PATTON, DOROTHY L. PhD*; COSGROVE SWEENEY, YVONNE T.*; RABE, LORNA K.†; HILLIER, SHARON L. PhD†
A VARIETY OF TOPICALLY APPLIED microbicidal agents (topical microbicides) are currently in preclinical testing for use in the prevention of sexually transmitted infections (STIs). A number of developed products, many of which contain nonoxynol-9, have been enrolled in clinical trials throughout the world. Nonoxynol-9 is approved for use in the United States as a vaginal contraceptive. Early efforts to develop a topical microbicide product focused on nonoxynol-9 as an active agent because it is readily available and has a long history of intravaginal use.
Numerous in vitro studies were designed to assess microbicidal activity of nonoxynol-9. Although many of these studies resulted in conflicting reports about anti-STI activity, a number of investigators reported cytotoxic effects on host cells. 1–3 Clinical safety studies of nonoxynol-9 products used vaginally in women have also yielded a wide spectrum of results. Several clinical studies have shown increased genital epithelial disruption and inflammation in the cervicovaginal environment in women using nonoxynol-9. 4–7 A recently reported yet controversial study of female sex workers who used a vaginally applied nonoxynol-9 gel up to 20 times/day showed an increased risk of HIV acquisition in comparison with the control arm. 8
Although there is accumulating evidence of nonoxynol-9-induced irritation to vaginal mucosal cells, the effects of nonoxynol-9-containing products on the rectal environment have not been clearly evaluated. Because 41% of men who have sex with men reportedly use nonoxynol-9-containing products during rectal intercourse, 9 it is imperative that all products being developed for topical microbicide use be assessed for safety in both the vaginal and rectal mucosal environment.
Preclinical safety evaluations—that is, studies conducted before human use—of topical microbicide product candidates should be standard practice. Few animal models are designed specifically to assess microbicide products for rectal use. In a mouse model for herpes infection, in which nonoxynol-9-containing products were shown to increase the likelihood of rectal infection with herpes simplex virus type 2, widespread sloughing of the rectal epithelium after nonoxynol-9 exposure was noted. 10,11
These results prompted the investigators to conduct a small study assessing the effects of rectally applied nonoxynol-9-containing products in humans (n = 4). They examined rectal lavage specimens collected before product use and at 15 minutes and 8 to 10 hours after rectal application of commercially available nonoxynol-9-containing products. Gross and microscopic observations included exfoliation of large sheets of rectal epithelium 15 minutes after exposure to the 2% nonoxynol-9 product. 12 Less dramatic results were noted after the use of products containing lower concentrations of nonoxynol-9, and all lavage samples appeared normal 8 to 10 hours after product use. A different study of 35 male couples showed that there was no evidence of mucosal damage in biopsy specimens collected after 5 to 6 weeks of daily rectal use of a 3.5% nonoxynol-9 product. 13
In order to establish the predictive value of our recently established monkey model for rectally applied topical microbicides, we initiated studies involving the use of Conceptrol (Advanced Care Products, Skillman, NJ), a commercially available product that is used extensively in this laboratory for vaginal safety and efficacy studies.
We recently reported that rectal flora and epithelial tissues of the pig-tailed macaque are similar to those in humans. In addition, we showed that the rectum of the pig-tailed macaque is naturally susceptible to human rectal isolates of Chlamydia trachomatis. 14 We have documented the usefulness of this animal model for evaluating topical microbicide products for vaginal and rectal use. 14–18 In this study, the effects of repeated rectal applications of Conceptrol (containing 4% nonoxynol-9) on the microflora and epithelium in the rectal macaque model were assessed.
Materials and Methods
Research colony–raised, sexually mature female pig-tailed monkeys (Macaca nemestrina) were used in these experiments. All monkeys were housed at the University of Washington National Primate Research Center. Prior approval for use of monkeys in this protocol was obtained from the Institutional Animal Care and Use Committee at the University of Washington. Animals were handled humanely, and experiments were carried out according to the Animal Use Guidelines of the National Institutes of Health.
