Table of Contents

• Introduction
• Natural History of HIV Infection and Response to Treatment in Women
• Disease Progression in Women Versus Men
• Gender-Based Differences in Treatment Response
• Does Pre-Nutritional Status Predict Response to Antiretroviral Therapy?
• Causes of Death in HIV-Infected Women
• Treatment-Related Adverse Events
• HIV Shedding in the Female Genital Tract
• Early Menopause
• Prevention of HIV Transmission to Women
• Vaginal Health and Viral Load
• Male Circumcision as a Method of HIV Intervention
• Herpes Suppression as a Method of HIV Intervention
• Maraviroc Levels in the Genital Tract
• Hormonal Contraception and HIV Transmission
• Hormonal Contraception and HIV Disease Progression
• Factors That Influence HIV Transmission
• Conclusion
• Footnotes

Introduction

I'm going to review new findings from CROI 2008 specific to HIV-infected women. First I'll talk about HIV disease progression and response to therapy. Then I'll talk about some very specific issues that are always of concern to HIV-infected women, specifically, reproductive cycling. Finally I will review the growing science regarding the prevention of HIV transmission to women.

Natural History of HIV Infection and Response to Treatment in Women

Disease Progression in Women Versus Men

Click to enlarge

I'll begin with a study that assesses gender-based differences in viral load, specifically at different CD4+ cell count levels.¹ This study presented by Beatriz Grinsztejn follows up on a previous fund of knowledge that, at the time it was first elucidated, was so surprising that many people thought it was not true, and that the researchers must be wrong. And that is that viral loads in women are clearly lower than viral loads in men.

This study that was presented at CROI 2008 extends our previous fund of knowledge in that it includes women from many parts of the world and it includes women at lower CD4+ cell counts in a single study, which has never been done before. The people enrolled in this randomized clinical trial were from eight resource-limited countries and the United States. There were 831 men and 740 women. Everybody who was enrolled had a CD4+ cell count of less than 300 cells/mm³.

The study researchers found that there was an almost perfect linear relationship between a lower CD4+ cell count and a smaller difference between women and men in terms of their viral loads -- that is, their HIV-1 RNA levels. Among study participants who had a CD4+ cell count of up to 300 cells/mm³, there was approximately a 0.2-log difference between women and men in their viral loads. Among participants with a CD4+ cell count below 50 cells/mm³, the difference was less than 0.1 logs.

This is in keeping with all previous data, which has indicated that there's a greater difference between women and men at higher CD4+ cell counts, meaning women have much lower HIV-1 RNA levels at the higher CD4+ cell counts, but we have never before been able to see this sort of direct linear relationship.

Gender-Based Differences in Treatment Response

Click to enlarge For additional images from this study, click here.

When people are treated for HIV infection, is there a gender-based difference in the response? One group from Nashville, Tenn., presented a study of 617 women and almost 2,000 men that addressed this.² The researchers were looking to see whether there were any differences by race and gender in response to treatment. The outcome that was measured was the ability to reach a viral load below 400 copies/mL.

As is almost always the case, there were differences between the women and men at study entry. The women were more likely to be black, were younger and had higher CD4+ cell counts at entry. In general, that's always true, both because women access care earlier and also because women without HIV infection generally have a higher CD4+ cell count than men. The women were also less likely to have had an AIDS diagnosis and were less likely to have had HAART [highly active antiretroviral therapy] during the course of their care.

The bottom line with regards to the study findings were that, if you did an unadjusted analysis, both blacks and women were less likely to have a good virologic response. However, when you adjusted for a variety of factors, especially with regards to whether the patient received HAART, the racial difference in terms of virologic response disappeared, but the gender-based difference in virologic response became more pronounced. After adjusted analyses, women in the study were found to be 50% more likely to die compared to the men in the study, whereas people of color had a similar risk of death compared to the non-blacks.

Click to enlarge For additional images from this study, click here.

However, I want to contrast those results with the results of a study by Amy Weintrob et al of a military cohort in the United States.³ The importance of this military cohort is that many of the sociodemographic differences you find between women and men, or between people of color and whites, in the United States, are almost obviated in the military. Everyone has a similar income. Everyone, at least until recently, had graduated from high school. Everyone has access to care. The providers of this care are the same for everyone.

What was found in this study from the military was that there was no difference between women and men in terms of achieving a good virologic response on HAART. But there was a difference by race. Specifically, the people of color in this study were only half as likely to achieve an undetectable viral load. Half as likely -- that is really a huge difference. It was highly statistically significant, with a P value of less than .001.

As I alluded to earlier, we find that we have contradictory results. Do women respond better, the same or less well to treatment than men? In fact, you could ask the same question about differences by race. The short answer to both questions is that, currently, we don't have enough information to know.

