbers of individuals were quite small).
"I think everyone suspected viral load to be a factor, and no one had looked at gender differences," states Sten Vermund, a UAB researcher who worked on the Zambian study. "Our finding was not at all subtle. It was very strong and consistent."
The Zambian study followed on the heels of a similar analysis in serodiscordant couples in Rakai, Uganda N Engl J Med 2000 Mar 30;342(13):921-9). There, a high level of HIV in the blood was also linked to an increasing risk of transmission. But the Rakai study reported no significant differences between men and women.
Why the discrepancy? It could well lie in the different study designs, says Thomas Quinn of John Hopkins University in Baltimore, who was lead author of the Rakai paper. Quinn was quick to praise the Zambian study as "fascinating and important." But he cautioned that his group didn't analyze viral load data above 100,000 copies, so a true comparison of the two studies is impossible. "We just lumped them all together above 50,000 copies," he said. The Rakai team also included fewer women in its study population, and they measured viral load in serum, not plasma, which is slightly lower. Besides these factors, says Quinn, "I could probably list ten variables, all of which play some modifying role in sexual transmission." He ticks off a few: condom usage, circumcision, viral load levels in blood versus genital secretions, age, presence of other STDs, HIV clade, duration of partnership, and the presence of acute seroconversion in the infected person. And of course, immune factors—the big gray area.
Clinicians are familiar with the variability."We know cases of women who have undetectable plasma viral loads who transmit, and there are certainly cases of women with very high plasma loads who do not transmit," says Kathleen Squires, an HIV clinician at the University of Southern California. Similar observations have been made for men. And transmission is clearly also affected by factors influencing HIV acquisition in the uninfected partner, as perhaps best illustrated by the cohorts of female sex workers who remain seronegative despite repeated, intensive exposure to many different HIV-infected male partners.
Recognizing that there is more to infectivity than viral load in the infected person's blood, several lines of research are trying to identify additional factors that affect both transmissibility and acquisition—information that could be key for understanding how vaccines which lower viral load may affect transmission overall.
An obvious place to look is in the genital compartments, where sexual transmission of HIV takes place. Susan Cu-Uvin is an Associate Professor of Obstetrics and Gynecology at Brown University who is studying HIV and immunity in the female genital tract. Says Cu-Uvin: "The majority view is that the higher the plasma viral load, the higher your genital tract viral load. But there are outliers. It's not everybody. Depending on the method you use to measure, and on the subcompartment of the genital tract, you will find that—at least in my work-up to 50% of the time, the genital tract viral load is higher than in plasma. So I think we can say they are separate compartments." Others have reported different figures of discordance in these reservoirs, but come to the same conclusion. Adds Cu-Uvin: "If you only study the plasma, you make assumptions that might not reflect what is going on and where the [transmission] action is taking place. Because you don't have sex in your plasma—HIV is a mucosal infection."
But even viral load in the directly-transmitting compartment is not the whole story, says USC's Squires. "Even if you look at maternal-fetal transmission studies, it's clear that there is not a one-to-one correlation" between high viral load and transmission, she says. To understand infectivity, "we need to look at genital tract viral load and do a head-to -head comparison between men and women. Given the right circumstances, both can be infected with HIV; it's not a gender-selective infectious disease. But what happens once they are infected may be different. Viral load in the genital tract is not the end of the story, either."
Backing these statements are a flurry of data showing that distinct viruses evolve in plasma versus the rectum, vagina, semen or breast milk. For example, Ted Ellerbrock of the Centers for Disease Control (Atlanta) recently found that plasma and vaginal secretions of HIV-positive women contained different viral populations, based both on genotype and on the emergence of drug-resistant subpopulations (J Infect Dis 2001 Jul 1;184(1):28-36). In his study, only 2% of the HIV in vaginal secretions were common to blood. In another example, Carmen Zorrilla (University of Puerto Rico) reported different patterns of HIV resistance in paired plasma versus vaginal swabs taken from HIV-positive women on HAART therapy (Abstract No. 719, 8th Conference on Retroviruses and Opportunistic Infections).
Immune components in the genital tract are also likely to be important modulators of infectivity, although the evidence so far is mostly suggestive rather than definitive. The data come partly from studies of SIV in primates and HIV in mice (see IAVI Report, May-June 2001, page 1). In these models, HIV-specific CD8+ T-cells (cytotoxic T-lymphocytes, or CTLs) have emerged as key players in mucosal defenses to HIV, and CD4+ T-helper cells also have a role. Other clues come from cohorts of commercial sex workers who do not seroconvert despite repeated exposure to the virus. High HIV-specific CTL levels in the genital tract may be contributing to this viral resistance, alongside humoral responses like secretory IgA (sIgA) (see article, "Closing in on Immune Protection in the Women of Pumwani").
At the Fearing Laboratory of Brigham and Women's' Hospital in Boston, Deborah Anderson is studying HIV in the male genital tract and blood. Working closely with Cu-Uvin, she has found that the most important mediators of T-cell responses are present at the urethra near the opening of the penis. "The mucosal underside of foreskin is a very important area to look at," she says. Some key questions: how mucosal responses compare with those in the vaginal mucosa and whether they can be enhanced with a vaccine.