The effects of multiple applications of rectally applied Conceptrol (containing 4% nonoxynol-9), placebo gel, or no product were assessed on rectal flora and mucosal tissues in the pig-tailed macaque model. Animals were randomly selected for inclusion in 1 of 3 groups (8 monkeys per group): Conceptrol (test product), placebo gel (to control for active agent, nonoxynol-9), or no product (to control for topical product application). Each group received 3 daily intrarectal applications of product, at 24-hour intervals.
At each visit, a preapplication rectal pH swab, a swab specimen for rectal microbiology, and a rectal lavage specimen were collected. The pH was determined by rolling a rectal mucosal swab onto a pH indicator stick with resolution of 0.5 pH units. A second rectal swab was collected and immersed in a transport tube (Port-a-Cul, Becton Dickinson Microbiology Systems, Cockeysville, MD) for eventual microbiologic characterization. The rectal lavage was performed with a 10-ml syringe attached to an infant feeding tube. The tube was carefully inserted 1.5 inches into the rectum, and 8 ml of saline wash was expelled. Twenty seconds later, the saline wash was recovered and deposited into a vial. These sample collections were followed by a 2.5-ml application of nonoxynol-9-containing product or placebo gel, past the anal sphincter into the rectum. Fifteen minutes later, samples were again collected. Animals assigned to the no-product group underwent sample collection again 15 minutes after completion of the preapplication sample collections. On the fourth day, a final rectal pH swab, microbiology sample, and rectal lavage specimen were collected from all animals.
Microbiologic Characterization of Rectal Microflora
Rectal swabs were collected at baseline and 15 minutes after product application, daily for 3 days. On the fourth day, a follow-up rectal swab specimen was collected. Swabs were individually placed in Port-A-Cul tubes, designed for transport of anaerobic, facultative, and aerobic specimens on swabs. This transport device has been shown to preserve viability of aerobic and anaerobic bacteria in clinical specimens. 19 Daily samples were packaged in accordance with biohazardous material shipping regulations and sent via express mail. Within 24 hours of collection, samples were delivered to the microbiology laboratory at Magee–Womens Research Institute (Pittsburgh, PA).
Swab specimens from all animals were evaluated for aerobic and anaerobic microorganisms with microbiologic assays previously described. 15 Semiquantitative analyses were performed for each organism detected, and the findings were graded on a scale of 0 to 4 (0 = not detected, 4 = growth on all quadrants of a plate). Species belonging to the genera Bacteroides, Porphyromonas, and Prevotella were grouped together as either anaerobic gram-negative rods (nonpigmented) or black gram-negative rods for ease of presentation.
Rectal washes were examined under a dissecting microscope at 7× magnification. Contents of the lavage were evaluated for the presence of individual cells, sheets of epithelium, and gross blood. After gross examination, lavage samples were processed for examination by routine light microscopy.
The frequencies of groups of microorganisms were assessed at each sampling point to determine trends. Generalized estimating equations were used to compare the changes over the sampling periods in the prevalence of H2O2-producing viridans streptococci, H2O2-producing lactobacilli, and black-pigmented anaerobic gram-negative rods between the Conceptrol group and the placebo group, with adjustment for baseline colonization status. The model was developed with use of Stata statistical software release 6.0 (Stata Corp., College Station, TX), and an unstructured working correlation matrix was specified. Statistical inference was based on the Wald chi-square test statistic, evaluated at the 0.05 significance level.
The range of normal pH in the macaque rectum is 5.5 to 8.0. The pH of the Conceptrol (nonoxynol-9-containing product) and of the placebo gel was 5.0. Among the animals receiving no product, the average pH from baseline to day 4 follow-up was 6.7 to 6.9; for the placebo gel animals the averages were 6.3 at baseline and 6.6 at follow-up. The Conceptrol group averaged 6.0 at baseline and 6.8 at follow up. All groups tended to have slightly higher rectal pH values at visits 15 minutes after application, in comparison with their preapplication values. The nonoxynol-9 product did not cause fluctuations in pH that were different from those in either control group.