Does Pre-Nutritional Status Predict Response to Antiretroviral Therapy?

Click to enlarge For additional images from this study, click here.

There were other studies from CROI 2008 that added some information to our knowledge, sometimes fairly important information. The first is a study of women in Rwanda that looks at nutritional status prior to treatment and whether that predicted a good or poor response to antiretroviral therapy.⁴ This study, which I was involved in, was restricted to HIV-infected women who had initiated antiretroviral therapy. The 457 women in the study were seen every six months, at which time multiple measures were taken, including some laboratory tests and measures of body composition.

The authors found that none of the measures of nutritional status were predictive of either the CD4+ cell count response or the virologic response to antiretroviral therapy. Specifically, serum albumin, hemoglobin, total body fat and fat-free mass did not predict the response to therapy. This is an encouraging finding, because in many parts of the world poor nutritional status is common among HIV-infected people initiating therapy. This study suggests that mild, pre-treatment malnutrition does not impair the response to HAART.

Causes of Death in HIV-Infected Women

Click to enlarge For additional images from this study, click here.

A study from France looked at the causes of death in HIV-infected women, and how they've changed since 2000.⁵ It actually looked at the causes of death in HIV-infected men, also. It's a fairly large study. There were 1,042 people included in 2005, compared to 964 people included in the year 2000.

Approximately 22% to 24% of these participants were women. The information was gained, first from public health death data, followed by chart abstraction in France. The remarkable finding in this study is that in 2000, women and men were equally likely to die from AIDS, but by 2005, women were much more likely to die from AIDS and much less likely to die from other causes.

In uncontrolled comparisons that we have from other studies -- for example, the U.S. Women's Interagency HIV Study [WIHS] -- we see that women remain very likely to die from AIDS, right through to the present.⁶ In WIHS, of those women who are dying, more than 50% of them are dying from AIDS. This is in contrast to other studies of men in general, which tend to show that AIDS is no longer the leading cause of death.

What accounts for that? Certainly, neither of these studies tells us the answer for why that is happening, but we have to consider that there's truly a biologic difference based on gender, and not just a difference in access to care.

Treatment-Related Adverse Events

Click to enlarge For additional images from this study, click here.

There's been some indication in the past, especially from WIHS, that the NRTIs [nucleoside reverse transcriptase inhibitors] may act somewhat differently in women than they do in men.⁷ Specifically, there seems to be more NRTI-specific toxicity in women than is generally described in studies of men, although there are very few studies that have direct comparisons.⁸ WIHS provided information at CROI 2008 regarding the development of insulin resistance among women.⁹

WIHS, in general, has enrolled a little over 3,700 women, of whom almost 1,000 were HIV uninfected, and therefore provide a good comparison group over time. This particular study included 2,218 women who contributed a total of 11,000 visits, in which they had had fasting glucose and insulin measured.

The study found, using the HOMA [homeostasis model assessment] index as the measure of insulin resistance, that women reporting the use of any kind of HAART since their prior visit were more likely to have insulin resistance. This was true whether or not the HAART included a PI [protease inhibitor]. In addition, the usual things that affect insulin resistance were found to be predictive -- those are hepatitis C infection, family history of diabetes, increasing age, increasing body mass index and having gone through menopause.

Taking a closer look at the components of HAART, it was found that it's really the NRTIs that were predicting the insulin resistance. The PI use and the NNRTI [non-nucleoside reverse transcriptase inhibitor] use were non-associated with insulin resistance.

Then, in further analysis, the researchers looked at which components of the NRTIs were contributing, and it was found to be almost entirely stavudine [d4T, Zerit].

The authors concluded that, although any use of HAART was associated with a higher likelihood of insulin resistance, the cumulative exposure to NRTIs, especially stavudine, but not PIs or NNRTIs, was independently associated with higher insulin resistance. This suggests that we need further study to understand this better.

Click to enlarge

Another study, this one by Sharon Riddler et al, comparing women and men in ACTG trial 5142, also had findings suggesting there may be some NRTI effects in women that may be different from those in men.¹⁰

The primary purpose of ACTG 5142, a randomized clinical trial of 753 people, of whom 20% (151) were women, was to compare three kinds of antiretroviral treatment in terms of efficacy, which has previously been reported.¹¹ The researchers compared efavirenz [EFV, Sustiva, Stocrin] plus two NRTIs to lopinavir/ritonavir [LPV/r, Kaletra] plus two NRTIs to a third regimen, which had no NRTIs, and was just efavirenz and lopinavir/ritonavir.

With regards to differences in gender, it was found that virologic failure was associated with being female. Virologic failure was also associated with younger age, a lower CD4+ cell count and black race.

