HIV Articles  
Microbicides to Prevent HIV Transmission: Overcoming Obstacles to Chemical Barrier Protection  
  The Journal of Infectious Diseases Jan 1, 2006;193:36-44
Darpun Dhawan1 and Kenneth H. Mayer1,2,3
1Brown Medical School, and 2Infectious Disease Division, The Miriam Hospital, Brown University, Providence, Rhode Island; 3Fenway Community Health, Boston, Massachusetts
Background. Microbicides are topical compounds that could prevent sexually transmitted infections. Several compounds have demonstrated activity both in vitro and in animal models, but none has been approved for use in humans.
Methods. A review of >100 recent publications from MEDLINE (through October 2005) and abstracts presented at recent conferences was undertaken to describe the current status of microbicide research and to delineate why microbicides are not yet available.
Results. More than 15 candidate microbicides are currently being studied in clinical trials. Their mechanisms of action include disruption of the viral membrane by surfactants, maintenance of an acidic vaginal pH, binding to the viral envelope to block receptor binding, and blocking of receptors; they may also be combined with antiretroviral drugs. The development of safe and effective microbicides has been delayed by limitations in understanding the biological processes of human immunodeficiency virus (HIV) transmission, difficulties in extrapolation from animal models, lack of established correlates of protection, and the need to enroll and follow large cohorts of high-risk participants for several years in order to demonstrate efficacy.
Conclusions. Safe and effective topical microbicides are biologically plausible. Several trials that are under way may demonstrate the ability of microbicides to protect against transmission of HIV, but multiple challenges remain.
The majority of HIV-infected women acquire the virus through unprotected heterosexual intercourse [1], which reflects inherent biological and social vulnerabilities to infection [2]. Although topical microbicides have been touted as a means of female-controlled HIV infection prevention for more than a decade [3], results from efficacy trials of active agents may only become available during the next 3-10 years.
The importance of product safety was the major lesson learned from the first efficacy trials of nonoxynol-9 (N-9), which is the active ingredient of many over-the-counter spermicides. N-9 nonspecifically disrupts cell membranes and has in vitro activity against bacterial membranes and the HIV envelope [4]. However, randomized controlled trials failed to demonstrate that N-9 has a protective effect against HIV infection [59]. Moreover, studies found that, with frequent use, N-9 increases inflammation of the cervicovaginal epithelia, which increases susceptibility to HIV infection. Thus, monitoring clinical safety is a priority in current microbicide research [1014].
Researchers are now investigating >30 other microbicides with diverse mechanisms of action. As of October 2005, there were 6 unique compounds in phase 1 trials, 4 compounds in phase 2 trials, 1 compound in phase 2/2b trials, and 4 compounds in phase 3 trials [15]. This new group of randomized controlled trials, which will include about 30,000 women worldwide, will collect data regarding the safety, efficacy, and acceptability of microbicides. This review draws from recent research on the mechanisms of HIV transmission and developments in the science of HIV prevention to provide an evaluation of the current status of and challenges to the development of a safe and effective microbicide.
There remain several unresolved issues with regard to the sexual transmission of HIV. These issues include identification of the types of cells and receptors in the female genital tract that are necessary for transmission to occur, determination of the unique vaginal ecological characteristics that influence transmission, and definition of innate host mechanisms that either protect against or facilitate infection.
Cellular targets. There are several cells that serve as sites of HIV entry in the genital tract, including CD4+ lymphocytes, dendritic cells, and macrophages [16-18]. HIV establishes infection by attaching its gp120 to the CD4 surface protein of helper T lymphocytes, in conjunction with the tandem engagement of CXCR4 or CCR5 chemokine coreceptors. Subepithelial dendritic cells play an important role in cervicovaginal transmission because they express DC-SIGN (dendritic cell C-specific intercellular adhesion molecular-3-grabbing nonintegrin), a ligand that efficiently binds, concentrates, and mediates transfer of HIV to CD4+ T cells [19, 20]. However, blockade of DC-SIGN alone is insufficient to inhibit HIV infection, because mannose-binding C-type lectin receptors continue to allow for HIV uptake [21]. At the mucosal surface of the genital tract, the availability of cells that express CD4, CCR5, or DC-SIGN receptors promote the selective transmission of CCR5-binding HIV strains [22, 23]. In animal models, either free virions or cell-associated virus can access transepithelial dendritic cells, facilitating efficient HIV amplification and dissemination [19, 24, 25]. Vaginal epithelia abraded by sexual trauma and/or sexually transmitted infections (STIs) may allow pathogens to pass directly into the lamina propia. Because transmission of HIV may occur through a variety of mechanisms, it is important to develop microbicidal agents that not only inhibit viral uptake by target cells but also provide a "second-line" defense that prevents replication and spread to lymph nodes [26]. Once HIV is intracellular, stopping progressive infection likely will require the use of antiretroviral medication.