Microbiologic assessments did not reveal significant shifts in the rectal flora after repeated no-product or placebo gel applications. However, a decrease in detection of black-pigmented anaerobic gram-negative rods was noted after repeated Conceptrol application (P = 0.004). Animals that received the nonoxynol-9 product were also more likely to have a decrease in H2O2-producing microorganisms than were animals that received no product or placebo gel. The presence of H2O2-producing lactobacilli remained far more constant than that of H2O2-producing viridans streptococci, which diminished in all animals in which these organisms were detected at baseline (P = 0.01;Figure 1).
Gross observation of rectal lavage samples indicated an increase in the presence of epithelial cells and sheets of epithelium 15 minutes after nonoxynol-9 application, as compared with preapplication samples and as compared with the placebo gel and no-product groups. By the end of the experiment, 5 of 8 animals receiving Conceptrol, versus 2 of 8 receiving placebo gel and 3 of 8 receiving no product, had evidence of blood detected in one or more rectal lavage specimens. The rectal mucosal tissues were somewhat delicate, prone to slight irritation as a result of repeated lavage, even in the absence of topical product.
On day 1, at baseline, only a scanty number of free cells appeared in samples of all products evaluated. Sheets of intact epithelial cells were observed, although rarely, and were as large as 40 mm2. By contrast, 15 minutes after nonoxynol-9 exposure, large pieces of intact epithelium were frequently observed.
On day 2, the preapplication samples looked similar to those at baseline. Again, free-floating cells and a few epithelial sheets as large as 32 mm2 could be seen. Evidence of sheets of intact epithelium was not significantly increased in samples from day 2, time 0, as compared with day 1 baseline observations. The sample collected 15 minutes after the second daily application of nonoxynol-9 contained numerous red blood cells and polymorphonuclear cells, as well as large intact epithelial sheets. At this visit, 6 of 8 animals had measurable epithelial sheets, some as large as 60.5 mm2.
On day 3, some preapplication washes in the nonoxynol-9 group still contained epithelial sheets (as large as 150 mm2), a trend noted with successive exposures to nonoxynol-9. Fifteen minutes later, the frequency of noted sheets of intact epithelium (up to 80 mm2) roughly doubled. Tinges of blood were noted in the rectal lavage samples from three nonoxynol-9-exposed animals at this collection. Follow-up lavage specimens on day 4 continued to show large sheets of sloughed epithelium (as large as 60.5 mm2) in the nonoxynol-9 group only.
Clearly, the evidence of sloughed epithelium observed 15 minutes after product application denotes a potential threat to the integrity of the mucosal tissues at the time of pathogen introduction. The continued presence of sloughed epithelial sheets 24 hours after repeated product application indicates a cumulative effect of nonoxynol-9 on rectal tissues. These results may indicate a higher risk of STI acquisition to frequent nonoxynol-9 users.
Histopathology showed an increase in epithelial sheets with stroma attached in rectal lavage specimens obtained on successive days after Conceptrol use. By the fourth day (24 hours after the third daily application), epithelial sheets were still observed in rectal lavage samples (Figure 2). Recurrent observations of large sloughed epithelial sheets (>20 mm2 in area), epithelium with stromal cells attached, and red blood cells were noted only after application of nonoxynol-9-containing product, as opposed to placebo gel or no-product applications.
In order to protect themselves during each sexual encounter, men and women will use repeated applications of a topical microbicide product. It is desirable that such a product be effective in preventing HIV and STI acquisition while not irritating the mucosal surface or adversely affecting normal flora of the vagina or rectum. For this reason, it is imperative that studies of multiple perturbations be performed on topical microbicide candidates.