That raises the practical question of how to do these studies. The big obstacles are a lack of standardized diagnostic tools to sample viral load in the genital tract, and quality control of samples. "I can do a viral load on a finger prick with just a filter paper and a drop of blood," says Quinn. "It's well standardized. The way genital tract samples are collected is much more difficult."
Part of the problem is variability. "The female tract is so dynamic that it's different from today to tomorrow," states Cu-Uvin, noting that hormones, menstrual cycle and STDs all affect the delicate balance of this microenvironment. The other part is the collection method. "If you use (cervical) lavage, you're diluting your sample," she says. "With the snow strip, the volume is so small you can't do additional studies. With the cytobrush, 97% of samples are contaminated with blood. We just haven't found the best way of doing it [in women]. In men, you masturbate, that's it." Another issue in women is ensuring that the collected virus reflects the woman's own infection rather than that from the semen of an HIV-positive sex partner.
In the meantime, some insights can be gained from in vitro studies. For instance, Anderson's laboratory and others have succeeded in culturing pieces of vaginal mucosa to study antibody transport at the molecular level. Other groups have developed laboratory models of sexual transmission to track how and where HIV breaches the mucosa of the human genital tract and then infects underlying cells, and what immune activation takes place. As Anderson notes, "We need in vitro systems to complement these in vivo observations, which are so difficult to interpret."
As they look to explain gender effects in HIV, researchers usually find themselves turning to hormones. "Why do women have lower plasma viral loads?" asks UAB's Vermund rhetorically. "Nobody has a clue. But women and men's makeup are nearly identical, and the key differences are hormones [and anatomy]. Are there unknown hormonal influences that either up-regulate HIV in men or down-regulate it in women? That could have therapeutic and other implications." As more questions pile up, HIV researchers are hampered by an overall lack of knowledge about basic reproductive biology and mucosal immunology. "There is not a whole lot of data available on what happens in the 'normal' vaginal tract," says Squires. "You can look at infections in pregnancy and endometriosis and those kinds of models as a corollary to what happens in the genital tract when a woman has an infection. But a big problem is that the hormonal milieu [in these circumstances] is so different."Compared to the peripheral immune system in the blood, the human mucosal system consists of a complex, integrated network of tissues, lymphoid and mucous-membrane cells, plus effector molecules such as IgA antibodies, cytokines, chemokines and their receptors—all of which act in concert with innate host factors such as defensins. Both cell-mediated and antibody responses occur in the mucosa, where adhesion molecules allow immune cells to "home" to sites of infection. Studies of sexual transmission show that HIV targets the cervix and that compared to men, the cervix offers a larger target cell area for the virus than the head of a penis. But key features that distinguish the male and female immune system are sex hormones and regulatory cytokines that control the level of antibodies like secretory IgA and IgG in vaginal fluid and tissues.
As Squires points out, pregnancy is a time when the immune system in a woman must somehow allow immunologically distinct fetal cells to survive. Sex hormones like estrogen and progesterone are thought to direct that response. They're also the fingered culprit in studies on why women are so prone to autoimmune disorders. In his textbook review of genital tract immunity, William Kutteh notes, "The composition of genital secretions (including IgG and IgA levels) is dependent on hormonal and inflammatory factors" (Mucosal Immunology, 1999). He and others report that the endocervix and ectocervix have a very high accumulation of immunoglobulin—forming cells that produce predominately sIgA antibodies. Other HIV and SIV transmission studies in female mice and primates show that these viruses initially target the cervix, including the lamina propria (just under the epithelium), before spreading to other sites.
Hormones can also affect immune dynamics at the epithelial surface of the cervix. At the Tulane University Primate Center, Preston Marx has reported that progesterone thins the epithelial cell layer of the cervix in primates, allowing SIV to more easily infect target cells. This monkey model may shed light on what is seen in young girls, where the lining of the cervix is known to be thinner than in older women and thickens with the onset of puberty. Such biological factors may play a role in making teenage girls more susceptible than older women to HIV infection. They might also help explain observations that women on birth control pills have a somewhat elevated risk of HIV acquisition, although a biological (as opposed to behavioral) basis for this has not yet been conclusively shown.
"The progesterone link is so interesting," says Squires. "I think it's short-sighted of us not to try to figure out what happening because we might be able to utilize this."
Taken together, what do these findings mean for vaccine research? With so many questions still unanswered, the clearest implication is that gender should be firmly on the agenda as clinical trials begin to evaluate how HIV vaccines that lower viral load affect disease progression and transmission. Will these vaccines have different clinical benefits in men and women, given the gender differences surrounding viral load? And will lower viral load levels affect infectiousness in men and women differently?
With most of the current HIV vaccine candidates unable to block infection but apparently working by lowering virus levels, these become critical questions. The more answers we have, and the more information on factors contributing to infectivity—including viral load, STDs, male circumcision status and mucosal responses—the easier it may be to answer these questions and guide future vaccine design and testing.
010710
IAVI2001-0711
©2001. The IAVI Report.
AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., John M. Lloyd Foundation, the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2001. This material is designed to support, not replace, the relationship that exists between you and your doctor.
AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.
Copyright ©1990, 2001. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.
