BIOMEDICAL PREVENTION OF HIV IN WOMEN: CHALLENGES AND APPROACHES, WITH PARTICULAR REFERENCE TO THE VAGINAL MICROBIOME

Abstract

Although great progress has been made in treating HIV-1 infection globally, incidence remains high in many locales, especially sub-Saharan Africa. Two-thirds of the burden of disease is there, and among those infected, 60% are women. Pre-exposure prophylaxis, the administration of a drug to prevent acquisition of infection, has proven to be a promising prevention modality for HIV-1 when administered as a daily oral regimen of tenofovir disoproxil fumarate/emtricitabine. However, data suggest that oral tenofovir disoproxil fumarate/emtricitabine may not achieve protective concentrations in the cervicovaginal tissue as rapidly as it does in the rectum. Moreover, a relative paucity of the beneficial vaginal bacteria Lactobacillus crispatus, along with a commensurate increase in the vaginal anaerobes that characterize bacterial vaginosis, strongly increases risk of HIV-1 acquisition and may even modify efficacy of topical tenofovir when used as pre-exposure prophylaxis. bacterial vaginosis, a globally prevalent infection that increases women’s risk of HIV-1 acquisition, and presents serious other reproductive health risks and management challenges, presumably involves survival of treatment-resistant biofilm communities. Methods to effect sustained improvement in the vaginal microenvironment are urgently needed.

INTRODUCTION

The World Health Organization estimates that in 2016, approximately 1.8 million individuals were newly infected with HIV-1 every year, predominantly through heterosexual intercourse; eastern and southern Africa accounted for 43% of the global total of new infections (1). Worldwide, women continue to carry a disproportionate burden of HIV-1 infection, particularly in sub-Saharan Africa where 59% of people living with HIV-1 are women. Pre-exposure prophylaxis (PrEP), the administration of a drug to prevent acquisition of infection, has proven to be a promising prevention modality for HIV-1 when administered as a daily oral regimen of tenofovir disoproxil fumarate/emtricitabine (TDF/FTC). However, data suggest that oral TDF/FTC may not achieve protective concentrations in the cervicovaginal tissue as rapidly as it does in the rectum, thus providing better and faster protection for unprotected receptive anal intercourse than vaginal intercourse (2).

Success and Challenges in the Current PrEP Landscape

Among the first indications for the biological plausibility of oral PrEP was the marked reduction in perinatal HIV transmission among HIV-infected women taking oral antiretrovirals. Prevention of maternal-to-child transmission is now a widely promoted and established strategy. Moreover, PrEP as a general concept is supported by its routine use for several infectious diseases, for example, malaria and Pneumocystis jirovecii pneumonia.Tenofovir and TDF/FTC were logical candidates for PrEP because of their potent inhibition of the HIV-1 reverse transcriptase, generally good safety and tolerability profiles in HIV-infected persons, high barrier to HIV-1 resistance, and favorable pharmacokinetics (once daily dosing that need not occur with food intake). Importantly, when taken orally, these drugs achieve concentrations in the genital secretions that are 2- to 6-fold that of blood plasma (3). For tenofovir (TFV) to be active, it must be taken up by relevant mononuclear cells and phosphorylated to its active form, TFV-diphosphate. The intracellular half lives of both TDF and FTC are long (40 to >100 hours), again providing support that substantial concentrations of these drugs in the genital tract could theoretically provide adequate activity against “incoming” HIV.

PrEP uptake globally has been highest among men who have sex with men, and has driven HIV incidence in this group to the lowest levels seen since the beginning of the HIV epidemic (4). The story for women is not so encouraging. First, the randomized clinical trials that evaluated oral and topical PrEP in women have, with the exception of the Center for the AIDS Programme of Research in South Africa (CAPRISA 004) study, been disappointing. CAPRISA 004 randomized women to use placebo or vaginal TFV gel (1%) before and after intercourse and showed a significant reduction in HIV acquisition by 49% (5). To advance this concept towards licensure, two subsequent trials — the vaginal & oral interventions to control the epidemic (VOICE) Trial (MTN 003) and Follow-on African Consortium for Tenofovir Studies (FACTS-001) — assessed daily and periocoital use of the same formulation of TFV gel, respectively (6,7). Unfortunately, product adherence was low in both of these trials, although in secondary analysis, women who used TFV gel as measured by plasma TFV levels in VOICE did experience a 50% reduction in HIV-1 incidence (8). Nonetheless, poor adherence is a principal reason for the inability of these studies to show TDF-based efficacy as PrEP for HIV-1 in women.

