CERVICAL MUCUS PENETRATION TEST FOR IN VITRO ASSAY OF VAGINAL CONTRACEPTIVE AGENTS*

Transcription

1 _ 2±± 1 223i& FERTILITY AND STERILITY Copyright c 1981 The American Fertility Society Vol. 36, No.4, October 1981 Printed in U.sA. CERVICAL MUCUS PENETRATION TEST FOR IN VITRO ASSAY OF VAGINAL CONTRACEPTIVE AGENTS* NANCY J. ALEXANDER, PH.D.t:j: DAVID L. FULGHAM, B.A.:j: BETrINA ZATUCHNI LOURENS J. D. ZANEVELD, D.V.M., PH.D. Reproductive Biology, Oregon Regional Primate Research Center, Beaverton, Oregon 97006, Departments of Anatomy, Obstetrics and Gynecology, and Urology, The Oregon Health Sciences University, Portland, Oregon 97201, Departments of Physiology and Biophysics, and Obstetrics and Gynecology, University of Illinois at the Medical Center, Chicago, Illinois An in vitro penetration test employing bovine cervical mucus to evaluate vaginal contraceptives is presented. Its usefulness for the evaluation of the potency of a formulation in preventing sperm penetration was assessed by comparison of five commercially available preparations, with the use of both human and primate spermatozoa. The results were compared with those obtained with a spermicidal test (Sander-Cramer method) and in vivo with the use of the primate Macaca arctoides. Analysis of the two in vitro tests and the one in vivo test showed a significant degree of concurrence. The cervical mucus test much more closely reflected in vivo results than the Sander-Cramer test. A recommendation is made regarding the type and order of in vitro and in vivo methods to be used for the development and/or testing of new formulations. Fertil Steril36:516, 1981 With varying degrees of effectiveness, vaginal contraceptives (spermicides) have been used for thousands of years. At present, with the concerns expressed regarding hormonal contraceptive agents and the problems associated with intrauterine devices, there is an increased interest in the general use of vaginal contraceptives. However,those presently on the market are not optimally effective; Pearl index failure rates range from 1.75 to 27.9 in the United States and even higher elsewhere (e.g., Puerto Rico to 38.3).1 Received April 3, 1981; revised and accepted June 17, *The work described in this article, Publication No of the Oregon Regional Primate Research Center, was supported in part by NIH grants RR and BRS RR treprint requests: Nancy J. Alexander, Ph.D., Oregon Regional Primate Research Center, 505 N.W. 185th Avenue, Beaverton, Oregon :j:reproductive Biology, Oregon Regional Primate Research Center, Beaverton, Oregon Departments of Physiology and Biophysics, and Obstetrics and Gynecology, University of Illinois at the Medical Center, Chicago, Illinois When developing a vaginal contraceptive one should evaluate its efficacy before initiating clinical tests. The most commonly used test for evaluation ofthe in vitro spermicidal activity of a compound is the International Planned Parenthood Federation (lppf) agreed test; in which 1.0 ml of a 1:11 solution of the test substance must immobilize all the spermatozoa in 0.2 ml of semen within 40 seconds after being mixed. 2 An improved method is that of Sander and Cramer, in which the strength of a formulation is assessed through determination of the smallest concentration at which complete sperm immobilization occurs within 20 seconds. 3 These in vitro tests for spermicidal activity do not necessarily reflect the in vivo properties of a compound, and it was recently determined that a primate species, Macaca arctoides, could be used to assess the postcoital vaginal spermicidal activity of various substances.4, 5 Spermicidal tests only assess sperm motility and do not take into consideration the quality of motility or other factors that may affect sperma- 516

2 Vol. 36, No.4 IN VITRO ASSAY OF VAGINAL CONTRACEPTIVES 517 tozoa. For example, certain formulations may not be immediately lethal to the spermatozoa but may nonetheless prevent cervical mucus penetration and thus conception. The objective of the study reported here was development of a test of the abilities of compounds to inhibit sperm penetration of cervical mucus. In this paper we describe the cervical mucus penetration test and the results of several trials with five commercially available vaginal contraceptives, and we compare this test with the Sander-Cramer test for spermicidal activity and the primate vaginal spermicidal test. We used bovine cervical mucus to permit the testing of several agents against a standard of cervical mucus. 