Physiological factors. The large mucosal surface area that is vulnerable to microabrasions during sexual intercourse renders women significantly more susceptible to HIV infection than men [27]. Ectopy of monostratified endocervical epithelium (more common in younger women) and the use of progestins (known to think the vaginal epithelia) increase the risk of HIV infection by altering genital epithelial integrity [28-32]. Physiological variations in female genital tract mucosa necessitate testing microbicides in diverse reproductive tract conditions.
Vaginal milieu. The normally acidic vaginal pH protects against the pathogens that cause STIs, including Haemophilus ducreyi, herpes simplex virus2 (HSV-2), and Chlamydia trachomatis [33-35]. Bacterial vaginosis, which is the most common cause of abnormal vaginal discharge, is associated with depressed levels of H2O2-producing lactobacilli. The resulting increase in vaginal pH facilitates the growth of anaerobic gram-negative rods, Gardnerella vaginalis, and genital mycoplasmas [36-38]. The presence of lactobacilli and an acidic vaginal environment are correlated with reduced HIV proliferation [39-42]. These findings have led to the development of microbicidal compounds that maintain an acidic vaginal pH by facilitating the persistence of H2O2-producing lactobacilli or by buffering the basic properties of semen.
Concomitant genital infections. The presence of genital lesions increase women's susceptibility to HIV by disrupting the epithelial barrier [43, 44]. Ulcerative infections (e.g., those due to HSV-2, H. ducreyi, or Treponema pallidum) have been associated with greater risk of HIV acquisition than are inflammatory infections (e.g., gonorrhea, chlamydia, and trichomoniasis) [45-47]. However, proinflammatory cytokines associated with chronic genital inflammation may dysregulate immune mechanisms that typically limit HIV replication [48]. The evaluation of microbicides will require delineating whether broad-spectrum activity against concomitant genital infections may enhance protection against HIV.
Viral inoculum. The efficiency of heterosexual HIV transmission has been related to plasma levels of HIV RNA (which usually correlate with genital tract levels) [46, 49-51]. In the rhesus macaque model, high-dose challenges with highly pathogenic simian immunodeficiency virus (SIV) were localized to lymph nodes within 18 h of intravaginal exposure [18]. When macaques were inoculated with low doses of HIV, they became infected only after multiple exposures to SIV, although SIV-specific T cell responses were detected earlier, indicating that the localized infection may precede peripheral blood infection by several weeks [52]. In light of mounting data on increased rates of HIV transmission during acute infection and advanced disease [51, 53], microbicides will need to be validated in multiple clinical situations.
First- and second-generation microbicides currently in clinical trials tend to be broad-spectrum compounds that kill HIV by disrupting the viral envelope, altering the vaginal pH, or binding the virus to inhibit entry. Third-generation compounds, which can inhibit HIV replication or bind specific HIV receptors, are being studied in preclinical and early pilot human studies.
Viral envelope disruption. The first widely studied agent in this class was N-9, which was found to be ineffective against HIV infection, probably because it disrupts human mucosa. Another surfactant, Savvy/C31G (Cellegy), has demonstrated amphoteric properties in vitro that disrupt HIV and sperm, as well as chlamydia and HSV-2 [54, 55]. Participants in phase 1 trials of 0.5% and 1.0% C31G formulations had fewer symptoms and signs of irritation than did groups using higher doses of C31G [56-58]. To date, clinical studies have included nearly 250 women and have demonstrated safety and tolerability, thus setting the stage for upcoming phase 3 trials that will evaluate 1.0% C31G gel.
Maintenance of normal vaginal defenses. The acidic pH of the vagina has been found to have a protective effect against HIV infection; however, the pH is neutralized when semen is present. Products that buffer vaginal pH and enhance lactobacilli colonization are being evaluated. In addition, bioengineered lactobacilli are under study that secrete proteins that inhibit HIV entry [59, 60].
