The use of colposcopy in assessing vaginal irritation in research.

Since the early 1990s, colposcopy of the vagina and cervix has been used in the development of vaginal products in order to detect epithelial changes that may increase the likelihood of HIV or acquisition of other sexually transmitted diseases. As part of a continued effort to examine and define the role of colposcopy in a research setting, the Contraceptive Research and Development Program (CONRAD) and the International Working Group on Microbicides (IWGM), in association with the United Nations Program for AIDS (UNAIDS) sponsored a conference entitled, 'The Use of Colposcopy in Assessing Vaginal Irritation in Research', held in Washington, DC in January 1999. This is a report of that conference. The World Health Organization's colposcopy procedure and nomenclature, published in 1995, were reviewed and changes were recommended. The revised procedure involves colposcopic examination of the external genitalia, naked eye examination of the cervix, fornices, and vaginal walls, followed by lavage and colposcopic examination of those areas, and sampling as appropriate for microscopic examination. Revised nomenclature replaces the terms used for findings with descriptions of what is actually seen. Digital video imaging and testing for inflammatory markers may be adjuncts to colposcopy and should be further studied. Other areas requiring additional research include the natural history of colposcopic changes, factors other than product use that may affect colposcopic findings, the clinical significance of findings, and the procedure which best assesses these findings.

The effect of one injection of Depo-Provera on the human vaginal epithelium and cervical ectopy.

Two studies in rhesus monkeys have shown that progesterone implants, Depo-Provera and Norplant, were associated with vaginal thinning. Progesterone implants have also been associated with an increased risk of simian immunodeficiency virus (SIV) acquisition. This study in 16 women was done to assess vaginal epithelial thickness and number of cell layers from biopsies taken in the untreated follicular and luteal phases, and at 1 month and 3 months after administration of Depo-Provera. There was no significant change over time in either parameter from biopsies obtained in the luteal phase compared with those at either time after Depo-Provera administration. There was also no change in the mean number of Langerhans cells in vaginal wall specimens and no change in cervical ectopy. It appears that women do not respond to exogenous progestins with the dramatic vaginal thinning seen in rhesus monkeys.

PIP: This study assesses vaginal epithelial thickness and number of cell layers from biopsies taken in the untreated follicular and luteal phases and at 1 month and 3 months after administration of Depo-Provera. Subjects were seen at the CONRAD Clinical Research Center at the Eastern Virginia Medical School, Norfolk, Virginia. Findings showed that there was no significant change over time in either parameter from biopsies obtained in the luteal phase compared with those at either time after Depo-Provera administration. There was also no change in the mean number of Langerhans cells in vaginal wall specimens and no change in cervical ectopy. The dramatic vaginal thinning seen in rhesus monkeys was not observed among these subjects.

The promise of public/private sector collaboration in the development of vaginal microbicides.

This paper describes the efforts of the public sector, primarily the CONRAD Program and its Consortium for Industrial Collaboration in Contraceptive Research (CICCR), to work with industry in the development of vaginal microbicides for the prevention of transmission of STDs including HIV. The results of a meeting on this topic held in Durham, NC, USA in October 1998, as well as market surveys, are also included. The projects described illustrate the opportunities for collaboration and the resulting progress.

Lea's Shield: a study of the safety and efficacy of a new vaginal barrier contraceptive used with and without spermicide.

The purpose of this study was to evaluate the safety, efficacy and acceptability of Lea's Shield, a new vaginal contraceptive barrier device, when used with either spermicidal or non-spermicidal lubricant. One-hundred-eighty-five (185) women enrolled at six centers. Half were randomized to use the device with spermicide and half with a non-spermicidal lubricant. To be eligible, volunteers had to be 18-40 years old (inclusive), in good health with regular menses, sexually active in an ongoing relationship and at risk for pregnancy, and willing to use Lea's Shield as their sole means of contraception for six months. Participants were seen at admission, one week, one month, three months and six months. Gross cumulative life table rates were calculated for pregnancy and others reasons for discontinuation. Adverse experiences and responses to an acceptability questionnaire were evaluated. One-hundred-eighty-two (182) volunteers contributed data to the analysis of safety and 146 to that of contraceptive efficacy. The unadjusted six-month life table pregnancy rate was 8.7 per 100 women for spermicide users and 12.9 for non-spermicide users (p = 0.287). After controlling for age, center, and frequent prior use of barrier methods, the adjusted six-month life table pregnancy rate was 5.6 for spermicide users and 9.3 for non-spermicide users (p = 0.086), indicating that use of spermicide lowered pregnancy rates, although not significantly, during typical use. For purposes of comparison, it is important to note that this study differed from the cap/diaphragm and sponge/ diaphragm studies in that a high percentage (84%) of volunteers were parous. For reasons that are unclear, pregnancy rates among parous women using barrier contraceptives tend to be higher than among nulliparous women. Indeed, in this study there were no pregnancies among nulliparous users of Lea's Shield. Standardization of parity of this study population on those of the cap/diaphragm and sponge/diaphragm studies suggests that unadjusted pregnancy rates for this device would have been considerably lower (2.2 and 2.9 per 100 users of spermicide and non-spermicide, respectively) had the study been done using the populations of earlier studies. Since no directly comparative study has been done, these figures provide a tentative estimate of the relative efficacy of Lea's Shield compared with the sponge, cap, and diaphragm. There were no serious adverse experiences attributed to the use of Lea's Shield. Acceptability was very good. Seventy-five percent (75%) of women responded to an end-of-study questionnaire; 87% of these reported that they would recommend Lea's Shield to a friend. Lea's Shield is a new vaginal contraceptive that does not require clinician fitting. Pregnancy rates in this study compare favorably with other studies of barrier contraceptive methods including the cervical cap, diaphragm, and sponge, even though this study was done with greater rigor and with a greater percentage of parous women than previous barrier studies. Lea's Shield appears to be safe and very acceptable to study volunteers.