6 MATERIALS AND METHODS Materials. Five nonprescription commercial vaginal contraceptives were evaluated. The compositions of contraceptives change from time to time without accompanying name changes, and so the composition of each contraceptive is indicated: Conceptrol Birth Control Cream (containing 5.0% nonoxynol-9 [N-9]; Ortho Pharmaceutical Corporation, Raritan, N. J.); Encare Oval (a suppository containing 2.5% N-9; Norwich-Eaton Pharmaceuticals, Norwich, N. Y.); Delfen Contraceptive Foam (containing 12% N-9; Ortho Pharmaceutical Corporation, Raritan, N. J.); Ortho Gynol Contraceptive Jelly (containing 1% Triton X-100; Ortho Pharmaceutical Corporation, Raritan, N. J.); and Ramses "10-Hour" Vaginal Jelly (containing 1% boric acid and 5% alcohol; Schmid Products Company, Little Falls, N. J.). Preparation of the Formulations. For the in vitro spermicidal test 3 an appropriate amount of the test substance was weighed and diluted in physiologic saline (wt:vol) to the desired stock solution in a test tube, and for the cervical mucus penetration test Baker's buffer was used as the diluent. 7 The preparations were mixed on a vortex unit at a high speed for two 30-second periods, and then were allowed to stand at 37 C for 30 minutes. Encare Oval required additional mixing during incubation for complete dissolution. Collection and Preparation of Semen Samples. Rhesus macaques (Macaca mulatta) underwent electroejaculation by the penile band technique. 8 The anal probe technique was used for cynomolgus macaques (Macaca fascicularis).9 The semen sample was incubated at 37 C for 30 minutes, and the exudate was collected. Human semen, ejaculated by masturbation into a clean glass vial, was allowed to liquefy at 37 C for 30 minutes. Only samples that liquefied completely within this period of time were used. The volume, sperm concentration, sperm motility, and forward progression of the ejaculated spermatozoa were measured. Occasionally, semen samples from the monkeys were pooled, but only those samples showing an overall sperm motility of more than 70% were used. The semen samples were diluted to a concentration of 60 x 10 6 spermlml. Only those samples that, after dilution, possessed an overall sperm motility of more than 60% and a forward progression of at least 2.5 (on a scale of 0 to 4) were used for the in vitro tests. Cervical Mucus Penetration Test. Cervical mucus was collected from cows during estrus. Round inicrocapillary hematocrit tubes 75 mm in length with inside diameters 1.1 to 1.2 mm (Sherwood Medical Industries, St. Louis, Mo.) were used. Cervical mucus was loaded into each capillary tube through application of a vacuum to one end of the tube. The tube was removed from the mucus, and the resulting strand of mucus was cut with scissors. The tube was then sealed at both ends with Parafilm (American Can Company, Greenwich, Conn.) and stored at - 20 C until use. Only tubes with more than 40 mm of cervical mucus and without bubbles or debris after thawing were used for the tests. Each tube containing cervical mucus was scored with a diamond pencil, broken at the mucus interface, and inserted into the semen-spermicide (0.05 ml of semen diluted to 60 x 10 6 sperm/ml plus 0.25 ml of the test substance) or semen-control mixture (0.05 ml semen plus 0.25 ml Baker's buffer).lo After 60 minutes, the capillary tube was removed from the 37 C incubator, and the migration of spermatozoa into the mucus column was assessed microscopically at a magnification of 100 x. The lowest concentration of a formulation that prevented sperm penetration or allowed it to occur for a distance less than 10 mm was determined. In Vitro Spermicidal Test. The Sander-Cramer test was performed to determine the minimum concentration at which a formulation immobilized spermatozoa within 20 seconds. 3 A drop of the mixture was wet-mounted and microscopically viewed at a magnification of 100 x. Vaginal Spermicidal Test. The postcoital spermicidal tests were performed in unsedated stumptailed macaques (Macaca arctoides).5 Different amounts of the contraceptives were used to obtain dose-response curves. Each agent was placed as close to the cervix as possible with a syringe. The