A Role for the Vaginal Microbiome

A second possibility for the apparently higher challenge to protection from HIV-1 in women has emerged: a role for the vaginal microbiome, particularly as it relates to modulation of local inflammation. In women of reproductive age, a vaginal environment that is quantitatively dominated by hydrogen peroxide–producing Lactobacillus species typically has a pH considered to be normal (<4.7), and has consistently been associated with healthy pregnancy outcomes, lack of abnormal vaginal symptoms, and reduced risk for acquiring several sexually transmitted pathogens, including HIV. The most commonly isolated Lactobacillus species associated with this healthy environment are L. crispatus and L. jensenii. In addition to the direct antimicrobial effects of lactic acid and the consequent acidic pH these bacteria engender, the natural defense mechanisms of the vagina include production of endogenous defensins [including human neutrophil peptide-1 (HNP1-3)], secretory leukocyte protease inhibitor, cytokines, and endocervical mucous (9). Bacterial vaginosis (BV) is a very common condition in which these protective lactobacilli are replaced by high quantities of commensal anaerobes, often resulting in symptomatic vaginitis. Relief from antibiotic therapy of BV — metronidazole or clindamycin — is often short-lived, with most women subject to recurrence in the next several months unless ongoing antibiotic therapy (biweekly vaginal metronidazole gel) is used to suppress it. Because BV is associated with additional adverse health consequences, including preterm delivery and pelvic inflammatory disease, and with deleterious effects on sexual and genital health (10), delineating the pathogenesis of recurrence and optimizing its prevention are major concerns. Critically, the initial event leading to the shift of the anaerobic predominance that characterizes BV is unknown, although data suggest that sexual intercourse likely contributes — at least in some women. Regardless, the role of beneficial Lactobacillus species is emphasized by the value (short-term and long-term) of vaginal birth over C-section. Infants born vaginally are rapidly colonized with vaginal bacteria (11). Transfer of optimal maternal microbiota to infants through this modality has been associated with lower rates of autoimmune diseases (asthma), and metabolic disorders (12).

While traditional cultivation has identified numerous BV-associated bacteria involved in these processes, many additional bacteria have been recently discovered using cultivation-independent methods. New tools in molecular biology have greatly expanded our understanding of microbial diversity in the human vagina by identifying fastidious or cultivation-resistant bacteria (13). Molecular methods have been widely adopted to study microbial populations because cultivation-based approaches do not detect or identify a large percentage of microbes present in most niches, including those associated with humans. The vaginal microbiota has been explored using several molecular methodologies. These include broad-range or consensus sequence polymerase chain reaction (PCR) of ribosomal RNA genes, in which primers that anneal with highly conserved regions of the bacterial 16S rRNA gene can be employed in a PCR to amplify segments of the gene from most bacterial species, as well as whole-genome sequencing. A complementary molecular approach to broad range PCR is taxon-directed PCR in which one designs primers to PCR amplify a specific microbe or microbial group (14). Fluorescently labeled probes may also be used in a real-time PCR format to increase the specificity of the assay and to generate quantitative data (15). Finally, fluorescently labeled oligonucleotide or peptide nucleic acid probes targeting bacterial ribosomes can be hybridized with fixed bacteria in tissues or body fluids to detect and identify bacteria by microscopy. For instance, several novel bacteria in the Clostridiales order are highly specific indicators of BV, and bacteria related to Megasphaera, Leptotrichia, Atopobium, and Dialister species are commonly found in women with BV. These advances have notable implications for research related to the pathogenesis, natural history, diagnosis, and adverse consequences of BV. In addition, largely prompted by investigations into the transmission of HIV, our understanding of innate and acquired immunity as it relates to the genital tract and the microbiota has improved significantly in the last several years.

