Global efforts are underway to develop a vaginal microbicide that women can use to protect themselves against HIV and other sexually transmitted infections (STIs). Safety of candidate vaginal microbicides has been assessed with inspection of the genital epithelium by naked-eye examination or colposcopy. Standardization of the documentation of genital findings is challenging, however. In addition, the clinical relevance of many of these findings is not well understood, and interpretation of study results remains subjective. Markers of toxicity would be helpful to assess subclinical inflammation of the vaginal and cervical epithelium1 and might predict safety of candidate microbicides before large efficacy trials are started.
Vaginal inflammation, which is generally associated with clearing of bacterial vaginal infections, may enhance susceptibility to HIV infection by attracting potential HIV-1 host cells to the vaginal mucosa.2 Inflammatory processes at mucosal surfaces are usually downregulated by anti-inflammatory factors, such as secretory leukocyte protease inhibitor (SLPI), which are produced by the healthy mucosal epithelium.2 Proinflammatory cytokines, interleukin (IL)-1, IL-6, and IL-8 have been associated with mucosal irritation after vaginal product use in an animal model3 and have been recommended as markers of vaginal inflammation.4 Cytokines may identify subclinical inflammation of the genital epithelium; a study among healthy volunteers using N-9 showed increased proinflammatory cytokines and decreased SLPI levels.2 In addition, other factors such as age, hormonal contraception, sexual intercourse, and genital tract infection may influence vaginal cytokine levels.4
Carraguard, the Population Council's lead noncontraceptive candidate vaginal microbicide, is currently being evaluated in large efficacy trials in South Africa.5 Carraguard is a gel formulation made from a mixture of λ- and κ-carrageenan, derived from red seaweed. Carrageenans have been shown in in vitro tests to prevent HIV-1 infection at concentrations significantly lower than those associated with cellular toxicity, presumably by nonspecific binding to pathogens or host cells to inhibit fusion or entry.6
Previous safety trials of Carraguard and similar carrageenan-based formulations did not show more cervical or vaginal epithelial findings among women after use of Carraguard as compared with placebo gel.7,8 In this study, we compared cervicovaginal cytokine and SLPI levels after use of Carraguard and placebo gels and hypothesized that Carraguard gel would not cause vaginal inflammation as indicated by similar or decreased levels of proinflammatory cytokines compared with baseline or placebo gel.
The study population consisted of 55 HIV-seronegative couples who participated in a 6-month, randomized, triple-blind safety trial of Carraguard (3% carrageenan) and placebo (2.5% methyl-cellulose) gel in Chiang Rai, Thailand.9 Couples were eligible to participate in the study if they were in a monogamous relationship (reported no other sexual partner in the previous year and expressed no intent to have another partner during the study period), were HIV-negative and free of STIs at screening, were not regular condom users (ie, used condoms <25% of the time), were planning to have vaginal sexual intercourse at least once per week on average, and had intact genital epithelium on examination.