3 518 ALEXANDER ET AL. October 1981 TABLE 1. In Vitro Spermicidal Potencies (Sander-Cramer Test) of Five Vaginal Contraceptives a Potency of vaginal contraceptives Species of spennatozoa Delfen Conceptrol Encare Ortho- Ramses Gynol mg/ml mg/ml mg/ml mg/ml mg/ml Human Monkey alowest concentration to produce complete immobilization of spermatozoa within 20 seconds. See the text for experimental details. Each test was performed twice, and the results were averaged if different. Kendall's coefficient of concordance resulted in P < 0.01, with concordance = 0.95, sets = 2, and N = 5. female was immediately placed in the cage of a male of proven fertility. The coital act normally took place within 2 to 3 minutes and was completed in 1 to 2 minutes. The female was immediately separated from the male, and the vaginal fluid was flushed and collected with a tuberculin syringe containing 0.25 ml of physiologic saline. The addition of 0.25 ml of saline facilitated the recovery of semen because M. arctoides has a small ejaculate volume. The collected fluid was immediately observed under a microscope, and the percentage of motile sperm was determined. Six female monkeys were tested at each concentration of the test substances. Four concentrations of each test substance were analyzed. Statistical Analysis. Kendall's coefficient of concordance was used to determine the degree of agreement for the in vitro and in vivo testsy The data generated from the primate experiments were plotted as dose-response curves and subjected to the parallel-line assay12 for comparison of relative potency. RESULTS The in vitro spermicidal test results (Sander Cramer) are shown in Table 1. The test gave repeatable results, although occasionally small differences (hardly ever more than twofold) were found in the potency of a formulation with different semen samples. Such differences were probably due to greater or lesser resistance of spermatozoa from different ejaculates to the spermicidal effects of a preparation. The results with Delfen and Encare compared well with earlier data: 1.0 and mg/ml, respectively, were spermicidal. within 20 seconds for human spermatozoa and 0.5 and 10.0 mg/ml, respectively, for monkey spermatozoa. 5 The spermicidal effects of the various formulations on human and monkey spermatozoa were quite similar, with the notable exception of Encare, which had a spermicidal potency with monkey sperm approximately 10 times that with human sperm. Delfen was the most potent spermicide for both human and monkey spermatozoa. Thereafter, in decreasing order of potency for human spermatozoa, came Conceptrol, Encare, Ortho-Gynol, and Ramses. The same order was found for the monkey spermatozoa, except that Encare and Ortho Gynol were reversed owing to the higher activity of the former with monkey spermatozoa. The rankings of the monkey versus human spermatozoa had a concordance of P < Even though the orders of effectiveness of the spermicides on monkey and human spermatozoa were approximately the same, large differences existed in the spermicidal effects of the compounds. For instance, with human sperm, Ramses was approximately 600 times less active than Delfen, but with monkey semen it was only 150 times less active. Similarly, there was a 100-fold difference between Ramses and Conceptrol with human spermatozoa, but only a 10-fold difference with monkey spermatozoa. Therefore, the relative susceptibilities of monkey and human spermatozoa to the formulations were significantly different. There was a greater consistency between the human and monkey spermatozoa regarding the rank order of potency of the contraceptives in the cervical mucus penetration test (Table 2). Also, the relative differences in potency of the preparations against human and monkey spermatozoa were smaller in the cervical mucus penetration test than in the Sander-Cramer test. All agents inhibited cervical mucus penetration at a concentration less than that necessary to immobilize spermatozoa in the Sander-Cramer test; this phenomenon was most obvious with the weaker spermicides. For instance, a concentration of Ramses 100 times smaller than that sper- TABLE 2. Potencies of Vaginal Contraceptives in Preventing Sperm from Penetrating Cervical Mucus a Species of spennatozoa Delfen Potency of vaginal contraceptives Conceptrol Encare Ortho- Ramses Gynol mg/ml mg/ml mg/ml mg/ml mg/ml Human Monkey alowest concentration to prevent sperm entry into cervical mucus or entry ofless than 10 mm. See the text for experimental details. Each test was performed in duplicate. Kendall's coefficient of concordance resulted in P < 0.01, with concordance = 0.90, sets = 2, and N = 5.