PIP: At six US centers, between August 1991 and October 1993, a prospective, double-blind, randomized clinical trial was conducted to evaluate the contraceptive efficacy of Lea's Shield (a new vaginal contraceptive barrier device) used with and without a spermicidal lubricant as well as its safety and acceptability. The clinical researchers enrolled 185 healthy women aged 18-40 who menstruated regularly, were sexually active, and at risk for pregnancy. 84% of the women were parous. They were asked to use Lea's Shield as their only contraceptive method for 6 months. The analysis of safety included 182 women, while that of contraceptive efficacy included 146 women. The adjusted 6-month life table pregnancy rate for spermicide users was lower than that for non-spermicide users (5.6% vs. 9.3%), but not significantly so (p = 0.086). None of the nulliparous women conceived, however. When the researchers standardized parity in this study population comparable to the parity of cap/diaphragm and sponge/diaphragm studies, the unadjusted and adjusted pregnancy rates would have been much lower (2.2% for spermicide users vs. 2.9% for non-spermicide users and 1.3% vs. 2.6%, respectively). No woman experienced serious or unexpected adverse effects using Lea's Shield. Discontinuation rates for device-related reasons were low (7.8% for spermicide users and 6.7% for non-spermicide users). 87% of women who completed an end-of-study questionnaire and 69% of men would recommend Lea's Shield to a friend. 84% of women and 55% of their partners liked Lea's Shield. 84% of women who had an opinion on the diaphragm preferred Lea's Shield. The aspects most liked were convenience and ease of insertion. These findings suggest that this new vaginal barrier contraceptive (available in one size fits all) has a relatively good contraceptive efficacy compared with other barrier methods and is safe and acceptable.

Biodegradable systems for the sustained release of fertility-regulating agents.

PIP: Biodegradable or erodible delivery systems are represented by contraceptive devices where the matrix also dissolves and both the drug and the system's components reach systemic circulation. One promising feature of such a system is the possibility of achieving programmed release whereby the active component is made available only during specified periods. A major drawback of the system is that if the device were made of microparticles, the occurence of a serious adverse reaction can be counteracted only by administration of other active principles, and not by simply removing the implant. Another drawback is that the components diffuse into systemic circulation, requiring careful long-term toxicological evaluation of hyrolysis products of the biopolymer making up the matrix. Various mechanisms are involved in the release of the active drug: these include erosion; diffusion; a combination of both, and the cleavage of a covalent linkage between the drug and the polymer backbone. Systems currently being developed include devices made of of: 1) polylactin and/or poly-glycolic acid; 2) polymers derived from e-caprolactone; 3) polypeptide polymers; 4) ploy-glutamic acid to which steroids are covalently bound, and 5) polyorthoester known as Chronomer.^ieng

Effects of cyclophosphamide and adriamycin on the healing of surgical wounds in mice.

Administration of therapeutic dose levels of cyclophosphamide as a single dose or as daily treatments for 5 days during the perisurgical period resulted in a significant decrease in the strength of surgical skin wounds in mice as measured 21 days after surgery. Administration of a single dose of 200 mg/kg either at the time of surgery or up to 4 days prior to or after surgery impaired 21-day wound strength, with the most extensive depression observed when the drug was given 1 or 2 days after surgery. Earlier stages of wound healing (day 3 or day 7) were not as sensitive to cyclophosphamide. Adriamycin in the therapeutic dosage range for mice did not significantly impair wound healing. This drug had an effect only at the LD10 dosage level. Combination treatment with cyclophosphamide plus adriamycin at the time of surgery impaired 21-day wound strength to a greater degree than observed with either agent alone, but did not significantly depress wound strength 3 or 7 days after surgery. These studies indicate that dosage level, the time of drug administration relative to surgery, and the time at which wound strength measurements are made are important parameters in determination of the effects of antineoplastic agents on wound healing.

Effects of antineoplastic agents on wound healing in mice.

The influence of antineoplastic agents on wound healing was evaluated in an experimental model employing surgical incisions in the skin of mice. A single injection in the therapeutic dose range of various clinically active antitumor agents was administered immediately following operation and the breaking strength of skin wounds was measured 3, 7, and 21 days later. Vincristine and methotrexate decreased wound strength at 3 days but not on days 7 or 21. Actinomycin D interfered with early phases of wound healing and had less effect on later phases. Treatment with bleomycin had no effect on wound strength at days 3 or 21 but prevented an increase in wound strength from day 3 to day 7; 1,3-bis(2-choroethyl)-1-nitrosurea (BCNU) decreased wound strength at all time points after operation, whereas 5-Fluorouracil had no significant effect at any time. The presence of a large tumor mass had no direct effect on wound strength nor did it modify the effect on wound strength of an antineoplastic agent, cyclophosphamide. These findings are discussed, taking into account the experimental and clinical literature on the effects of antitumor drugs on wound healing.