icon-folder.gif   Conference Reports for NATAP  
 
  11th Annual Retrocirus Conference
(CROI-Conference on Retroviruses and Opportunistic Infections)
San Francisco
Feb 8-11, 2004
Back grey_arrow_rt.gif
 
 
 
HIV Microbocides at the 11th Retrovirus Conference
 
 
  The ideal microbicide will reliably prevent sexually transmitted diseases and HIV concomitantly, without inhibiting conception. It will be cheap to manufacture, prove simple to store and apply, and prove palatable to women and their sexual partners. A wide range of products, including detergents, charged compounds, acidic compounds, antibodies, antivirals, and protobiotics are in development. However, none of these compounds fulfills all of the criteria desired. In addition, in his lecture Dr. Shattock predicted that using an optimistic timeline, no product would be clinically available before 2010. In general, the funding stream for microbicide development has improved and enthusiasm is high, but development of a successful product will require overcoming a variety of major biological restrictions and very large, expensive clinical trials. The models described, however, should facilitate development of microbicides. Challenge of animals with HIV (SHIV) in concentrations found in human semen should help to provide a good idea of the capacity of candidate microbicide compounds.
 
Microbicides have become a "cause célèbre " in HIV prevention. This topic was emphasized in a passionate lecture in the Opening Ceremony by Stephen H. Lewis, the United Nations Special Envoy for HIV/AIDS in Africa, in the opening plenary talk Dr. Robin Shattock of St. George's Hospital Medical School in London, England, and in both a poster session (#121) and a microbicide symposium (Session 30). The microbicide symposium included 3 lectures.
 
Qu from St. George's Hospital Medical School used a variety of methods to block the entry and "spread" of HIV-1 strains into an ex-vivo cervical explant model. Release of p24 antigen into the supernatant, or detection of HIV DNA in the cervical tissue, was used to measure primary infection. This could be blocked through inhibition of CD4 alone, or CCR5 and CXCR4 together, suggesting infection of macrophages and/or lymphocytes. Detection of HIV in dendritic cells and lymphocytes in the culture supernatant was taken to represent spread of HIV in tissue. This could be blocked by inhibition of CD4 and mannose C-type lectin receptors (eg, DC-SIGN). Neutralizing antibodies (mAB b12) and sCD4 fusion protein CD4-IgG2 blocked both localized and disseminated infection. These results were used to relate the explant model to infection in vivo, with the assumption that this model can be used to test microbicides. The model suggests initial infection of macrophages with subsequent infection of dendritic cells, which in turn infect lymphocytes.
 
Ron Veazey from Tulane University in New Orleans, Louisiana, reported on the use of PSC-RANTES (one of a class of CCR5 receptor inhibitors) in a macaque challenge model. PSC-RANTES is a potent amino-terminus-modified RANTES that inhibits replication of HIV, with concomitant and durable down regulation of CCR5. Macaques were given medroxyprogesterone (Depo-Provera) to increase susceptibility to HIV infection through changes in the endocervix. Animals were challenged intravaginally with 300 TCID50 (about 1 million copies) of SHIV162P3, which led to infection of 4/5 animals. Five macaques treated 15 minutes before HIV exposure with the highest dose (1 mM) RANTES remained uninfected, compared with 4/5 animals treated with 330-mcM PSC-RANTES and 3/5 treated with the 100-mcM PSC. The investigators concluded that PSC-RANTES in high concentration is sufficient to block vaginal SHIV infection. In addition, clinical and tissue examination of the cervix failed to reveal evidence for inflammation after RANTES exposure.
 
Ron Otten of the US Centers for Disease Control and Prevention in Atlanta, Georgia, reported on the use of vaginal cellulose acetate phthalate (CAP) to prevent SHIV in a pig-tailed macaque model . Animals (n = 12) were repeatedly exposed over 16 weeks to 3 concentrations (1.2 x 105, 1.2 x 106, or 5.8 x 106) SHIV162P3. Macaques exposed to these concentrations became infected after 3, 4-8, and 12 exposures, respectively. The higher concentration of SHIV was used to test the efficacy of a 2 mL of a 13% preparation of CAP provided 15 minutes before SHIV exposure. Three of 4 macaques treated with CAP remained virus-free after 8 virus exposures, but 1 animal became infected after 3 exposures. The peak plasma viral RNA level in this macaque was approximately 2 log10 lower than in the control animals. At the concentrations used CAP caused no clinical inflammation.
 