BufferGel (ReProtect) is a compound that has the ability to preserve a vaginal pH of 3.8-4.0. Two phase 1 trials that followed 125 women in 5 countries found the product to be safe and well tolerated [61, 62]. A significant decrease in bacterial vaginosis was also noted [61]. Although some participants reported self-limited local genitourinary signs (e.g., erythema) and symptoms (e.g., pruritis or dysuria), the majority expressed an interest in using the product if it were shown to be effective [62]. Phase 2/2b trials recently began to compare BufferGel with PRO 2000 (Indevus Pharmaceuticals). Acidform (Instead Corporation) is another microbicide being evaluated for acid-buffering and bioadhesive properties [63, 64].
Broad entry/fusion inhibitors. Sulfonated polyanions are negatively charged agents that block viral entry by interacting with gp120 viral envelope proteins and preventing coreceptor activation [65, 66]. The polyanion PRO 2000 (naphthalene sulfonate polymer) demonstrated in vitro activity against C. trachomatis, Neisseria gonorrhoeae, and HSV [67]. In a trial including HIV-infected women, cervical vaginal secretions were collected before and after application of PRO 2000 or a placebo gel. The cervical vaginal secretions obtained 1 h after application of PRO 2000, in comparison to placebo gel, were significantly more active against in vitro HIV and HSV infection [68]. Although another study found that 73% of low-risk women who applied PRO 2000 daily for 2 weeks reported mild self-limited adverse effects [69], the safety and tolerability profile of the gel will be further assessed in 2 large-scale efficacy trialsone will compare 0.5% PRO 2000 with BufferGel, and the other will compare 0.5% with 2% PRO 2000 gel.
Carraguard (Population Council) is a sulfated polymer derived from carrageenan, which has been found to prevent HIV-infected mononuclear cells from binding to vaginal epithelia [70]. On the basis of laboratory studies, carrageenan is assumed to be noncontraceptive [71], and the agent is classified by the US Food and Drug Administration as "GRAS" (generally regarded as safe), because congeners have been used in food products. Phase 3 trials have been initiated, after safety studies that found mild symptoms, such as genital irritation and bladder fullness [72].
Cellulose sulfate (Ushercell; Polydex Pharmaceuticals and TOPCAD) is another sulfonated polyanion that has been found to have in vitro activity against N. gonorrhoeae, C. trachomatis, human papillomavirus, and G. vaginalis [73-75]. A phase 1 study found that the safety profile of cellulose sulfate was comparable to that of K-Y Jelly (McNeil-PPC) and a contraceptive gel containing N-9 and that women preferred the 2.5-mL dose to the 5.0-mL dose [76]. Two phase 3 trials are under way at multiple international sites, to evaluate the efficacy of this product.
CCR5 inhibitors. CCR5 is being evaluated as a possible drug target, because there are in vitro data demonstrating the ability of RANTES (a natural ligand that binds CCR5) to block HIV infection [77], in addition to the clinical finding that individuals with a CCR532 polymorphism are noted to have slower progression of HIV disease [22, 78]. Recently, researchers demonstrated that the vaginal application of a coreceptor-binding mimetic, PSC-RANTES (Nalfa (n-nonanoyl-des-Ser1-[L-thioproline2, L-alfa-cyclohexylglycine3]RANTES), blocked intravaginal challenge with SHIV-SF162 (a chimeric simian/human immunodeficiency virus) in 12 of 15 macaques that had been pretreated with progesterone to thin the genital epithelium [79]. With the success of blocking CCR5 receptors in an animal model, this approach seems promising, although concerns have been raised that there could be selection for viral variants that are CXCR4 trophic or that bind CCR5 with greater affinity [80]. Recent experiments suggest that combinations of agents that inhibit CC5 or gp41-mediated fusion are protective against SHIV challenges [81].
Inhibition of reverse transcription. Antiretroviral drugs, such as nucleotide reverse-transcriptase inhibitors (NRTIs) and nonnucleoside reverse-transcriptase inhibitors (NNRTIs), are being evaluated as topical microbicides. Tenofovir (PMPA; 9-[2-(phosphonomethoxy)propyl]adenine), a nucleotide analogue, is being evaluated because it is easily activated and has a high barrier to resistance, compared with other reverse-transcriptase inhibitors [82]. Topical NRTIs have also been shown to protect against HIV challenge in vitro and in simian models [82]. A 14-day phase 1 trial of vaginal tenofovir gel in 84 women demonstrated high levels of positive safety, tolerability, and acceptability [83].