All couples provided informed consent for participation in this trial and were randomized to use Carraguard or placebo gel. Women were asked to use the gel each time they had sexual intercourse and to record intercourse and gel use in a coital diary. Women were screened for HIV (Genetic Systems HIV 1/2 Enzyme Immunoassay, Redmond, WA, with confirmation by NovaPath HIV-1 Immunoblot; BioRad Diagnostics Group, Hercules, CA) and STIs, including Chlamydia trachomatis and Neisseria gonorrhoeae (COBAS Amplicor PCR; Amplicor, Branchburg, NJ), Trichomonas vaginalis (InPouch culture test; Biomed Diagnostics, San Jose, CA), and Treponema pallidum (Macro-Vue rapid plasma regain card test; Becton-Dickinson, Franklin Lakes, NJ, with confirmation by a T. pallidum passive particle agglutination test; Fujirebio Diagnostics, Tokyo, Japan) and were not eligible if they tested positive for any of these infections. Women were tested for bacterial vaginosis (Nugent criteria) and vaginal yeast (microscopy) at screening and at the visit 1 month after enrollment (1-month follow-up).9 Colposcopy was performed at screening, enrollment, and 1-month follow-up visits if genital findings were seen by naked-eye examination. Genital findings were recorded in a standardized way and were categorized as having intact epithelium or superficial or deep (into stroma) epithelial disruption.10
Cervicovaginal lavage (CVL) specimens were collected at the enrollment visit (baseline) and at the 1-month follow-up visit after enrollment (1-month follow-up). CVL samples were obtained according to procedures described in the DAIDS Virology Manual for HIV Laboratories;11 5 mL of saline was used to flush the cervix and vaginal walls. A transfer pipette was used to recollect as much of this fluid as possible after 30 to 60 seconds; areas with visible gel were avoided because of concern about a possible effect of the gel on the cytokine assay. Within 30 minutes of collection, the fluid was centrifuged for 10 minutes at 1000 g. Separate supernatants and pellets were prepared from CVL samples, transported on dry ice to the laboratory, and stored at −70°C. Supernatants were tested for the following cytokines: IL-1β, IL-6, IL-8, and SLPI using a microwell plate-based enzyme immunoassay according to the manufacturer's instructions (Quantikine EIA; R&D Systems, Minneapolis, MN). Sensitivity and coefficient of variation were as follows: IL-1β (<1 pg/mL and 4.8%, respectively), IL-6 (<0.7 pg/mL and 2.7%, respectively), IL-8 (<10 pg/mL and 6.7%, respectively), and SLPI (<25 pg/mL and 6.2%, respectively).
Factors Associated With Cytokine Levels
Median age, use of hormonal contraception, number of vaginal sex acts, diagnosis of bacterial vaginosis, detection of yeast, and presence of genital findings were studied in relationship to IL-1β, IL-6, IL-8, and SLPI levels.
Median cytokine levels were assessed at 2 time points (baseline and 1-month follow-up visit) for Carraguard and placebo separately. Nonparametric tests were used for comparisons. The mean change in cytokine levels between the baseline and 1-month follow-up within study groups was tested using a paired t test (the change in levels between time points was normally distributed, allowing use of parametric statistical tests), and a difference in change between study groups was tested. A Wilcoxon rank sum test was used to compare median cytokine levels after Carraguard use with levels after placebo use. A random effects mixed model was fit to the log-transformed cytokine data to study cytokine levels from baseline to 1-month of follow-up in the combined study group. Factors thought to be associated with cytokine levels were dichotomized at the median levels, and a Wilcoxon rank sum test was used to test the association using data from both study groups and both visits.
Fifty-five women were enrolled in the study (28 in the Carraguard group and 27 in the placebo group), and 53 women completed the full study. All 55 women had a CVL collected at enrollment, and 52 women had a CVL collected at their 1-month follow-up visit. Three women missed their 1-month follow-up visit and were seen at an unscheduled visit (n = 2) or at the 2-month follow-up visit (n = 1), during which a CVL sample was collected. These 3 visits were combined with the 52 scheduled 1-month visits in the analysis. Study nurses noted a small amount of gel visible in 2 women (1 in each study group).
Median age among women was 32 years, and current use of hormonal contraception was reported by 33 (60.0%) of women. The median number of sex acts during the first month of the study was 5 (range: 2 to 14 sex acts), and reported adherence to instructions for gel use during each act of penile-vaginal sex was 95.9%; these were not significantly different for the 2 study groups. No STIs were diagnosed at follow-up visits during the study. At the 1-month follow-up visit, bacterial vaginosis was diagnosed in 7 (12.7%) women and yeast in 8 (14.5%) women; there was no significant difference between the 2 study groups. The number of women diagnosed with bacterial vaginosis or yeast was similar at screening and at the 1-month follow-up visit. Genital findings with intact epithelium at the 1-month follow-up visits, such as petechial hemorrhage and erythema, were found in 16 (29.1%) of the 55 women. No genital findings with disrupted epithelium were found.