The gender-based differences were predominantly by regimen. Thus, the risk of virologic failure was lower for women than men in the NRTI-sparing arm. There was no difference seen between women and men in terms of the risk of virologic failure in the other arms. Similarly, the risk of toxicity was lower for women than men in the NRTI-sparing arm. Therefore the women had a longer time to virologic failure and a longer time to treatment-limiting toxicity with the NRTI-sparing arm.

This suggests, similar to the previous study from WIHS,⁷ that the NRTIs may have different effects in women than men. The frequently shown differences in toxicity that we see with HAART between women and men may, in fact, mostly be driven by the NRTIs. Further study is needed for us to be able to understand this fully.

HIV Shedding in the Female Genital Tract

Let me mention two studies that are not easily categorized, but which generally are of great interest to HIV-infected women. The first is a study that looks at some of the predictors of HIV shedding in the female genital tract.¹² Meaning, what is it that creates higher viral loads in a female patient's genital tract than in her blood plasma? The study specifically was looking at whether immune activation systemically -- not locally in the genital tract, but systemically throughout the body -- was associated with higher viral loads in the genital tract.

Click to enlarge For additional slides from this study, click here.

All past studies have indicated that the single strongest predictor of HIV viral load in the genital tract is the systemic HIV viral load (the viral load in the plasma). The question of what does systemic immune activation do had not really been previously studied. Systemic immune activation, in this study, was measured by looking at specific kinds of CD8+ cells, and whether they were DR-positive or DR4-positive.

This was a study of 225 HIV-infected WIHS women who had paired plasma and cervical vaginal lavage (CVL) specimens available. The researchers found that systemic immune activation is significantly associated with viral load in the genital tract, and that the lack of systemic immune activation, which is measured in a different way, was protective.

We had, in the past, thought that it was mostly the local -- meaning, just compartmentalized in the genital tract -- immune activation that could be driving the genital viral load. This study suggests that, in fact, the pathophysiologic mechanism might actually be systemic immune activation, and therefore this needs to be studied further.

Early Menopause

Another question that HIV-infected women frequently have, and in fact, which the women in WIHS have repeatedly asked for data regarding, is whether there is early menopause, also referred to as early ovarian deficiency, in HIV-infected women compared to uninfected women. One would think that this is easy to study, but it's not, in part, because the ability to study menopause is an evolving field. Only really in the last 10 years have we had research methodology that allows us to start addressing this question.

At CROI, there was one French study by Marialuisa Partisani et al looking at premature ovarian deficiency in HIV-infected women, which included 78 HIV-infected women.¹³ The researchers looked at four markers of ovarian function: follicle-stimulating hormone [FSH]; inhibin B; antimullerian hormone [AMH], also called mullerian-inhibiting substance or MIS; and antral follicular count [AFC]. They did ultrasounds of the ovary to look at the antral follicular count, and then did a descriptive analysis according to age.

The researchers found that in 85% of the women, at least one marker was abnormal. There was no effect of the women's treatment status on these ovarian markers.

Now this study does not have a comparative group of HIV-uninfected women. However, any historical comparator of women in this age group -- and the mean age was 34 years -- would not find that 85% of women had at least one abnormal marker. Thus, this essentially should be looked at as preliminary data, suggesting that better controlled studies should be performed using these sort of new and multiple markers of ovarian function.

Of note, there have been other studies, not of ovarian function in HIV-infected women, but of clinically diagnosed menopause.¹⁴ Clinically diagnosed menopause almost always means that the female patient tells the provider whether she's gone through menopause. Those studies have not found a difference, actually, between the HIV-infected and the HIV-uninfected women. This study, however, has more sensitive measures, and therefore should be expanded.

Prevention of HIV Transmission to Women

Now, let me talk about prevention of HIV transmission to women. I'll start with a somewhat encouraging observational study, but then I'm going to move into the clinical trials. In this area, as in the vaccine trials, the results have been very disappointing in terms of the ability to actually prevent HIV transmission to people.

Vaginal Health and Viral Load

I'll start with a look at a normal, healthy vagina, which is really defined in great part by the presence of Lactobacillus, or plural lactobacilli. Lactobacilli are hydrogen peroxide-producing bacteria present in women's vaginas in greater or lesser degree and which, for many decades, have been known to promote genital health in women, or vaginal health, in particular. For example, back in the '60s or '70s, we learned that the presence of lactobacilli would prevent yeast infections in women, or make them less likely.

Click to enlarge

This study by Jane Hitti et al that was presented at CROI 2008 looked at 57 women, who were seen multiple times. The researchers assessed the presence of lactobacilli, as well as the presence of the bacteria that cause bacterial vaginosis.¹⁵ They also looked at the HIV-1 RNA levels in the vagina.