TMC-120, an NNRTI found to be safe in vitro and effective in mouse models, is also currently in development as a slow-release intravaginal ring that can be left in place for several weeks [84, 85]. TMC-120 has been found to be effective against NNRTI-resistant viruses, because 2 different viral mutations are needed for inactivation [86]. Cellular and tissue explant models demonstrate that UC-781, another NNRTI, is protective against HIV and may induce persistent HIV inactivation during subsequent exposures [87, 88]. A recent study suggests that tenofovir is synergistic with NNRTIs, setting the stage for possible combination antiretroviral microbicides that avoid the rapid development of resistance [89]. Antiretroviral formulations are also being studied in HIV-infected women, to evaluate the ability to decrease HIV vaginal shedding, the efficiency of transmission to male partners, and the development of resistance.
Recent developments have raised great hopes with regard to the possible use of microbicides to decrease the spread of STIs and HIV. Although 6 major clinical trials of microbicide efficacy will have begun by the end of 2005 (table 1), concerns have been noted with regard to the need for enhanced scientific coordination to analyze clinical results of trials of different compounds [90]. Evaluating the nuances of different compounds that have similar mechanisms of action will further our understanding of specific chemical structures and their toxic and therapeutic effects. Moreover, further studies may lead to the development of combination products that can maximize effectiveness, increase the spectrum of STI activity, and potentially reduce the drug concentrations necessary to prevent HIV transmission. In this section, we present unique preclinical and clinical challenges that researchers are overcoming in the design and development of microbicides.
Surrogate markers to evaluate safety. One of the major causes of delay in the evaluation of microbicides in clinical trials has been the lack of validated surrogate markers that correlate with clinical findings [91]. Outcomes from the clinical trials of N-9 underscore the need for rigorous safety evaluation; however, no reference standard exists that predicts ultimate safety in vitro, in animal, or in early clinical models. Currently, clinical safety testing utilizes pelvic examinations and coloposcopy to evaluate patients for signs of macroscopic and microscopic irritation, inflammation, and ulceration [92]. In an effort to standardize colposcopic assessments by multiple practitioners, study sponsors have developed common staff training, have shared multimedia resources for the description of epithelial lesions, and have emphasized receiving second opinions when clinical findings are noted [93].
Immunoassays are under investigation, to test whether genital cytokines could be used as biomarkers to assess mucosal inflammation in vitro and in animal models [94]. Proinflammatory cytokines may activate the recruitment of target cells to the cervicovaginal tract, enhancing macrophage and T cell susceptibility to HIV infection [95]. In HIV-infected individuals, genital cytokines may induce viral replication and increase genital HIV shedding [95]. The effects of microbicides on genital tract microflora and other innate antimicrobial peptides continue to be monitored [96]. For example, secretory leukocyte protease inhibitor, a serine protease inhibitor found in mucosal fluids, has been shown to inhibit HIV infection of macrophages [97]. Additional research in mucosal immunology could ultimately provide useful indicators for safety and efficacy.
Recently, animal and tissue explant models have been developed to mimic the physiological conditions of sexual HIV transmission. Progesterone-treated female rhesus macaques are used to demonstrate the biological efficacy of microbicides [79, 98], and cervical tissue explants prepared from women who have undergone hysterectomy allow for the study of antiviral activity in the setting of subepithelial migratory cells [88, 99]. Although microbicides currently in efficacy trials have not been tested in a uniform cell culture, tissue explant, or animal model, future clinical trials will likely benefit from a more systematic developmental pipeline that makes use of preclinical testing modalities [25, 98, 100].
Resistance in the genital tract. Because microbicides may need to protect against high titers of HIV under a variety of physiological conditions, combination agents and broad-spectrum products could demonstrate the greatest clinical efficacy by protecting the greatest number of targets. Although the rationale of combination therapy is to avoid the emergence of drug-resistant virus, the long-term, potentially inconsistent use of microbicides that target one part of the viral life cycle will need to be closely monitored. In HIV-infected individuals, exposure to subinhibitory monotherapy could lead to resistance to an entire class of drugs. For HIV-uninfected women, the use of fusion and reverse-transcriptase inhibitors will necessitate ongoing genotypic evaluation in order to detect the development of resistance during seroconversion if HIV is not effectively prevented. In particular, women exposed to entry inhibitors that block CCR5 will need to be monitored for the selection of CXCR4 virus [81].