Given the high prevalence of BV — and of some BV-associated bacteria such as Gardnerella vaginalis — in many populations, it is reasonable to ask what “normal” vaginal microbiome really comprises. In recent years, many experts have shifted to use the term “optimal” to describe what many continue to call normal. In this context, the optimal vaginal microbiome would promote 1) absence of abnormal vaginal symptoms, especially unusual discharge or odor; 2) relative protection against infection with reproductive tract pathogens, including sexually transmitted infections and HIV; and 3) healthy pregnancy and delivery of healthy, at-term infants. Whatever term is used, however, nearly all investigators agree that certain vaginal Lactobacillus species probably help women to resist vaginal and cervical infection, and to attain healthy pregnancy outcomes. The mechanism for this, however, is less clear. Some vaginal lactobacilli inhibit G. vaginalis, Mobiluncus, and anaerobic Gram-negative rods in vitro. Some vaginal strains of Lactobacillus produce hydrogen peroxide (H₂O₂), and several studies have shown that H₂O₂-producing strains of lactobacilli more frequently colonize the vagina of normal women, compared to women with BV. Women colonized vaginally with H₂O₂-positive lactobacilli less often developed BV than did those colonized with H₂O₂ negative lactobacilli (16). H₂O₂ may inhibit the growth of anaerobic rods, Gardnerella, Mobiluncus, and Mycoplasma in the vagina either directly through its toxic activity or by reacting with a halide ion in the presence of cervical peroxidase as part of the H₂O₂-halide-peroxidase antibacterial system. More recent attention has focused on the role of lactic acid and its direct effect on anaerobic vaginal bacteria, and some investigators, citing work in in vitro models, have suggested that it has a distinct, and perhaps more important, role than H₂O₂ (17). Regardless, the shift to BV is accompanied by production of volatile amines, a rise in vaginal pH, and marked depletion of key Lactobacillus spp. such as L.crispatus. L.crispatus appears to play an important role in defense against pathogens through the production of lactic acid, bacteriocins, and other antimicrobial molecules. Several studies have detected a polymicrobial biofilm in women with BV that is adherent to vaginal epithelial cells and absent in healthy controls (18).

When discussing optimal vaginal microbiota, an additional layer of complexity is the dimension of time. Molecular methods have been used to study how the composition and concentrations of vaginal bacteria change over the course of the menstrual cycle. For example, one study used quantitative PCR to assess changes in concentrations of key vaginal bacteria over the course of a month (19). This study showed that there are statistically significant decreases in concentrations of several Lactobacillus species with onset of menses, such as with L. crispatus and L. jensenii. Conversely, there is a significant increase in concentrations of G. vaginalis during menses, although individual women can manifest very different patterns. These data, and data from other longitudinal studies, suggest that the vaginal microbial ecosystem is dynamic in many women, and subject to both internal factors (menses, hormones) and external factors (sexual activity, intravaginal product use). The changes in bacterial populations seen during menses likely reflect the availability of different nutrients for bacterial growth.

RELATIONSHIP BETWEEN BV AND HIV-1 ACQUISITION

A growing body of evidence suggests that the presence of BV, or absence of vaginal lactobacilli, may increase a women’s risk of acquiring HIV via heterosexual intercourse (20,21). The relationship between BV and HIV acquisition may be most synergistic — and most alarming — in areas of the world where both are highly prevalent. The prevalence of BV among many women in sub-Saharan Africa has been 50% or greater in many studies; with HIV seroprevalence up to 38% in some of these countries, even if BV confers only a small increase in the relative risk for HIV acquisition, the population attributable risk could be considerable. Critically, BV in women infected with HIV confers an increased risk of HIV transmission to male partners (22). BV may elevate these risks in several ways, including upregulation of relevant T-cell populations by the high burden of associated vaginal anaerobes; reduced levels of protective factors (defensins, secretory leukocyte protease inhibitor) and increased levels of inflammatory factors (certain cytokines); directly increasing quantity of HIV shed genitally; and loss of H₂O₂, which is virucidal.