Written by Myron Cohen, MD, University of North Carolina; medscape.com
 
Oral & Poster Abstracts from 11th Annual Retrovirus Conference
 
Reported by Jules Levin
 
In Vitro Microbicide Testing- poster abstracts
 
TMC120 Spreads Its Wings: Pre-treatment Blocks Infection of a Range of Cell-free Primary Isolates
 
H F Njai*1, K Andries2, M P de Béthune2, L Kestens1, P J Lewi3, P A J Janssen3, and G Vanham1 1Inst. of Tropical Med., Antwerp, Belgium; 2Tibotec, Mechelen, Belgium; and 3Janssen Pharmaceutica, Vosselaar, Belgium
 
Despite existing preventive measures (condoms) the spread of the HIV epidemic still continues to grow. A successful microbicide will prevent substantial new HIV infections and for it to be used world-wide candidate microbicides must be able to neutralise a wide range of circulating HIV-1 subtypes, especially those circulating the third world. A pre-treatment system previously developed was used to study the neutralising capacity of TMC120 (R147681, dapivirine) and UC781, on Ba-L (NSI/R5) virus and six primary isolates (subtypes A, A/G, B, C and CRF02 AG) mostly from seropositive African individuals.
 
An in vitro co-culture system of monocyte derived dendritic cells (MO-DC) and autologous CD4 T cells were used representing early in vivo targets of the HIV sexual transmission. Virus was first immobilised on a poly-l-lysine (PLL) treated 96-well plate and pre-treated with drug for 1 hour. Afterwards, the drug is washed away, target cells are added and cultured for 2 weeks. Viral production is measured on supernatant with HIV antigen ELISA. Negative results are confirmed by showing absence of proviral DNA in the cells.
 
TMC120 inhibited replication of Ba-L with an EC50 value of 49 nM, it had a more potent inhibition capacity on the primary isolates with an EC50 value ranging from 22 nM to <10 nM. Complete inhibition of replication was observed between 10 nM and 100 nM in all the virus strains. UC781 inhibited replication of the reference strain Ba-L with an EC50 value of 500 nM. EC50 values on the primary isolates were comparable to Ba-L and ranged from 34 nM to 588 nM. Complete inhibition of replication for all strains was observed between 1000 nM to 10,000 nM.
 
The outcome can be used to select compounds for studies in animal models of vaginal infection (SCID mice and macaques). A successful microbicide will give women, especially those in the Third World, more control over exposure and prevention to HIV.
 
Topical Estrogen Protects against SIV Vaginal Transmission without Evidence of Systemic Effect
 
S M Smith1, M Mefford2, Z Klase1, D Sodora3, N Alexander4, D Hess5, and P A Marx*2 1St. Michael's Med. Ctr., Newark, NJ, USA; 2Tulane Natl. Primate Res. Ctr., Covington, LA, USA; 3Univ. of Texas Southwestern Med. Ctr., Dallas, USA; 4Organon, Roseland, NJ, USA; and 5Oregon Natl. Primate Res. Ctr., Beaverton, USA
 
Accumulating evidence suggests that the state of the vaginal epithelium affects a woman"s risk of HIV vaginal transmission. Several human and non-human primate studies have established that the rate of HIV or SIV vaginal transmission is decreased when estrogen is dominant. Women treated with depo-medroxyprogesterone acetate (DMPA), an estrogen antagonist, have low levels of estrogen and are at increased risk of HIV vaginal transmission. Moreover, women, who become infected with HIV through the vaginal route while on DMPA, have more viral variants and higher viral loads than women not on a contraceptive agent.
 
In this study, we examined the protective efficacy and safety of vaginal estriol in ovariectomized macaques against SIV vaginal transmission. Vaginal estriol, in a cream form approved for human use, applied twice weekly resulted in minimal serum estriol levels and had no effect on serum LH levels, which decline rapidly in the presence of systemic estrogen and serve as the most sensitive marker for systemic estrogen effect. Vaginal epithelia cornified and thickened significantly in response to topical estriol therapy. Twelve macaques on vaginal estriol and 8 control animals on placebo cream were challenged vaginally with pathogenic SIVmac251.
 
One of 12 estriol treated became infected after this single challenge, while 6 of 8 control animals became infected (p = 0.0044). These data demonstrate that topical vaginal estriol can strongly protect against SIV vaginal transmission, while having no detectable systemic effect. The effect of estriol on the vaginal epithelia persisted for over 12 days after the last treatment.
 