4 Vol. 36, No.4 IN VITRO ASSAY OF VAGINAL CONTRACEPTIVES 519 TABLE 3. Results of Parallel-Line Analysis of the Postcoital Test and Relative Potencies a Vaginal preparation Potency Confidence limits Delfen Conceptrol 0.239b Ortho-Gynol b Encare 0.056b Ramses 0.042b afour dose levels of each formulation were tested. Each dose level was evaluated in six primates. bsignificantly different from Delfen (P < 0.05). micidal in the Sander-Cramer test prevented cervical mucus penetration. As such, the differences in potency between the agents were also reduced; Ramses was only 20 times less potent than Delfen in preventing sperm penetration of cervical mucus, even though it had the spermicidal potency. Since in vitro and in vivo results are sometimes quite different, a comparison was important. Using the primate model, the relative vaginal spermicidal potencies of the compounds were determined on the basis of weight in milligrams. The following ranking was obtained: Delfen, Conceptrol, Ortho-Gynol, Encare, and Ramses (Table 3). For each milligram of Delfen, one would require 4.2 mg of Conceptrol, 7.6 mg of Ortho-Gynol, 17.9 mg of Encare, and 23.8 mg of Ramses to achieve the same effect. All these products were significantly less effective than Delfen (P < 0.05). By means of Kendall's coefficient of concordance, we evaluated the degree of similarity between the results of the two in vitro tests and the results of the in vivo test; there was close agreement. All possible combinations yielded a P value of less than 0.0l. It is notable, however, that the results obtained with the primate test were more similar to those of the cervical mucus penetration test than the Sander-Cramer test. For instance, Ramses and Ortho-Gynol were, respectively, approximately 20 and 4 to 8 times less active than Delfen in both the primate and cervical mucus tests but, respectively, 150 to 600 and 38 to 114 times less potent in the Sander-Cramer test. Even though some quantitative differences are also present between the cervical mucus test and the primate test, it is apparent that the cervical mucus test much more closely reflects in vivo results than the Sander Cramer test. DISCUSSION To assess the contraceptive potential of a substance in vitro, one must not only evaluate its spermicidal activity but also determine its ability to prevent entry of spermatozoa into cervical mucus. The data reported here have revealed that relatively weak spermicidal substances are much more effective in preventing the passage of spermatozoa into cervical mucus than their spermicidal activity would indicate. Although certain compounds are much less effective than others when tested only for their spermicidal effects in vitro, the differences are sometimes much smaller when the cervical mucus data are compared. Alternatively, two agents may have equal spermicidal potency, but one may be more effective in preventing cervical mucus penetration and thus may be the better contraceptive in vivo. In vivo spermicidal tests are important for the same reasons. As shown by our data, differences in spermicidal potency may be present in vitro (as measured by the Sander-Cramer test) or in vivo. Only one animal model is currently available for such studies: the stump-tailed macaque (Macaca arctoides).5 Some mixtures already studied are relatively good spermicides in vitro but have no, or almost no, potency in vivo (unpublished results). The cervical mucus penetration test described in this article was designed for ease of performance and low cost. When one uses bovine cervical mucus, one can obtain large amounts from the same donor and so perform several tests with the same sample, an important feature, because cervical mucus collected from separate individuals or on different days of the cycle could differ in composition. Only small quantities of cervical mucus can be collected from a woman. It can be argued that bovine and human cervical mucus have different properties and that their penetration by human spermatozoa may differ from that by bovine sperm, but all data accumulated thus far negate this argument. 6 An additional advantage of the technique is that the mucus-containing tubes can be kept frozen. Therefore, all the tubes can be loaded at one time, and the test can be performed at a later date. Also, it allows shipment of the tubes to various laboratories. The freezing procedure does not appear to alter the ability of sperm to penetrate bovine cervical mucus. 6 The susceptibilities of human and monkey spermatozoa to the spermicidal action of contraceptives are similar, although small differences do exist. The exception is Encare, which shows major differences when human and monkey spermatozoa are used. Since the active ingredient in Encare is the same as that in Delfen and Conceptrol,