Exciting recent data indicate a role for local inflammation in mediating the increased risk of HIV-1 acquisition in women with BV. Using samples from the CAPRISA 004 study, Mlisana et al. (23) were the first to define an inflammatory signature derived from nine cytokines measured in vaginal fluid. Women who had this profile were at considerably increased risk for acquiring HIV-1. These data have been recently expanded in a paper exploring the role of inflammation as it relates to the microbial communities in the vagina and surprisingly, to a role for modifying the efficacy of TFV-based topical PrEP. McKinnon et al. (24) recently showed that in women without genital inflammation, tenofovir was 57% protective against HIV (95% confidence interval: 7% to 80%) but was only 3% protective (95% confidence interval: -104% to 54%) if genital inflammation was present. Even among women who adhered well to using the gel, TFV protection was 75% in women without inflammation compared to -10% in those with inflammation. Using samples from the same-study participants, Klatt et al. (25) showed that women whose vaginal microbial communities (defined using mass spectrometry/proteomics) were dominated by Lactobacillus species experienced reduced rates of HIV-1 acquisition (61% reduction; P = 0.013) relative to those dominated by G. vaginalis and other anaerobes (18% reduction; P = 0.644). Detectable mucosal TFV was lower in women in the latter group, and in vitro experiments suggested that G. vaginalis itself might be degrading this drug. McClelland et al. (26) also recently reported that specific fastidious BV-associated bacteria are associated with susceptibility to HIV infection in African women in a nested case-control study derived from prospective cohorts of women in Kenya, Uganda, Tanzania, Botswana, South Africa, and Zambia.

Perhaps the most compelling observational data to date that emphasize the potential role the vaginal microbiome might play in modifying HIV-1 acquisition risk comes from the Females Rising through Education, Support, and Health (FRESH) Cohort, a group of young women (18 to 23 years old) enrolled in KwaZulu Natal, South Africa. Among 200 women followed for 1 year, HIV-1 incidence was high (more than 8%) (27). The investigators sequenced bacterial 16S rRNA V4 gene from cervical swabs and categorized the communities into four cervicotypes, ranging from one dominated by L. crispatus (CT1), one by L. iners (CT2), and increasingly diverse communities composed of a wide variety of anaerobes (CT3 and CT4). Risk of HIV-1 acquisition increased markedly with increased dysbiosis, and remarkably, none of the women in the CT1 group acquired HIV-1 throughout the course of the study.

DISCUSSION

Despite considerable research effort and recent advances, BV remains an enigmatic condition. Efforts to link BV to a single cultivated bacterial pathogen have been unconvincing. Molecular tools have confirmed that BV is most likely a heterogeneous syndrome caused by different communities of vaginal bacteria. If this view is correct, BV can be considered a dysbiotic condition caused not by a single pathogen but by a change in microbial composition and community structure. A woman’s individual risk for a acquiring a particular etiologic vaginal bacterial community might depend on specific practices, such as unprotected sex or douching. Future studies of BV and its associated adverse outcomes should determine if specific combinations of organisms are more pathogenic than others, and causally associated with different adverse events. These studies will shed light on the changes in vaginal microbiota that correlate with BV and the dynamic microbial ecology of the human vagina. Are there keystone species in BV that form the initial wave of bacteria that facilitate colonization with other species, or does the entire community of BV-associated bacteria become established simultaneously? What changes in the vaginal environment precede the development of BV? Finally, determination of causality will depend not only on more precise categorization of the vaginal microbiota, but also on variations in the host environment that may be associated with changes in bacterial communities over time. Although current treatments afford reliable symptom relief, failure rates for BV and vulvovaginal candidiasis (VVC) are as high as 83% and 60%, respectively, and in which the survival of resistant biofilm communities have been increasingly implicated. These failure rates represent an important public health concern because these infections are associated with a host of serious complications as noted.

In terms of provision of PrEP for women, we must make available products that work and that women will use. Sustained delivery of antiretroviral PrEP with an intravaginal ring may offer a viable alternative to oral tablets or vaginal gels. Intravaginal rings are currently used by millions of women to deliver contraceptives. Qualitative research performed in the VOICE and Preexposure Prophylaxis Trial for HIV Prevention among African Women (FEM-PrEP) trial participants post-trial indicated that even in places where HIV transmission is rampant, women at risk are more interested in obtaining contraception than biomedical products aimed solely at HIV prevention (28). Results of the first phase 3 clinical trials using a silicone intravaginal ring to deliver dapivirine, A Study to Prevent Infection with a Ring for Extended Use (ASPIRE) and ring studies, were published in 2016 (29,30). Both showed an approximately half-fold reduction in incident HIV-1 acquisition, with efficacy related to age (lowest in the youngest women). In subsequent analyses, adherence was deemed to be the reason for most of this lack of protective effect; however, other factors remain under exploration.