Although prevention of vaginally acquired STIs has been a primary focus during the development of topical microbicides, it is important to note that anorectal transmission has been increasingly recognized as a source of infection among men and women. 20 An optimal topical microbicide product should be safe and effective when used vaginally and/or rectally. Persons who engage in anal intercourse commonly report use of lubricants, and a large proportion of gay men having anal sex actively seek sexual lubricants that contain nonoxynol-9. 9 This suggests that products being evaluated as microbicides for vaginal use will likely be used during anal sex. The current study suggests that products widely considered safe for vaginal use may not be safe for rectal use and that rectal safety studies should be performed for all microbicides being developed for vaginal use. Animal models will be useful for performance of safety studies and preliminary proof-of-principle efficacy evaluations.
In this study, after repeated applications of Conceptrol (nonoxynol-9), placebo gel, or no product, little if any change in the frequency of lactobacilli was detected. Specifically, the presence of H2O2, which is largely produced by lactobacilli in humans, is believed to be a protective factor in the vaginal milieu. However, it has been shown that viridans streptococci are the predominant H2O2-producing species in the vagina of the pig-tailed macaque. 15 In this study, a statistically significant decrease in the H2O2-producing viridans streptococci was detected after repeated nonoxynol-9 exposures, in comparison with their presence after application of placebo. The clinical significance of this observation is unknown.
In this study multiple applications of a 4% nonoxynol-9 formulation led to exfoliation of rectal epithelial tissues as early as 15 minutes after exposure. One day after the first exposure (day 2, time 0), sloughing was no longer evident, suggesting that repair occurred rapidly. Increased amounts of sloughed epithelial tissues were detected after repeated applications, suggesting that the damage was cumulative after multiple exposures to nonoxynol-9.
Our findings corroborate the observations of Phillips et al, 12 who reported extensive exfoliation of rectal epithelium, as assessed by transmission electron microscopy, in four men exposed rectally to nonoxynol-9. Such tissue irritation may heighten the acquisition and spread of STIs. No such tissue exfoliation was detected in the animals assigned to the placebo gel and no-product groups.
By contrast, a phase I study of nonoxynol-9 safety and toxicity found that low-dose (3.5%) nonoxynol-9 gel was not associated with macroscopic rectal epithelial disruption or inflammation. 13 In the phase I clinical trial, rectal biopsies were collected for interpretation of safety on the Monday after 5 to 6 weeks of daily product use (personal communication), as opposed to the protocol in the current study, wherein rectal lavage samples were collected 15 minutes after product use.
Our observations in the primate model confirm the mouse model and human data suggesting that rectal exposure to nonoxynol-9 causes sheets of epithelium to be sloughed from the rectal mucosa, a potential site for STI transmission, 15 minutes after product use. The current study extends previous findings by demonstrating that the rectal epithelial disruption resolves within 24 hours of a single nonoxynol-9 exposure, which may suggest that human safety studies of rectal microbicide use should involve assessment of epithelial damage within minutes rather than hours after exposure. Furthermore, the findings in the current study suggest that multiple rectal exposures to nonoxynol-9 result in greater disruption and apparent inability of lesions to resolve within 24 hours, which may imply that repeated applications inhibit repair.
These alarming observations emphasize the importance of conducting safety studies of potential microbicide products in the animal model. In addition, these results underscore the importance of characterizing the effects of topical microbicides on the mucosal environment at the time of potential infection, i.e., shortly after product application. Finally, it is clear that the effects of repeated applications of topical microbicide products must be evaluated before human use is approved or recommended.
1. Patton DL, Wang SK, Kuo CC. In vitro activity of nonoxynol 9 on HeLa 229 cells and primary monkey cervical epithelial cells infected with Chlamydia trachomatis. Antimicrob Agents Chemother 1992; 36: 1478–1482.
2. Knight ST, Lee SH, Davis CH, Moorman DR, Hodinka RL, Wyrick PB. In vitro activity of nonoxynol-9 on McCoy cells infected with Chlamydia trachomatis. Sex Transm Dis 1987; 14: 165–173.