These results support the study of topical vaginal estriol in preventing HIV vaginal transmission in at-risk women. Topical estriol therapy will have no systemic effect in women, who can use the treatment twice per week (or less) and remotely from sexual encounters. Further, topical estriol therapy may be used in conjunction with an effective microbicide. This therapy should be studied in high-risk women, especially those who are on DMPA, as a means to reduce the rate of HIV vaginal transmission.
 
Sulfated K5 Escherichia coli Polysaccharide Derivative: Potential Candidate Microbicide for Prevention of HIV-1 Infection
 
E Vicenzi*1, F Pacciarini1, L Piemonti1, C Fortis1, S Ghezzi1, G Zoppetti2, P Oreste2, G Poli1, and M Clementi1 1San Raffaele Sci. Inst., Milan, Italy and 2Glycores 2000 s.r.l., Milan, Italy
 
The ideal microbicide should fulfill a number of criteria including a broad activity against transmission of HIV and other sexually transmitted agents. We have previously described that a derivative of K5 polysaccharide with high degree of sulfation (K5-N,OS[H]) inhibits HIV-1 attachment and/or entry. Here we have investigated the potential inhibitory effects of K5-N,OS(H) on the replication of several primary HIV isolates. In addition we have tested the efficacy of K5-N,OS(H) in preventing dendritic cells (DC)-mediated infection of peripheral blood mononuclear cells (PBMC).
 
The K5 polysaccharide was obtained by fermentation of the E. coli strain Bi 8337/41. The N,O-sulfated K5 (K5-N,OS[H]) was obtained by chemically sulfation of K5 polysaccharide in N-position after N-deacetylation, and in O-position. Primary T-cell blasts from several healthy donors were infected with R5, R5X4, and X4 primary isolates, randomly selected within our repository of more than 500 primary isolates. Monocyte-derived DC were infected with either R5 or X4 HIV isolates in the presence or absence of K5-N,OS(H). Autologous PBMC were cocultivated with monocyte-derived DC incubated for 24 hours with HIV. The kinetics of viral replication were measured by supernatant-associated RT activity.
 
K5-N,OS(H) potently inhibited the replication of all (30 out of 30) tested primary HIV isolates including 16 R5, 4 X4 and 10 R5X4 viruses in T cell blasts. The 50% inhibitory concentration (IC50) was 2.1±1.6 µg/ml, without evidence of cytotoxicity even at the maximal concentration tested (100 µg/mL). K5-N,OS(H) (100 µg/mL) abolished R5 HIV replication in either DC cultivated alone or in DC-PBMC co-cultures. X4 HIV replication was only detected in DC-PBMC co-cultures and it was also suppressed by this concentration of K5-N,OS(H).
 
K5-N,OS(H) antiviral activity against several primary HIV-1 isolates with different coreceptor usage in both T-cell blasts, DC, and DC-mediated PBMC infection coupled with the lack of penetration into cells, render this molecule a potential candidate as a topical microbicide for preventing sexual HIV-1 transmission.
 
Mucosal Delivery of HIV-1 Microbicides Using Bio-engineered Lactobacilli
 
O Pusch*1, D Boden1, S Hanniffy2, M Boyd3, J Wells2, and B Ramratnam11Brown Univ. Med. Sch., Providence, RI, USA; 2IFR, Norwich, UK; and 3USA Cancer Ctr., AL, USA
 
Lactobacilli are commensual organisms found on mucosal surfaces and can be engineered to secrete therapeutic proteins thereby serving as a potential mucosal drug delivery vehicle. We wanted to bio-engineer lactobacilli to secrete HIV-1 microbicidal proteins such as the prototypic virucidal compound cyanovirin (CV-N). CV-N is an 11-kD protein that binds to high mannose oligosaccharides found on the envelope glycoproteins of HIV-1 and prevents virus entry at nanomolar concentrations by blocking viral fusion with the cell membrane.
 
To achieve secretion of CV-N, we inserted the structural gene for CV-N into the pKS131 series of theta-replicating plasmids. These constructs are based on a non-self-transmissible plasmid (pIL253) that carries a broad gram-positive host replicon, which has been adapted for constitutive expression of heterologous proteins in lactobacilli. pKS131-CV-N was transformed into L. lactis MG1363 and L. plantarum NCIMB 8826. Bacterial supernatant was sampled during log and stationary growth phases and protein secretion was verified by Western blotting. Antiviral activity of secreted, recombinant (r)CV-N was assessed by HIV-1NL4.3 challenge of CEMx174 cells carrying a tat responsive luciferase reporter gene (LuSIV). Alternatively, inhibition of HIV-1 infectivity was determined by real-time PCR of total intracellular HIV-1 DNA.
 