Despite these advances, there are significant gaps in our knowledge concerning the vaginal microbiota. What are the consequences of untreated vaginal infections, particularly BV, and how do these complications arise? What interventions have and have not worked to prevent these consequences, and how do we leverage this knowledge to move forward with appropriate prevention strategies? How might advances in molecular biology contribute to an enhanced approach to diagnosing BV, and to a better understanding of how a disrupted vaginal microbiome may facilitate infection with other reproductive tract pathogens, such as N. gonorrhoeae and Trichomonas vaginalis? A more complete understanding of vaginal microbial populations and the host factors that regulate both pathogens and commensals may lead to better strategies to maintain healthy vaginal microbial communities — thus enhancing women’s health — and create opportunities to explore the role of novel bacteria in reproductive tract diseases.

DISCUSSION

Martin, New Orleans: Great talk as always and I certainly agree with you — a multicomponent approach is really important. I was struck though by the study of, in men in Holland, where they used pre-exposure prophylaxis (PrEP) followed by post-exposure prophylaxis (PEP) and had good outcomes, and I wonder if that approach might improve outcomes in women. I’d like your thoughts on that.

Marrazzo, Birmingham: Great, great comment. So, the study in Paris, actually in France, was the IPERGAY study. Those men used oral Truvada right before sex and after, and it worked beautifully. The only study that has done that in women used vaginal tenofovir gel and that was the CAPRISA-004 study and that did show a reduction of 50% of HIV acquisition in women. So, I think it is theoretically possible with the right product. I think it has to probably be topically delivered if you’re talking about tenofovir because it takes about 5 days for tenofovir to reach steady state in the cervical and vaginal tissues. And so, the pre-exposure dose orally is just not enough to protect that side. But I think a topical product like a ring or a gel would be very, very appropriate in that setting.

Licht, Gainesville: This reminds me — I just heard an interesting talk the other day from probably a colleague you might know, Erin Kobetz from Miami, who has done some ethnographic work that may help us understand the HIV epidemic in Haitians in South Miami and found that a lot of linkage had to do with feminine hygiene habits. I was wondering if some of the dysbiosis you’re seeing here could be due to that. And she learned about all kinds of herbs that are used in things that she had no idea about. So, wondering in the South African cohorts if some of this dysbiosis you’re seeing might be used to previously not well-known habits and personal habits.

Marrazzo, Birmingham: Great question and there’s a lot of work actually on vaginal hygiene and HIV risk in a very nice meta-analysis by Nicola Low recently in PLoS Medicine. I am certain that especially in Africa those practices contribute to a sort of constant challenge to maintaining an optimal vaginal environment. I think it’s a vicious cycle because women start to have symptoms, then they start to use those practices, and then they continue undermine the ability of the Lactobacillus crispatus to gain a foot hold and then reestablish its population. So, I’m almost certain it’s a huge player. Education about that is much needed, but it’s a challenging area.

Crowley, Boston: It’s a lovely talk. For many years it has seemed to me that vaginal rings are the way to go with a lot of this. And yet there’s been tremendous resistance in the restriction of polysilicone as the major one. They are large and clunky rings. Myself and Bob Langer have patents on ethylene-vinyl acetate rings working with a pharmaceutical company in Boston. I think that is what we should be targeting if we could just get it....but the resistance to this is enormous, and I don’t get it.

Marrazzo, Birmingham: You are a man after my own heart. So, there is a dapivirine vaginal ring that I did not have time to talk about that was reported in the New England Journal of Medicine in January. Two large studies showed a reduction in HIV acquisition in the most adherent women of 50%. If we had a vaccine that was 50% effective, I suspect that would have been front-page news. The other issue is that rings are beautiful because you can combine them with contraception. So, there are now tenofovir rings that include hormonal contraception and that way you can address young women who also need pregnancy prevention. So, I cannot agree with you more.