Efficacy and trial design. Without an animal model or surrogate marker that reliably predicts microbicide safety and efficacy, it is necessary to conduct large-scale clinical trials that include thousands of HIV-uninfected, high-risk women who will be randomized to active or placebo formulation and followed for several years. In reality, because women and their partners may be unaware of their HIV status, HIV-infected women should be studied in early clinical trials. The need to establish standards of care and informed consent is challenging in resource-poor countries and communities where access to health care is extremely limited [101]. Recent HIV prevention trial controversies [102] underscore the need for greater international consensus on study design and long-term follow-up and for sensitivity in dealing with scientific, technical, and local political issues.
Condom migration. One of the persistent concerns of microbicide development is that offering access to a partially effective microbicide might offset successful condom promotion efforts. To address this concern, in 2003, a US Food and Drug Administration advisory committee recommended that efficacy trials should include an additional randomized study arm for patients to receive only condom counseling. This recommendation has led to debate about the effect of an unblinded control (i.e., condom-using participants) on study reliability, and the requirement for increased recruitment and resources [103-106]. Mathematical projections suggest that a low-efficacy microbicide used consistently would provide greater protection over time than a less consistently used, high-efficacy product, such as a condom [107].
Optimal product design. On the basis of contraceptive product research, a substantial amount of data has been collected that evaluates the acceptability and potential use of microbicides [108]. International studies on hypothetical, prototypical microbicides have focused on acceptability, particularly with regard to women's perceptions about product application, leakage, and cultural factors [109]. Women tend to demonstrate interest in a product that preserves sexual intimacy and can be used with or without their partner's knowledge [110].
Many formulation types and applicators have been developed for microbicide delivery. Gels, creams, suppositories, and vaginal rings are being explored as potential formulations, which are loaded in prefilled, single-dose, disposable applicators that are easy to store, use, and dispose [111]. Efforts to link formulation properties to biological functionality incorporate magnetic resonance imaging to predict the intravaginal spread of a potential microbicide [112]. Other formulation issues under study include the assessment of adequate dosage and volume, activity and toxicity, effect with frequent use, and compatibility with barrier methods [113]. Because these products are designed for women of child-bearing age and are used in contexts in which conception may be desired, the effect of microbicides on sperm and pregnancy must be considered.
Rectal microbicides. For many women and for men who have sex with men, unprotected anal intercourse continues to be a major and efficient route of HIV transmission [114-117]. The single columnar layer of rectal mucosa is easily damaged during the trauma encountered during anal sex, even when lubricants are used [118]. Several studies have shown that N-9 increases the disruption of epithelial cells and renders the rectal mucosa susceptible to HIV transmission [119-122]. Introduction of HIV through inflamed rectal mucosa is particularly efficient given the vast gut reservoir of CD4 and other immune cells that become targets of HIV entry and replication [123]. Moreover, intestinal epithelial M cells constantly sample luminal microbes and transfer them to intraepithelial lymphocytes and dendritic cells [123].
Currently, efforts are under way to design assays to test markers of inflammation in the rectal mucosa, such as the measurement of fecal calprotectin levels. The development of a colorectal tissue culture and the use of macaque models will allow for further evaluation of product efficacy and toxicity [124]. Data demonstrate that men who have sex with men frequently use lubricants during unprotected anal sex, therefore suggesting that lubricating rectal microbicides could be an acceptable prevention option [125].
Funding. Although many major pharmaceutical companies have not been involved in microbicide research, public-private sector initiatives have been initiated to fund research and clinical trials. Since 1997, investment by the National Institutes for Health and the United States Agency for International Development for the research and development of microbicides has more than tripled, to nearly US $90 million dollars in fiscal year 2004 [126]. Of the US $19.4 million dollars in philanthropic investment, the majority is provided by the Bill and Melinda Gates Foundation and the Rockefeller Foundation. Both of these organizations spearheaded financing of the International Partnership for Microbicides (, an organization that seeks to strategically facilitate the development, testing, and availability of several candidate products. Communication among microbicide researchers, advocates, and potential funders has been facilitated by the Alliance for Microbicide Development ( [127]. Without the support of large pharmaceutical industry budgets, microbicide researchers are limited in their ability to develop safety and efficacy studies and perform the large-scale clinical efficacy trials.
Microbicide research has been built on decades of advances in understanding the sexual transmission of HIV and, increasingly, has incorporated women's complex biological and social experiences into the design of this preventive technology. As concurrent clinical trials mature to demonstrate product safety, effectiveness, and acceptability in the next decade, the discovery and distribution of an effective microbicide may offer new options to transform the trajectory of the HIV pandemic.
  icon paper stack View Older Articles   Back to Top