What they found was that the presence of lactobacilli was associated with a 0.5-log lower viral load in the vagina, whereas the presence of bacterial vaginosis was associated with a 0.3-log higher viral load in the vagina. That is based on cross-sectional study.

Longitudinally, the researchers followed women over time and found that, if a woman acquired more lactobacilli (had a higher level of lactobacilli) over time, that was associated with a decreasing vaginal viral load. In fact, it decreased by 0.7-log, which is actually a huge decrease, especially in the vagina, which tends to have low viral loads to begin with.

In addition, if the woman had had lactobacilli, but then lost them over time, there was a 0.5-log increase in the vaginal viral load. There was no effect of trichomonas or yeast infections, or their change over time, on the vaginal viral load. The authors' bottom line was that vaginal health decreases the HIV viral load.

Let me just say a word about why we would be studying this in the first place. The reason to study it is that these local factors in the vagina have been shown to actually be associated with HIV acquisition or HIV transmission. Thus, the presence of bacterial vaginosis, because it's a very common infection, could have a very large population-based effect should it be found to increase HIV transmissibility. That is why this has been studied.

Could we use this information to promote changes in women's vaginal health? Specifically, what you'd want to do is get rid of the bacterial vaginosis bacteria and increase the lactobacilli. It's not entirely clear how that could be done. In theory, you can treat bacterial vaginosis, which is what we do after we diagnose it, but how do you get more lactobacilli into the vagina?

In the distant past, some 30 years ago, there was much discussion among women about whether putting live-culture yogurt into the vagina would increase vaginal health. That has not actually been studied, but there was discussion of it at the meetings.

Male Circumcision as a Method of HIV Intervention

Now, for somewhat discouraging news, let me talk about the intervention trials that have been done to assess whether certain kinds of interventions can prevent HIV transmission to women and men, though I'll specifically be talking about the outcomes in women here. Two different people presented at CROI 2008 from the Rakai study, which is a trial of male circumcision.

Click to enlarge

Aaron Tobian et al looked at whether circumcision in men would have any favorable outcomes in the mens' female partners.¹⁶ In this study, there were 3,516 men enrolled who were negative for both HIV and HSV-2 (herpes simplex virus type-2, also known as genital herpes infection). Of those, approximately half had immediate circumcision. The rest of the men had delayed circumcision, meaning 24 months later they would become circumcised. There was 24 months of follow-up, with 80% retention.

One of the outcomes that was studied in men was whether circumcision would prevent the acquisition of genital herpes infection, which it did. There was a 25% decrease in genital herpes infections in the men. The effect was even greater, a 50% decrease, in men who were high risk, which basically meant that they did not use condoms consistently.

Click to enlarge

The female partners who were also studied were defined as being the wives of some of these 3,516 men. There were 1,608 wives, of whom 825 were married to men in the immediately circumcised group, and 783 were married to men in the delayed circumcised group. There were 12 months of follow-up, with 95% retention. The outcomes that were looked at in these women were vaginal symptoms and vaginal infections.

Click to enlarge

There was no difference in vaginal symptoms based on whether the male partner had been circumcised. There was a difference in diagnosed genital ulcer disease among the female partners of men who had been circumcised, and an approximately 50% decrease in the incidence of trichomonas. There was not much of a difference in bacterial vaginosis in general, but in severe bacterial vaginosis, there was a 70% decrease. To sum up, male circumcision was associated with decreases in genital ulcer disease, severe bacterial vaginosis and trichomonas in the female partner.

A second presentation by Maria Wawer et al of this same trial on circumcision in men looked at the HIV-infected men who had been enrolled.¹⁷ There were 992 HIV-infected men enrolled. The 94 female partners of those men who were circumcised and the 71 female partners of those men who did not get circumcised were also included in the study.

Click to enlarge

The results: There was no difference at all in genital infections among the women. There was also no difference at all in HIV transmission to the women from their male partners, with one exception. That exception occurred among couples who had sex early -- meaning, prior to wound healing, when there was an enhanced risk of transmission to the women. The transmission rate among couples who had early sex was 27%, versus 9.5% among the couples who did not have early sex. This was essentially the same as the transmission rate seen in the uncircumcised men.

As we know, circumcision protects men from HIV acquisition,¹⁸ but it does not seem to be the case that circumcising HIV-infected men confers any protection to their female partner. In fact, if men are recently circumcised and there is sex prior to wound healing, that actually enhances the risk of transmission to the female partner.

Herpes Suppression as a Method of HIV Intervention

Another intervention that has been studied in clinical trials to prevent HIV acquisition, or HIV transmission, is to suppress herpes simplex genital episodes in people who are already herpes simplex infected. It is thought that this would work, because people who have herpes simplex are known to be more likely to have HIV infection, or maybe even be more likely to acquire HIV infection. The thinking was that, if we could prevent the ulcerations from herpes simplex, then we could prevent the HIV transmission.