We did not observe an increase in proinflammatory cytokines levels after use of Carraguard gel or placebo gel during the study (Table 1). Figure 1 shows a wide range of cytokine levels, with high levels among some women. SLPI levels tended to decrease from baseline to the 1-month follow-up visit among women using Carraguard (see Table 1). At the 1-month follow-up visit, median IL-6 levels were lower and median IL-8 levels tended to be lower after Carraguard use compared with placebo use, although levels of these cytokines were already somewhat lower in the Carraguard use group at baseline (see Table 1). Changes in median IL-6, IL-8, and SLPI levels between the baseline and 1-month follow-up visits were not significantly different between women who used Carraguard and those who used placebo gel. In the combined group of 55 women, IL-8 levels tended to decrease during the study (P = 0.06), which was similar in both groups (interaction term, P = 0.7); other cytokine levels did not increase or decrease during the study (all P values >0.1).
Factors Associated With Cytokine Levels
We did not find associations between cytokine levels and age, current use of hormonal contraception, number of vaginal sex acts in previous month, and presence of bacterial vaginosis (data not shown; all P values >0.1). Women with vaginal yeast present at the time of CVL collection tended to have somewhat higher IL-8 levels compared with women without yeast present (924.9 pg/mL vs. 337.0 pg/mL; P = 0.07). Women with genital findings with intact epithelium showed slightly higher IL-6 levels compared with those with normal epithelium, although this did not reach statistical significance (14.9 pg/mL vs. 8.8 pg/mL; P = 0.08).
This study provides additional safety data about the candidate vaginal microbicide, Carraguard. No increase in proinflammatory cytokines levels was observed after 1 month's use of Carraguard gel or placebo gel by 55 women, suggesting that neither gel causes inflammation. Our sample size might not have been sufficient to allow detection of significant changes in cytokine levels, however.
Comparisons of cytokine levels after gel use showed lower IL-6 and somewhat lower IL-8 levels after Carraguard use compared with placebo use. The decrease in IL-6 and IL-8 during the study was not significantly different between women using Carraguard and women using placebo gel, however. The lower baseline levels of proinflammatory cytokines among women randomized to the Carraguard group may have resulted from a difference in preexisting conditions of women between randomization groups. Although we did not test the CVL samples for the presence of semen, we assume that a potential effect of semen on cytokine detection4 would be similar across groups. Because semen contains IL-8, especially in men with prostatitis,12 semen contamination may be one of the possible explanations for the high IL-8 levels in some women in our study.
IL-8 levels tended to decrease during the study. The reason for this is not clear. Gels might have direct anti-inflammatory effects or indirectly affect levels by increasing lubrication during sexual intercourse or act as a physical barrier. Potential effects of Carraguard or placebo gels on the cytokine assays must also be considered, although care was taken to avoid collection of visible gel in our study. Consistent with our findings, others have shown decreased cytokine levels after use of candidate microbicides and after placebo.13 Decreased IL-8 levels were observed after use of cellulose sulfate,14 and decreased IL-1β, IL-8, and SLPI levels were observed after use of PRO 2000 gel (Indevus Pharmaceuticals, Lexington, MA).15 Although not done in our study, cytokine evaluation of a third group of study participants using no gel would have facilitated interpretation of study findings and evaluation of the inertness of the placebo. The potential effect of the gel on the cytokine assay and inclusion of a third group of women using no gel16 should be considered in evaluating cytokine levels in future microbicide safety trials. In addition, use of ultrasensitive assays to evaluate in vitro inhibition and detect additional biomarkers that remained low or undetectable in our study (including IL-10, IL-12, tumor necrosis factor-α, transforming growth factor-β, interferon-γ, macrophage inflammatory protein [MIP]-1α, MIP-1β, and soluble tumor necrosis factor receptor-1; data not shown) should also be considered in future studies.
In our study, median IL-6 levels tended to be higher among women with genital findings with intact epithelium compared with women without such findings, although this did not reach statistical significance in our limited sample size. Genital findings with intact epithelium may be noninflammatory and of uncertain clinical significance. None of the women had disrupted genital epithelium at the 1-month follow-up visit, so we were not able to study a potential association between cytokine levels and disrupted epithelium. Further research is needed to understand the association between cytokines and genital findings and to explore a possible role of cytokines in predicting epithelial safety of candidate vaginal microbicides.