Western blot analysis revealed that engineered lactobacilli secreted 2 to 5 mg (r)CV-N per liter bacterial supernatant. We demonstrate that secreted (r)CV-N exhibits full biological activity in vitro, demonstrated by the fact that addition of 10 mL of filtered supernatant of (r)CV-N secreting lactobacilli to cells prior to HIV-1 challenge rendered the cells resistant to HIV-1 infection.
 
Lactobacilli can be engineered to secrete HIV-1 microbicides. Based on this approach, we hypothesize that it may be possible to effectively block HIV-1 infection at the vaginal/rectal mucosa by direct mucosal implantation of bio-engineered lactobacilli.
 
A Human Colorectal Explant Model for HIV-1 Infection and Microbicide Evaluation
 
P Watts*1, J Elliott2, I McGowan2, P Anton2, G Griffin1, and R Shattock11St. George's Hosp. Med. Sch., London, UK and 2Univ. of California, Los Angeles, USA
 
With the HIV-1 epidemic now entering its third decade, suitable strategies for prevention need to be developed. The potential of microbicides to reduce transmission of HIV across mucosal surfaces has been clearly identified. Work to develop microbicides for vaginal application is now well advanced, with some potential agents entering clinical trials. However, the prevalence of anal intercourse among heterosexual couples highlights the urgent need to assess the effect of these microbicides on HIV infection of rectal tissue. Such compounds would need to be tested within a suitable tissue model to ensure their efficacy in prevention of transmission across the colorectal mucosa. We present a model of colorectal explant culture to evaluate HIV infection and the efficacy of potential microbicides.
 
Investigations were completed to assess culture media requirements for the maintenance of tissue viability and subsequent viral replication. Viability measurements were completed using both histology and MTT dye reduction. Tissue was disrupted into single cells suspensions and assessments made of T-cell (CD4/CD8) ratios, and cell surface receptor expression (CCR5/CXCR4). Tissue explants exposed to HIV in the presence of candidate microbicides were assessed for viral replication by the presence of p24 in culture supernatants, and proviral DNA within proteinase-K digested tissue.
 
Investigations demonstrated no benefit from the addition of growth factors, including FCS, to culture media (DMEM). Tissue did benefit from being cultured at the air:media interface supported on gelfoam rafts. Histology of the tissue following 24 hours in culture demonstrated a significant reduction in tissue structure, but distribution of immune cells was unaffected. Additionally, there was no further deterioration of tissue structure following a further 6 days in culture. These culture conditions supported HIV-1 infection in the absence or presence of immune stimulation (PHA, 10 mg/mL). Preliminary experiments using compounds such as PRO2000 and UC781 demonstrate that microbicides can block HIV-1 infection of colorectal tissue.
 
A model of colorectal explant culture has been developed and is suitable for the pre-clinical assessment of potential rectal microbicides to prevent HIV-1 infection.
 
Effects of Pre- and Post-Trichomonas Therapy CVL Supernatants Applied to Cervical Explant Tissues on the Susceptibility to HIV-1 Infection
 
B K Patterson*1, G Spear2, A Landay2, R M Novak3, and for the Chicago WHIN1Stanford Univ. Sch. of Med., CA, USA; 2Rush Univ. Med. Coll., Chicago, IL, USA; and 3Univ. of Illinois at Chicago, USA
 
Epidemiologic studies on the aggressive treatment of sexually transmitted infections for the prevention of HIV have yielded confounding results. Here, we demonstrate in an in vitro model of tissue infection that cervical-vaginal lavage specimens (CVL) from women effectively treated for trichomonas infection decrease the susceptibility of cervical explant tissue relative to matched untreated CVL.
 
Pretreatment CVL from 8 women with documented trichomonas infection were collected prior to treatment with 2 g metronidazole. CVL were again collected 2 to 3 weeks post-therapy and assessed for trichomonas infection. Pre-and post-therapy CVL (300 mL) were applied to ectocervical, endocervical, and transition zone biopsies obtained from benign hysterectomy specimens. These 4-mm punch biopsies were cultured in a transwell apparatus that maintained the polarity of normal tissue. Tissues were pre-treated (1 hour) with undiluted CVL then infected with 600 TCID50 HIV-1Bal (R5) and harvested after 24 hours. Supernatants from the lower chamber of the transwell were added to cMAGI cells to quantify HIV-1 infectivity as a percentage of the positive control (media alone). The CVL were also assessed for the ability to stimulate production TNF-a in a Toll-like receptor 4 (TLR4) responsive HEN mouse spleen cell assay. Toll-like receptor 4 (TLR4) non-responsive HEN mouse spleen cells were used as a negative control.
 