Click to enlarge

In this study by Connie Celum et al, 3,277 participants were randomized to either have their herpes simplex suppressed or not suppressed.¹⁹ There were two groups of people included in the study: 1) women from southern Africa, who represented about half of the study population, and 2) men who have sex with men from either Peru or the United States. The participants were randomized to receive oral acyclovir [Zovirax], or not to receive it.

There was clearly a decrease in herpes simplex episodes and ulcerations, but there was no difference in HIV acquisition, none at all, which was very disappointing. The question is: Why is that? The short answer is: We don't know.

There was some speculation that perhaps the association between herpes and HIV is that they are acquired simultaneously. Therefore, once somebody has HSV, suppressing it will not prevent the HIV transmission, because it would have already occurred by then. However, it was very clear that this particular intervention didn't work and that, therefore, suppressing herpes in people will not help them avoid HIV transmission.

Maraviroc Levels in the Genital Tract

There was what we might want to call preclinical information presented regarding the new drug maraviroc [MVC, Selzentry, Celsentri]. Maraviroc is a CCR5 antagonist. It's a relatively new drug. It works by inhibiting cellular entry using the R5 receptor. Almost all HIV transmission occurs through the R5 receptor. People who are already infected can have virus that utilizes either R5 or X4, but at the time of infection, transmission almost always occurs through the R5 receptor.

Click to enlarge For additional slides from this study, click here.

This study by Dumond et al looked at the use of maraviroc in HIV-uninfected women, and looked at the drug levels and other pharmacokinetic parameters in the genital tract.²⁰ The researchers studied 12 women, of whom 33% were African American. This is a U.S.-based study.

The researchers found very interesting results. Basically, the drug level in the genital tract -- specifically, undiluted cervical vaginal fluids -- was 11-fold higher than the drug level in the plasma.

The drug level in the plasma is an inhibitory dose. It's high enough to prevent viral replication, but the implications of having a 1-log higher drug level in the genital tract for a drug that specifically blocks the CCR5 receptor, where all transmission must occur utilizing the CCR5 receptor, does give us some real hope that this could be a very effective preventive measure in HIV-uninfected women.

In this study, the drug was taken by mouth. For preventive therapy, the drug can either be administered orally or locally. This study does not address the use of the drug locally. It did, however, have one additional piece of information that was very interesting: The researchers measured how much protein binding there was for this drug in the genital tract.

In the plasma, for almost all drugs, there is a huge amount of protein binding, like up to 90%. It is only the unbound portion of the drug that can work. For maraviroc, in these same women, the researchers looked at the blood plasma protein binding, which was 76%, in keeping with most drugs. However, the local genital tract protein binding was only 7.6%, suggesting that not only is the drug there in much higher concentrations, it's also much more available, which may, again, increase its ability to prevent transmission.

Hormonal Contraception and HIV Transmission

Click to enlarge For additional slides from this study, click here.

Let me finish up by presenting not primary data, but information that was presented at a symposium regarding prevention of HIV transmission in women. First, Elizabeth Stringer gave a summary talk about the effect of hormonal contraception on HIV acquisition, HIV disease transmission, and HIV shedding in the genital tract.²¹

This has been a very controversial area for quite a while and it's an enormously important area, because preventing pregnancy, for women everywhere, but especially in the developing world, is critical to women's health and wellbeing, and the health and wellbeing of their children, and it is mostly done through hormonal means. So, if there were effects of hormones on HIV acquisition or disease progression, or shedding, these could have enormous population-based implications, as well as real implications for the individual woman's health.

I'll first discuss the topic of HIV acquisition. The most recent study on this was published in 2007.²² It was a large study of more than 6,000 women. There was a fairly low seroconversion rate of 2.8 per 100 person-years. There was no influence of Depo-Provera or oral contraceptives on HIV acquisition. Specifically, the odds ratios of HIV acquisition were 0.9 for Depo, and 1.0 for oral contraceptives, which, clearly, was not statistically significant.

However, prior studies have indicated that there may be some effect. There was a study in the journal AIDS in 2004, which studied 1,272 women and demonstrated that with hormonal contraception, there was a 1.8-fold increased likelihood of HIV acquisition if it was Depo-Provera, 1.5-fold higher likelihood if it was oral contraceptives, and 1.6-fold more likely if it was Norplant.²³

The transmission rate in this study was higher, 4-fold almost; it was 8.5 per 100 person-years. It's really this study that in the past has raised the concern. Dr. Stringer also showed a summary slide of all the published studies in the past, most of which show no effect, i.e., that there is neither an increased nor a decreased transmission rate if the woman is using a hormonal contraceptive.