In conclusion, neither gel seems to cause inflammation as indicated by changes in cytokine levels. This supports earlier reports showing no increase in cervical or vaginal epithelial findings with Carraguard use as compared with use of placebo.7-9
The authors thank Dr. Raina Fichorova and Barbara Friedland for reviewing this manuscript and for providing useful suggestions. Chonticha Kittinunvorakoon and Thanyanan Chaowanachan are acknowledged for their contributions to the data collection and analysis. The staff of the Chiang Rai Health Club (Thailand MOPH-US Centers for Disease Control and Prevention Collaboration, Chiang Rai HIV/STD Prevention Research Field Station) is acknowledged for their contribution, and the authors also thank the couples who participated in this study.
1. Hillier SL, Moench T, Shattock R, et al. In vitro and in vivo, the story of Nonoxynol 9. J Acquir Immune Defic Syndr
2. Fichorova RN, Tucker LD, Anderson DJ. The molecular basis of Nonoxynol-9-induced vaginal inflammation and its possible relevance to human immunodeficiency virus type 1 transmission. J Infect Dis
3. Fichorova RN, Bajpai M, Chandra N, et al. Interleukin (IL)-1, IL-6, and IL-8 predict mucosal toxicity of vaginal microbicidal contraceptives. Biol Reprod
4. Fichorova RN. Guiding the vaginal microbicide trials with biomarkers of inflammation. J Acquir Immune Defic Syndr
. 2004;37(Suppl 3):S184-S193.
5. Phase 3 Study of the Efficacy and Safety of the Microbicide Carraguard® in Preventing HIV Seroconversion in Women (Population Council Protocol 322). New York: Population Council; February 2002. Available at: http://www.clinicaltrials.gov/ct/show/NCT00213083?order=6
. Accessed June 26, 2006.
6. Pearce-Pratt R, Phillips DM. Sulphated polysaccharides inhibit lymphocyte-to-epithelial transmission of human immunodeficiency virus-1. Biol Reprod
7. Coggins C, Blanchard K, Alvarez F, et al. Preliminary safety and acceptability of a carrageenan gel for possible use as a vaginal microbicide. Sex Transm Infect
8. Kilmarx PH, van de Wijgert JHHM, Chaikummao S, et al. Safety and acceptability of the candidate microbicide Carraguard in Thai women: findings from a phase II clinical trial. J Acquir Immune Defic Syndr
9. Kilmarx PH, Blanchard K, Chaikummao S, et al. A randomized, placebo-controlled trial to assess the safety and acceptability of use of Carraguard vaginal gel by heterosexual couples in Thailand. Sex Transm Dis
. (in press).
10. Bollen LJM, Kilmarx PH, Wiwatwongwana P. Photo Atlas for Microbicide Evaluation
. Bangkok, Thailand: Thailand MOPH-US CDC Collaboration; 2002.
11. National Institutes of Health, National Institute of Allergy and Infectious Diseases, Division of AIDS. DAIDS Virology Manual for HIV Laboratories
. Publication NIH-97-3828. Washington, DC: US Department of Health and Human Services; 1997.
12. Khadra A, Fletcher P, Luzzi G, et al. Interleukin-8 levels in seminal plasma in chronic prostatitis/chronic pelvic pain syndrome and nonspecific urethritis. BJU Int
13. Keller MJ, Guzman E, Hazrati E, et al. PRO 2000 elicits a decline in genital tract immune mediators without compromising intrinsic antimicrobial activity. AIDS
14. Schwartz JL, Mauck C, Lai JL, et al. Fourteen-day safety and acceptability study of 6% cellulose sulfate gel: a randomized double-blind phase I safety study. Contraception
15. 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
16. Fleming RF, Richardson BA. Control groups in microbicide trials: in defense of orthodoxy. J Infect Dis
Keywords:© 2008 Lippincott Williams & Wilkins, Inc.
Carraguard; cytokines; safety; vaginal microbicide