In the context of vaginal microbicide development there are currently four principal ways of assessing inflammation: naked eye inspection, colposcopy, pro-inflammatory cyto-kine expression and biopsy.
The technique of vaginal biopsy has been used to provide an objective assessment of inflammation in Phase I studies conducted at Imperial College, London, since 1995. The technique involves the use of Eppendorfer biopsy forceps, which reliably provide biopsy specimens of a length of 3–5 mm, and a depth of 1–2 mm of lamina propria. The biopsy site is standardized so that at baseline the biopsy is taken from the cervical aspect of the right fornix and at follow-up from the cervical aspect of the left fornix. The lateral fornices were chosen as the most appropriate biopsy sites as they were deemed likely to have maximum exposure to the gel.1 Baseline and follow-up biopsies are standardized for the phase of the menstrual cycle, being taken approximately one month apart. In all cases the study participants are counselled before the procedure, and in earlier studies1–3 xylocaine spray (100 mg/ml) was used as a topical local anaesthetic. When necessary, the biopsy site is cauterized with a silver nitrate stick. For all participants, healing of the biopsy site 2 weeks after the baseline biopsy is confirmed at colposcopy before the administration of gel. In the earlier studies1–3 the biopsies were preserved in formal saline, whereas in the more recent studies4 they were frozen at −70°C. The specimens are stained with haematoxylin and eosin, and when inflammation is present a score of 1 (mild inflammation) to 3 (severe inflammation) is given by the pathologist. In addition, the inflammatory response is characterized by immunohistochemical staining for CD4 cells, CD8 cells, B cells and macrophages. The evaluations are carried out by one histopathologist who is blinded as to the allocation group and to the timing of the biopsy in relation to gel use.
The vaginal microbicide programme at Imperial College, London, was set up in 1995, funded jointly by the UK Medical Research Council and by industry. Initial clinical trials were implemented with a view to taking forward the clinical development of the sulphated polymers dextrin sulphate (DS) and PRO-2000, both of which had shown promising results in laboratory studies5,6. Both are large molecular weight polymers that inhibit the entry of HIV into cells, are highly selective for HIV and other enveloped viruses, and exhibit a high therapeutic index (activity over toxicity) of greater than 100 000. This contrasts with nonoxynol-9, the only agent with anti-HIV activity licensed for vaginal use, albeit as a spermicide rather than a microbicide7. Nonoxynol-9 is a surfactant with non-specific activity against HIV and a low therapeutic index of 1. It seemed even at that time that nonoxynol-9 was likely to be toxic to the cells lining the vagina and cervix at concentrations that were therapeutically active. We felt it important that novel agents such as DS and PRO-2000 should be developed as vaginal microbicides only if they were shown in clinical trials to be safer than nonoxynol-9. It was therefore necessary to determine accurately and quantitatively the toxicity of nonoxynol-9 in human volunteers.
This reasoning led to a double-blind Phase I clinical trial of nonoxynol-9, given in gel form at a concentration of 20 mg/ml (100mg/dose) in healthy sexually abstinent women. The placebo consisted of the gel vehicle alone. Volunteers were aged between 18 and 45 years and were HIV negative. We dosed for 7 days with nonoxynol-9 administered twice a day and observed toxicity closely by symptoms, by visual colposcopic examination of the vagina and cervix, and by interaction with normal vaginal flora. Vaginal biopsy pre and post 7-day dosing was used to provide objective evidence of inflammation. The trial reported by Stafford et al.2 showed that although there was no observed disruption of the genital epithelium after nonoxynol-9 use, objective biopsy confirmed evidence of vaginal inflammation in 35% of women (7/20) receiving nonoxynol-9, in whom colposcopy and symptomatology were unremarkable. The biopsies revealed that the inflammatory response occurred in the superficial layers of the lamina propria and included CD8 T cells and CD4 macrophages. The increase in the number of these cells, thought to be the first cell type infected by HIV-1 in natural mucosal transmission, raised the possibility that nonoxynol-9 might not only be virucidal, but might also enhance infectibility through inflammation in the intact vagina. These findings are in keeping with subsequent work by Fichorova et al.,8 who characterized the molecular basis of nonoxynol-9-induced inflammation, and Van Damme et al.,9 who established through a large multi-centre randomized controlled trial of sex workers, that the nonoxynol-9 formulation COL-1492 was not protective against the transmission of HIV-1 and indeed could enhance transmission risk.
In our study of nonoxynol-9, a further analysis of the results revealed that the inflammatory response to nonoxynol-9 found on biopsy was frequently sub-clinical: colposcopy findings and symptomatology correlated poorly with histological findings. Taking histology as the most accurate measure of genital inflammation, symptomatology had a sensitivity of 33% and a specificity of 61%, whereas colposcopy had a sensitivity of 56% and specificity of 81%.2 For this reason we set as our ‘gold standard’ for future initial Phase I studies of novel agents the occurrence of biopsy-confirmed inflammation, in order to exclude any new microbicide that was as or more inflammatory than nonoxynol-9.