Hormonal Contraception and HIV Disease Progression

Let me next address the influence of hormones on disease progression. Again, the data is contradictory. There's the Mombasa cohort, which is a seroconversion study, that looked at the viral load set point in women who were taking hormonal contraception and found that it was 0.33-logs higher in the hormonal contraception users.²⁴ Now, 0.33-logs is a two-fold -- that's doubled -- viral load set point. A 0.33-log difference does predict, in general, in most populations, a faster disease progression.

In addition, those same authors found that women who were taking hormonal contraception had multiple viral variants, were basically 4-fold more likely to have multiple viral variants at the time of transmission, which was also associated with a more rapid CD4+ cell count decline.

There is a third study, a study from Zambia, in which women were randomized to receive either an IUD or hormonal contraception.²⁵ Of note, the women with an IUD had more effective contraception, whereas the women on hormonal contraception had a 2.2-fold higher likelihood of becoming pregnant during the study. The women on hormones were more likely to have a CD4+ cell count decline to less than 200 cells/mm³, and were more likely to have disease progression in toto measured as CD4+ cell count decline or death.

In a time variant analysis, meaning where the researchers looked at whether the woman remained on hormonal contraception, the researchers found that really only the Depo-Provera was associated with a greater likelihood of progression, defined again as death or a CD4+ cell count decline. It was about a 50% more likely decline.

Again, there was a summary slide shown of all the studies that have addressed this, and the findings are completely inconclusive -- meaning, some studies show an effect, and some studies show no effect.

Finally, Dr. Stringer addressed HIV shedding. The short answer is: There's probably an effect, but again, we really don't know. All of this suggests that we need better research in this area, which is now occurring, but for many years, was not occurring.

Factors That Influence HIV Transmission

There were two more presentations that addressed the underlying scientific paradigms that we should consider when looking at HIV prevention in women. In the first, Rupert Kaul discussed why we would think we could make interventions that could have an effect, that could decrease transmission.²⁶ What are the key concepts in transmission? One is: How much virus is there? What are the susceptible cells that are present? In other words, how much virus does the transmitter have? How much susceptibility is there in the person who is uninfected? What are the coinfections in each of the two parties?

Ninety percent of the transmission of HIV infection comes from only 20% of the HIV-infected people.²⁷ The factors that influence that are: viral load; the semen enhancer for viral infectivity, which is not really well defined; coinfections; and host genetics. Those are factors having to do with the transmitter.

In the susceptible person, the factors that influence transmission are: how many susceptible cells are around; what are the other inflammatory mediators that are present; and again, host genetics, which we think plays a large role, but we have not defined it very well.

All of this would suggest that preventing sexually transmitted infections should prevent HIV transmission, but, in fact, as you've seen from some of the data I presented earlier,¹⁹ mostly this doesn't seem to be true. So, treating herpes simplex does not prevent HIV acquisition.

Dr. Kaul presented data on bacterial vaginosis, which, again, seemed not to have an impact on HIV transmission. He had two points to make about this: 1) we need further study, and 2) condoms remain the single most effective way to prevent HIV transmission, and should continue to be promoted aggressively.

Click to enlarge

Finally, let me mention a little bit about the local environment in the female genital tract, and why we have not succeeded in our microbicide efforts to prevent transmission to women. This is information that was presented by Betsy Herold's group.²⁸ It goes back to the basic science of why something would work or not work. I think everyone knows the story of nonoxynol-9 -- because it kills HIV in vitro, it was expected that this would be a very effective way to prevent HIV transmission to women.²⁹ In fact, what happened in clinical trials was that it enhanced HIV transmission to women. It really significantly enhanced it. There was a lot of speculation about why this was true, but it's only recently that we have found information that tells us why this is true.

There are two really major factors here. One is that the epithelium provides a natural barrier against infection and there are things that can disrupt that barrier directly.

The second is that there are small proteins in the vagina that protect us from being infected with a variety of pathogens, including viruses. It turns out that nonoxynol-9 both disrupts the epithelial barrier by breaking up what's called the tight junctions between the cells, allowing virus to cross the barrier and it also elicits an inflammatory response that leads to a loss of some of the protective proteins and an increase in inflammatory proteins. The inflammatory proteins could recruit more HIV target cells into the genital tract and, when combined with the disruption in the epithelial barrier, could enhance HIV infection.

Dr. Herold's group has found this to be true, both for nonoxynol-9 and for cellulose sulfate. Cellulose sulfate is the second microbicide that went into clinical trials and was also found to enhance transmission, and therefore the studies were stopped early.³⁰ Now there is some good news here, which is that there are other microbicides coming along -- specifically something called PRO 2000,³¹ tenofovir [TDF, Viread],³² and maybe we should consider maraviroc, which has not been studied yet.