5 520 ALEXANDER ET AL. October 1981 Le., nonoxynol-9; perhaps the base into which the nonoxynol-9 is incorporated or the total amount of the detergent used causes the lesser susceptibility of human spermatozoa. For an as yet unknown reason; Encare does not show this species difference in the cervical mucus penetration test. The evidence that the spermicidal and antipenetration actions of agents depend, at least in part, on different principles suggests that it is important to test for both activities. The only definitive test for vaginal contraceptive activity is evaluation in women. However, because such tests are fraught with technical problems and difficulties in interpretation, consistency among different investigators in different parts of the world is not to be expected. Also, new compounds not yet approved for human use by the U. S.Food and Drug Administration cannot be tested clinically. Finally, clinical studies are very complicated, time-consuming, and expensive. One should, therefore, employ in vitro and animal tests to ensure thaf a substance has a good chance of being contraceptive before beginning clinical studies. A number of techniques are now available. We recommend that the following tests be performed, in the indicated order, by persons developing and testing new formulations: (1) an in vitro spermicidal test (preferably the Sander-Cramer test)3; (2) an in vitro cervical mucus penetration test (as described here); (3) an in vivo primate vaginal spermicidal and cervical mucus penetration test (the vaginal spermicidal test5; one can collect cervical mucus simultaneously to evaluate the capacity of the sperml! tozoa to penetrate the cervix); (4) breeding experiments, either first with rabbits13 and then with monkeys,14, 15 or with monkeys alone; and (5) clinical experiments if the contraceptive is safe for human use, either first in the vaginal spermicidal tests described by Masters and associates16 and then in use-effectiveness studies, or in use-effectiveness studies alone. We suggest this order because it starts with simple and inexpensive tests and proceeds to more expensive and complicated ones. At each step; the agent being tested should be compared with contraceptives already on the market. Acknowledgment. We appreciate the cooperation of Dr. Do Won Hahn,.of Ortho Pharmaceutical Corporation, in perfonning the statistical analyses of the in vivo primate data. REFERENCES 1. Sobrero AJ: Spennicidal agents: effectiveness, use, and testing. In Vaginal Contraception: New Developments, Edited by GI Zatuchni, AJ Sobrero, JJ Speidel, JJ Sciarra, Hagerstown, Md, Harper & Row, 1979, p Kleinman RD (Ed): International Planned Parenthood Federation Medical Handbook, Second Edition. London, International Planned Parenthood Federation, Sander FV, Cramer SD: A practical method for testing the spermicidal action of chemical contraceptives. J Hum Fertil 6:134, Zaneveld LID, Bhattacharyya AK, Kim DS, Schumacher GFB, Beluhan Z: Primate model. for the evaluation of vaginal contraceptives. Am J Obstet Gynecol 129:368, Zatuchni B, Hahn DW, Zaneveld LID: Postcoital, vaginal spermicidal potency offonnulations: the Maeaca arctoides (stumptailed macaque) as animal model. Fertil Steril 35:683, Alexander NJ: Evaluation of male infertility with an in vitro cervical mucus penetration test. Fertil Steril36:201, Baker JR: The spennicidal powers of chemical contraceptives. IV. More pure substances. J Hyg (Camb) 32:171, Mastroianni L, Manson WA: Collection of monkey semen by electroejaculation. Proc Soc Exp BioI Med 112:1025, Fulgham DL, Alexander NJ: Unpublished data 10. Alexander NJ, Fulgham DL: Antibodies to spennatozoa in male monkeys: mode of action. Fertil Steril 30:334, Siegel S: Nonparametric Statistics for the Behavioral Sciences. New York, McGraw-Hill Company, 1956, p Finney OJ: Statistical Methods in Biological Assay. Cambridge, Cambridge University Press, 1971, p Homm RE, Doscher GE, Hummel EJ, Greenslade FC: Relative antifertility activity of three vaginal contraceptive products in tbe rabbit: relationship to in vitro data. Contraception 13:479, Alexander NJ: Vasect.omy and vasovasostomy in rhesus monkeys: the effect of circulating antispenn antibodies on fertility. Fertil Steril 28:562, Zaneveld LID, Beyler SA, Kim DS, Bhattacharyya AK: Acrosin inhibitors as vaginal contraceptives in the primate and their acute toxicity. BioI Reprod 20:1045, Masters WH, Kolodny RC, Johnson-Masters VE: In vivo testing of intravaginal contraceptives. In Vaginal Contraception: New Developments, Edited by GI Zatuchni, AJ Sobrero, JJ Speidel, JJ Sciarra. Hagerstown, Md, Harper & Row, 1979, p 256