Through the use of a common clinical trial protocol design we then studied increasing concentrations of DS (0.0003% and 0.001% DS) and PRO-2000 (0.5% and 4% PRO-2000) in healthy sexually abstinent female volunteers. All studies were randomized and double-blinded, and the placebo gel consisted of the gel vehicle without the active agent. For the DS study we dosed for 7 days with vaginal biopsies taken before and after gel exposure. DS was not associated with any inflammatory response in the vaginal biopsies post dosing.1 The PRO-2000 study was undertaken at the Institute of Tropical Medicine in Antwerp and at Imperial College, London, by Van Damme et al.3 The study showed that PRO-2000 like DS was well tolerated and was not associated with the induction of vaginal inflammation on biopsy. From these early Phase I studies,1–3 we concluded that the novel HIV entry-inhibitors DS (doses up to 0.001%) and PRO-2000 (doses up to 4%) were likely to be more tolerable and less toxic than nonoxynol-9.
We then performed a study of 0.125% DS in sexually active HIV negative women. Although biopsies were not used in this study, laboratory data and colposcopy suggested that DS at this concentration was well tolerated. We subsequently proceeded with a further Phase I study of higher concentrations of DS in sexually active women and in HIV-positive women.4 This was conducted at the Institute of Tropical Medicine, Antwerp and at Imperial College, London. The study was in two parts. In part A 50 sexually active HIV-negative women were randomly assigned (double-blind) to one of three study groups: 1% DS, 4% DS gel or placebo (gel vehicle alone). Analysis of part A determined the concentration of DS to be used in part B, in which the highest tolerated concentration selected in part A was further evaluated in HIV-negative and HIV-positive women. Part B involved 30 women (20 HIV positive and 10 HIV negative) who received the same concentration of DS gel and an additional 20 HIV-negative women who were included as a no-intervention observation group. In both parts, dosing was twice a day for 14 days, and as in the previous trials we observed toxicity closely by symptoms, by visual colposcopic examination of the vagina and cervix, and by interaction with normal vaginal flora. Vaginal biopsies were taken (from UK participants only) to provide objective evidence of inflammation: pre and post 14 days of dosing. All participants were counselled that vaginal spotting may occur for up to 24 h and that intercourse should be avoided if spotting occurred. In addition, sexually active HIV-positive women were counselled to avoid sexual intercourse for one week after their biopsy because of the risk of HIV transmission to an HIV-negative partner. For all women, healing of the pre-gel biopsy site was confirmed before gel administration. We found that the gel was well tolerated by all women with no genital epithelial disruption on colposcopy, although intermenstrual bleeding was reported frequently by women using gel, including those using placebo.
In this study, the biopsy procedure was found to be safe and well tolerated by both sexually active HIV-negative women and by HIV-positive women. Analysis of paired biopsy data (n = 42) revealed inflammation (mild to moderate) in 12 women at baseline and in 8 at follow-up. Of the 4 women who had inflammation only on their follow-up biopsies, two were in the 4% DS group and two were in the no-intervention observation group. Once again, we found that there was a poor correlation between symptomatology, colposcopic findings and histology. We concluded from this study that DS gel at concentrations up to 4% was safe and well tolerated in sexually active HIV-negative women and in a small cohort of HIV-positive women. We also concluded that in view of the high frequency of baseline inflammation in sexually active women, post exposure vaginal biopsies are potentially more difficult to interpret in this group.
Vaginal biopsy is a simple, safe and effective procedure in selected populations. It is an effective measure of inflammation in the genital tract (gold standard). It is most useful in early Phase I studies in sexually abstinent populations. It has the advantage that the inflammatory infiltrate can be characterized. It allows comparison between agents with different modes of action. Care must be taken in sexually active women and in HIV-positive women (counselling is essential).
The author would like to thank the study participants, and from Imperial College, London: Miranda Cowen, Charles Lacey, Jonathan Weber, Valerie Kitchen; from the Institute of Tropical Medicine, Antwerp: Vicky Jespers, Lut Van Damme; and from the Medical Research Council Clinical Trials Unit, London: Rhian Gabe, Sheena McCormack, Andrew Nunn.
1. Stafford MK, Cain D, Rosenstein I, et al. A placebo-controlled, double-blind prospective study in healthy female volunteers of dextrin sulphate
gel: a novel potential intravaginal virucide. J Acquir Immune Defic Syndr Hum Retrovirol
2. 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
3. Van Damme L, Wright A, Depraetere K, et al. A phase I study of a novel potential intravaginal microbicide
, PRO 2000, in healthy sexually inactive women. Sex Transm Infect
4. Low-Beer N, Jespers V, McCormack S, et al. A safety study of dextrin sulphate
gel as a novel vaginal microbicide
: data from HIV negative and HIV positive women. In: 14th International AIDS Conference. Barcelona, Spain, June 2002.
5. Shaunak S, Gooderham NJ, Edwards RJ, et al. Infection by HIV-1 blocked by binding of dextrin 2-sulphate to the cell surface of activated human peripheral blood mononuclear cells and cultured T-cells. Br J Pharmacol
6. Rusconi S, Moonis M, Merrill DP, et al. Naphthalene sulfonate polymers with CD4-blocking and anti-human immunodeficiency virus type 1 activities. Antimicrob Agents Chemother
7. Hicks DS, Martin LS, Getchell JP, et al. Inactivation of HTLV-III/LAV-infected cultures of normal human lymphocytes by nonoxynol-9 in vitro. Lancet
8. 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
, 2001; 184:418–428.
9. Van Damme L, Ramjee G, Alary M, et al. Effectiveness of COL-1492, a nonoxynol-9
vaginal gel, on HIV-1 transmission in female sex workers: a randomised controlled trial. Lancet