At the same dose, PRO 2000 also has effects on the tight junction proteins, but the effects are substantially less than those observed with cellulose sulfate or nonoxynol-9 and do not lead to increased migration of HIV across the epithelium. Importantly, in the ongoing clinical studies, PRO 2000 is formulated at a lower concentration (0.5%) than was cellulose sulfate (6%).^31,33 Tenofovir, which is also being studied as a gel in several studies, did not disrupt the epithelial barrier at all and did not activate the inflammatory pathways.^32,34 Therefore, tenofovir is not likely to increase HIV replication or recruit in new cells. Going back to what Dr. Kaul mentioned, you have to have susceptible cells there, whereas cellulose sulfate and nonoxynol-9 do recruit new cells and increase HIV replication.

Basically, Dr. Herold's group is suggesting that the safety evaluation of anything that's going to be considered as a microbicide needs to be evaluated prior to going into clinical trials in women. There are ways to evaluate it. Specifically, their group has developed both a cell and explant culture model and a mouse model -- and remember, safety includes not getting HIV. You don't want to be giving people something to prevent HIV when it's actually enhancing HIV transmission.

Our understanding of the basic science explains why we have not succeeded with nonoxynol-9 and cellulose sulfate and, importantly, gives us some hope about the new substances that we'll be able to test in the near future. Also, importantly, it gives us a mechanism to assess safety prior to using these compounds in women who are at risk for HIV infection.

Conclusion

In summary, we've reviewed data that looks at disease progression in women, response to therapy, and some basic pathophysiology in how to prevent HIV transmission in women. We have way more information than we have ever had in the past, but I think you can see we still don't have enough. There need to be more studies that look at differences in women and men, and by ancestral history, for response to therapy, that look at some really specific issues that are of concern to women -- for example, reproductive hormones in HIV-infected women, and equally importantly, that assess how we can prevent HIV transmission to women, who continue to be at higher risk than men for acquiring HIV.

This transcript has been edited for clarity.