3. Kappus EW, Quinn TC. The spermicide nonoxynol-9 does not inhibit Chlamydia trachomatis in vitro. Sex Transm Dis 1986; 13: 134–137.
4. Roddy RE, Cordero M, Cordero C, Fortney JA. A dosing study of nonoxynol-9 and genital irritation. Int J STD AIDS 1993; 4: 165–170.
5. Stafford MK, Ward H, Flanagan A, et al. Safety study of nonoxynol-9 as a vaginal microbicide: evidence of adverse effects. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 17: 327–331.
6. Martin HL Jr, Stevens CE, Richardson BA, et al. Safety of a nonoxynol-9 vaginal gel in Kenyan prostitutes: a randomized clinical trial. Sex Transm Dis 1997; 24: 279–283.
7. Rustomjee R, Abdool Karim Q, Abdool Karim SS, Laga M, Stein Z. Phase I trial of nonoxynol-9 film among sex workers in South Africa. AIDS 1999; 13: 1511–1515.
8. Van Damme L, Chandeying V, Ramjee G, et al. Safety of multiple daily applications of COL-1492, a nonoxynol-9 vaginal gel, among female sex workers. COL-1492 Phase II Study Group. AIDS 2000; 14: 85–88.
9. Gross M, Buchbinder SP, Celum C, Heagerty P, Seage GR 3rd. Rectal microbicides for U.S. gay men: are clinical trials needed? Are they feasible? HIVNET Vaccine Preparedness Study Protocol Team. Sex Transm Dis 1998; 25:296–302.
10. Phillips DM. Intravaginal formulations to prevent HIV infection. In: Fantini J, Sabatier JM, eds. Perspectives in Drug Discovery and Design. Dordrecht, The Netherlands: Kluwer Academic Publishers, 1996: 213–223.
11. Phillips DM, Zacharopoulos VR. Nonoxynol-9 enhances rectal infection by herpes simplex virus in mice. Contraception 1998; 57: 341–348.
12. Phillips DM, Taylor CL, Zacharopoulos VR, Maguire RA. Nonoxynol-9 causes rapid exfoliation of sheets of rectal epithelium. Contraception 2000; 62: 149–154.
13. Tabet SR, Surawicz C, Horton S, et al. Safety and toxicity of nonoxynol-9 gel as a rectal microbicide. Sex Transm Dis 1999; 26: 564–571.
14. Patton DL, Cosgrove Sweeney YT, Rabe LK, Hillier SL. The pig-tailed macaque rectal model: microflora and chlamydial infection. Sex Transm Dis 2001; 28: 363–366.
15. Patton DL, Cosgrove Sweeney YT, Rabe LK, Hillier SL. The vaginal microflora of pig-tailed macaques and the effects of chlorhexidine and benzalkonium on this ecosystem. Sex Transm Dis 1996; 23: 489–493.
16. Patton DL, Ganzle GM, Cosgrove Sweeney YT, Clark AM, Hillier SL. Effects of nonoxynol-9 on vaginal microflora and chlamydial infection in a monkey model. Sex Transm Dis 1996; 23: 461–464.
17. Patton DL, Cosgrove Sweeney YT, McKay TL, Clark AM, Rabe LK, Hillier SL. 0.25% Chlorhexidine gluconate gel: a protective topical microbicide. Sex Transm Dis 1998; 25: 421–424.
18. Patton DL, Kidder GG, Cosgrove Sweeney YT, Rabe LK, Hillier SL. Effects of multiple applications of benzalkonium chloride and nonoxynol-9 on the vaginal epithelium in the pigtailed macaque. Am J Obstet Gynecol 1999; 181: 1080–1087.
19. Baron EJ, Strong CA, McTeague M, Vaisanen M-L, Finegold SM. Survival of anaerobes in original specimens transported by overnight mail services. Clin Infect Dis 1995; 29: 5174–5177.
20. Wexner SD. Sexually transmitted diseases of the colon, rectum and anus: the challenge of the nineties. Dis Colon Rectum 1990; 33: 1048–1062.