Explant cultures from 7 of the 8 women demonstrated 10 to 80% decrease in susceptibility following treatment of trichomonas infection. The percent decrease in susceptibility in 7 of 8 and the percentage increase in susceptibility in 1 of 8 correlated with the change in production of TNF- a in the HEN mouse spleen cell assay (corr coef--0.75) but not in the HEN mouse spleen cell assay.
 
Trichomonas is the most common sexually transmitted infection in the United States. CVL from trichomonas infected women increase susceptibility of cervical explant cultures to HIV possibly through a mechanism involving the TLR4. Effective treatment of trichomonas using inexpensive therapy reduces susceptibility to HIV infection by 10 to 80%. Aggressive screening and treatment of Trichomonas-infected women in high HIV risk populations may be an effective and cost-effective HIV-1 prevention strategy.
 
ORAL ABSTRACTS
 
Blockade of Attachment and Fusion Receptors Inhibits HIV-1 Infection of Human Cervical Tissue
 
Q Hu*1, P Watts1, I Frank2, V Williams2, J Moore3, M Pope2, and R Shattock1 1St. George"s Hosp. Med. Sch., London, UK; 2Population Council, New York, NY, USA; and 3Weill Med. Coll. of Cornell Univ., New York, NY, USA
 
The majority of HIV-1 infection occurs via mucosal exposure, with heterosexual transmission as the predominant mode of infection worldwide. However, the early events of HIV-1 entry and transmission are still poorly understood. Identification of cellular factors governing sexual transmission of HIV-1 is critical not only for understanding viral pathogenesis, but also for development of effective microbicides.
 
To investigate the first line co-receptors in cervix, co-receptor-targeted inhibitors were used to block the entry of a panel of HIV-1 strains (R5, R5X4, and X4) in an ex vivo cervical explant model. To investigate whether the migratory cells from cervix are involved in dissemination of infectious HIV-1 from mucosal sites, human cervical tissues were infected with HIV-1 in the presence of CD4 antibody, co-receptor antagonists, mannose C-type lectin receptor antagonist, alone or in combination. Migratory cells from cervical tissues were collected and co-cultured with indicator cells.
 
Blockade of CD4 alone, or CCR5 and CXCR4 together, inhibited localized mucosal infection. However, simultaneous blockade of both CD4 and mannose C-type lectin receptors including, but not limited to, DC-SIGN, was required to inhibit HIV-1 uptake and dissemination by migratory cells. In contrast, the direct targeting of HIV-1 by neutralizing mAb b12 and sCD4 fusion protein CD4-IgG2 was sufficient to block both localized infection and viral dissemination pathways. The observations were further confirmed in moncyte-derived dendritic cells and DC-SIGN+ cell line.
 
The findings in this study provide the first demonstration that HIV-1 infects human cervix via CCR5 and CXCR4, and that both CD4 and MCLR on migratory cells may be involved in dissemination of infectious HIV-1 to other tissue sites. These studies have identified key targets for novel intervention strategies designed to prevent sexual transmission of HIV-1.
 
Intravaginal PSC-RANTES Protects against Vaginal Transmission of SHIV162P to Macaques; Implications for HIV Microbicide Strategies and Pathogenesis
 
R Veazey*1, R Offord2, O Hartley2, A Blauvelt3, J Dufour1, M Piatak Jr4, J Lifson4, D Mosier5, and M M Lederman6 1Tulane Natl. Primate Res. Ctr., Covington, LA, USA; 2Univ. of Geneva, Switzerland; 3Dermatology Branch, NCI, NIH, DHHS, Bethesda, MD, USA; 4AIDS Vaccine Prgm., SAIC Frederick Inc., NCI, MD, USA; 5Scripps Res. Inst., La Jolla, CA, USA; and 6Case Western Reserve Univ., Cleveland, OH, USA
 
Despite years of research, no cure or effective vaccine for HIV infection currently exists. This has led to the push for strategies that could prevent HIV infection, such as microbicides, that could be applied to the vagina and protect women from infection. However, compounds that damage or destroy the lipid envelope of HIV may also damage host cells, resulting in increased mucosal tissue damage and increase the rate of HIV-1 transmission. We are therefore developing microbicides that may block HIV-1 infection by binding to receptors on susceptible cells and interfere with HIV attachment or fusion on mucosal surfaces. Since HIV-1 utilizes 1 or more chemokine receptors for attachment, we are testing the potential for RANTES analogs to block HIV-1 transmission using the rhesus macaque model. RANTES is the natural ligand for the CCR5 receptor, which is utilized by most vaginally acquired strains of HIV. We have developed an amino-terminus-modified RANTES analog called PSC RANTES that is several orders of magnitude more potent than RANTES for inhibition of HIV replication in vitro, and also provides more durable downmodulation of CCR5 expression.
 