Footnotes

1. Grinsztejn B, Smeaton L, De Gruttola V, et al, and the PEARLS study team of the ACTG. Gender-associated differences in pre-ART plasma HIV-1 RNA in diverse areas of the world vary by CD4+ cell count. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 672.
   View poster: Download PDF
2. Lemly D, Shepherd B, Hulgan T, et al. Race and sex differences in HAART use and mortality among HIV-infected persons in care. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 810.
3. Weintrob A, Grandits G, Agan B, et al. Disparate virologic response to HAART between ethnicities. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 809.
4. Anastos K, Shi Q, Lu D, et al, and Rwanda Women's Interassociation Study and Assessment (RWISA). Association of pre-treatment nutritional status with CD4 and viral load response to ART in Rwandan women. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 667.
5. Hessamfar-Bonarek M, Chene G, Cacoub P, et al, and Mortalité 2000 & 2005 Study Group. Causes of death in HIV-infected women and their evolution since 2000: the Mortalité 2000 and 2005 surveys, ANRS EN19. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 666.
6. WIHS Data Management and Analysis Center. Women's Interagency HIV Study (WIHS) dossier October 2007. Women's Interagency HIV Study Web site.
7. Tien PC, Schneider MF, Cole SR, et al. Antiretroviral therapy exposure and incidence of diabetes mellitus in the Women's Interagency HIV Study. AIDS. August 20, 2007;21(13):1739-1745.
8. Anderson PL, Kakuda TN, Lichtenstein KA. The cellular pharmacology of nucleoside- and nucleotide-analogue reverse-transcriptase inhibitors and its relationship to clinical toxicities. Clin Infect Dis. March 1, 2004;38(5):743-753.
9. Tien P, Schneider M, Cole S, et al. ART exposure and insulin resistance in the Women's Interagency HIV Study. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 933.
10. Riddler S, Haubrich R, DiRienzo G, et al, and AIDS Clinical Trials Group 5142 Study Team. Effect of baseline characteristics on treatment outcomes in ACTG 5142: a prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 776.
11. Haubrich RH, Riddler S, DiRienzo G, et al, and the AIDS Clinical Trials Group 5142 Study Team. Metabolic outcomes of ACTG 5142: a prospective, randomized, phase III trial of NRTI-, PI-, and NNRTI-sparing regimens for initial treatment of HIV-1 infection. In: Program and abstracts of the 14th Conference on Retroviruses and Opportunistic Infections; February 25-28, 2007; Los Angeles, Calif. Abstract 38.
12. Spencer L, Karim R, Landay A, et al. Association between immune activation and HIV RNA-1 genital tract shedding. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 674.
    View poster: Download PowerPoint
13. Partisani M, Ohl J, Demangeat C, Binder-Foucard F, Nisand I, Lang JM. Premature ovarian deficiency in HIV-infected women. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 669.
14. Conde DM, Pinto-Neto AM, Costa-Paiva L. Age at menopause of HIV-infected women: a review. Gynecol Endocrinol. February 2008;24(2):84-86.
15. Hitti J, Paul K, Agnew K, et al. Protective effect of vaginal Lactobacillus on genital HIV-1 RNA concentrations: longitudinal data from a US cohort study. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 27LB.
16. Tobian A, Serwadda D, Quinn T, et al. Trial of male circumcision: prevention of HSV-2 in men and vaginal infections in female partners, Rakai, Uganda. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 28LB.
17. Wawer M, Kigozi G, Serwadda D, et al. Trial of male circumcision in HIV+ men, Rakai, Uganda: effects in HIV+ men and in women partners. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 33LB.
18. Quinn TC. Circumcision and HIV transmission. Curr Opin Infect Dis. February 2007;20(1):33-38.
19. Celum C, Wald A, Hughes J, et al, and HPTN 039. HSV-2 suppressive therapy for prevention of HIV acquisition: results of HPTN 039. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 32.
20. Dumond J, Patterson K, Pecha A, et al. Maraviroc (MVC) pharmacokinetics (PK) in blood plasma (BP), genital tract (GT) fluid and tissue in healthy female volunteers. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 135LB.
21. Stringer E. Impact of hormonal contraception on HIV acquisition, cervico-vaginal viral shedding, and disease progression. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 94.
22. Morrison CS, Richardson BA, Mmiro F, et al, for the Hormonal Contraception and the Risk of HIV Acquisition (HC-HIV) Study Group. Hormonal contraception and the risk of HIV acquisition. AIDS. January 2, 2007;21(1):85-95.
23. Lavreys L, Baeten JM, Martin HL Jr, et al. Hormonal contraception and risk of HIV-1 acquisition: results of a 10-year prospective study. AIDS. March 5, 2004;18(4):695-697.
24. Baeten JM, Lavreys L, Sagar M, et al. Effect of contraceptive methods on natural history of HIV: studies from the Mombasa cohort. J Acquir Immune Defic Syndr. March 2005;38(Suppl 1):S18-S20.
25. Stringer EM, Kaseba C, Levy J, et al. A randomized trial of the intrauterine contraceptive device vs hormonal contraception in women who are infected with the human immunodeficiency virus. Am J Obstet Gynecol. August 2007;197(2):144.e1-144.e8.
26. Kaul R. Vaginal flora, co-infections and immune cells in the genital mucosa: implications for HIV transmission. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 96.
27. Lloyd-Smith JO, Schreiber SJ, Kopp PE, Getz WM. Superspreading and the effect of individual variation on disease emergence. Nature. November 17, 2005;438(7066):355-359.
28. Mesquita P, Wilson S, Cheshenko N, Keller M, Galen B, Herold B. Comprehensive pre-clinical assessment of microbicide safety using in vitro and murine models. In: Program and abstracts of the 15th Conference on Retroviruses and Opportunistic Infections; February 3-6, 2008; Boston, Mass. Abstract 26.
29. Hillier SL, Moench T, Shattock R, Black R, Reichelderfer P, Veronese F. In vitro and in vivo: the story of nonoxynol 9. J Acquir Immune Defic Syndr. May 1, 2005;39(1):1-8.
30. Van Damme L, Govinden R, Mirembe F, et al. Phase III trial of 6% cellulose sulfate (CS) gel for the prevention of HIV transmission. In: Program and abstracts of the 4th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention; July 22-25, 2007; Sydney, Australia. Abstract WESS301.
31. Fletcher PS, Shattock RJ. PRO-2000, an antimicrobial gel for the potential prevention of HIV infection. Curr Opin Investig Drugs. February 2008;9(2):189-200.
32. Peterson L, Taylor D, Clarke EEK, et al. Findings from a double-blind, randomized, placebo-controlled trial of tenofovir disoproxil fumarate (TDF) for prevention of HIV infection in women. In: Program and abstracts of the XVI International AIDS Conference; August 13-18, 2006; Toronto, Canada. Abstract THLB0103.
33. Keller MJ, Zerhouni-Layachi B, Cheshenko N, et al. PRO 2000 gel inhibits HIV and herpes simplex virus infection following vaginal application: a double-blind placebo-controlled trial. J Infect Dis. January 1, 2006;193(1):27-35.
34. Mayer KH, Maslankowski LA, Gai F, et al, HPTN 050 Protocol Team. Safety and tolerability of tenofovir vaginal gel in abstinent and sexually active HIV-infected and uninfected women. AIDS. February 28, 2006;20(4):543-551.