To assess the efficacy of PSC RANTES as a microbicide, macaques were given 30 mg Depo-provera, and 30 days later, 4 mL of PSC RANTES diluted in saline at concentrations of 1 mM (n = 5), 330 mM (i>n = 5), 100 mM (i>n = 5), or 1 to 10 mM (i>n = 5) were applied intravaginally; 5 received saline alone as controls. Animals were exposed intravaginally, 15 minutes after dosing, to 300 TCID50 of the CCR5-using SHIV162P3. Viral loads in plasma were monitored weekly by RT-PCR.
 
All 5 macaques pretreated with the highest dose (1 mM), 4/5 treated with the 330-mM dose and 3/5 treated with the 100-mM dose of PSC RANTES were completely protected against vaginal transmission, as evidenced by undetectable virus in plasma. In contrast, 7/10 macaques in the low-dose groups, and 4/5 in the control group became infected, indicating that protection from vaginal challenge was dose-dependent.
 
These results clearly demonstrate that topical PSC RANTES provides complete and dose-dependent protection against vaginal transmission of a CCR5-utilizing SHIV. Since CCR5-utilizing strains predominate in early mucosal transmission, these findings suggest that targeting CCR5 is sufficient to block vaginal SHIV infection, and that this may be an effective strategy for preventing mucosal HIV-1 transmission.
 
Cellulose Acetate Phthalate Protects Macaques from Multiple, Low-dose Vaginal Exposures with an SHIV Virus: New Strategy to Study HIV Pre-clinical Interventions in Non-human Primates
 
R A Otten*1, D R Adams1, C N Kim1, E Jackson1, J K Pullium2, K Lee1, L Grohskopf1, M Monsour1, N Jivani3, E Serbinova3, S T Butera1, and T M Folks1
1CDC, Atlanta, GA, USA; 2Emory Univ., Atlanta, GA, USA; and 3Dow Pharm. Sci., Petaluma, CA, USA
 
A nonhuman primate model for HIV-1 infection that more closely mimics sexual transmission by repeated low-dose exposure is critical to better understand and design intervention strategies using microbicides or vaccines.
 
Here we describe such an in vivo system using female pig-tailed macaques infected by multiple, low-dose SHIVSF162P3 inoculations given at weekly intervals via the intravaginal route. Macaques given 10 tissue culture infectious doses (TCID) were systemically infected after 3 exposures. Those given 2 TCID were infected after 4 to 8 exposures. No overt infection was observed using 0.2 TCID after 12 exposures. We have applied this low-dose exposure strategy using the 10-TCID dose to evaluate cellulose acetate phthalate (CAP, 13% formulation) as a topical microbicide for vaginal use.
 
CAP safety was demonstrated by colposcopy of the cervicovaginal region of macaques (n = 2) after applying 2 mL of product daily for 4 consecutive days. Efficacy was evaluated by applying 2 mL of CAP to the cervicovaginal region of macaques (n = 4) 15 minutes before each virus exposure. Systemic infection, defined by detectable virus in the peripheral blood leading to seroconversion, was observed in mock-treated control macaques (n = 3/3) following 3 virus exposures. Mean peak plasma vRNA levels reached ~106 copies/mL and vRNA was also detected in cervicovaginal lavage specimens (~ 105 copies per lavage). Interestingly, 3 of the 4 macaques (75%) treated with CAP have remained virus-free through 8 virus exposures to date and we intend to give a total of 12 exposures overall. Similar to the results with the controls, 1 of 4 CAP-treated macaques (25%) was infected after only 3 virus exposures. The peak plasma vRNA level in this breakthrough macaque was ~2-log10 lower than that of control macaques, suggesting a blunted peak of plasma viremia.
 
Our findings provide a basis to further refine nonhuman primate mucosal challenge models to mimic more closely natural human HIV exposures. This unique study provides data highly relevant to the design of preclinical trials to investigate the efficacy of therapeutic interventions, including microbicides and vaccines.