Vaginal Ring Hormone Delivery Systems in Contraception and Menopause

The vagina has been identified as a drug absorption area since the early 1900s. 1 During the 1960s, Dziuk and Cook 2 discovered that sex steroid-containing polysiloxane (silicone rubber) tubes or disks would release a drug into a saline bath at a continuous rate. The vaginal delivery of sex steroids was first patented by Duncan at Upjohn (Peapack, NJ) in 1970. Since then, more than a dozen other patents for vaginal contraceptive rings have been filed. Contraceptive rings were formulated of polysiloxane. The first to be studied released medroxyprogesterone acetate and demonstrated the advantage of self-insertion. 3 Other early progestin-only products tested included norethindrone in 50 and 200-μg release rates per 24 hours, 4 norgestrel, 5 and progestin R-2323. 6 Three main sources for fabrication of rings were the Population Council Laboratories, the World Health Organization Program in Human Reproduction, and Organon International (BH Oss, Netherlands). Ring Design The first rings studied were homogenous devices with the steroid mixed uniformly through the polysiloxane matrix. The design was largely abandoned because of a high initial release of drug with rapid decrease of drug release thereafter. To optimize performance, several other ring designs were developed. One had steroid located in a polysiloxane core in the center of the device (core design). Steroid release was proportional to the surface area of the core and inversely related to the thickness of the polysiloxane overcoat. Duration of action depended on the total steroid load in the ring. Another design affecting stable, prolonged release of steroid was the shell design, in which a thin membrane of concentrated steroid was sandwiched between an unmedicated polysiloxane core and a metering layer of polysiloxane to control steroid release. Another compound used for vaginal devices is ethylene vinyl acetate. Roumen et al 7 studied unmedicated devices with a diameter of 54 mm and cross-sections of 3, 3.5, and 4 mm. The widest diameter was most acceptable. The devices had a 21-day continuation rate of 91% (59 of 65). Contraceptive Pharmaceutical Agents PROGESTIN ALONE Early progestin-only rings that contained 100 mg of medroxyprogesterone acetate successfully suppressed ovulation. 3 They were removed after 21 days of use to allow withdrawal bleeding and were replaced on cycle day 5. Early norethindrone-containing rings were associated with ovulation and unacceptable bleeding. 4,5 These early compounds were superceded by others. PROGESTERONE Vaginal rings of homogenous design with up to 2.2 mg of progesterone release per day and 0.22 mg of estradiol were evaluated in cycling women, 8 but they demonstrated poor cycle control and unreliable ovulation inhibition. The devices did demonstrate the potential for high-dose vaginal steroid delivery. A larger study using devices releasing 5 or 10 mg of progesterone alone in lactating women found one pregnancy in 739 women-months of use compared with 19 pregnancies in the 677 women-months using lactation only. 9 A multicenter study comparing a vaginal ring releasing 10 mg of progesterone daily with a copper intrauterine device in 1,500 lactating women found a 1-year pregnancy rate of 1.7 and 0.5%, respectively. 10 The product was licensed by the Population Council to a Chilean company. The commercial product was fabricated for 3-month continuous use with average daily release of 10 mg of progesterone. It was similarly compared with the copper intrauterine device in 500 lactating Chilean women. No pregnancies occurred during more than 1,200 months of exposure with either the intrauterine device or the vaginal ring. 11 This lead to product registration in Chile. Synthetic Progestins ETONOGESTREL The first published report in 1989 described ethylene vinyl acetate rings measuring 4 x 56 mm and releasing 15 or 30 μg of etonogestrel (ETG) per day. 12 The pharmacokinetics of ring insertion and removal were assessed. No burst progestin release was seen. In the plateau phase, release rates decreased 5% per 21-day cycle. The estimated bioavailability of ETG (3 keto-desogestrel) from vaginal rings was 97% to 102%. 13 In vitro release data correlated well with in vivo release. LEVONGESTREL/NORGESTREL Early norgestrel devices provided excellent cycle inhibition. 5 The World Health Organization tested levonorgestrel-releasing vaginal rings extensively, culminating in a multinational trial of 20-μg levonorgestrel-releasing rings. 14 The devices were to be worn for 3 consecutive months by more than 1,000 women. The devices were effective, with a 12-month pregnancy rate of 3.6 per 100 women, despite ovulatory function in one-third of the cycles. Discontinuations for menstrual problems were 17.2 per 100 women at 1 year. As with other progestin-only methods, menstrual blood loss was actually two-thirds of baseline with the method. The device showed an expulsion rate of 7.1 per 100 women and was associated with vaginal epithelial lesions 15 that halted further product development. MEGESTROL The Chinese have studied a vaginal ring that released megestrol. Ji-Hua et al 16 found that thicker mucus and impaired endometrial progesterone receptor production lead to a pregnancy rate of only 6 of 100 woman-years despite minimal inhibition of ovulation. NESTORONE Nestorone (ST-1435 or 17-acetoxy-16-methylene-19-norprogesterone), a novel progestin with low androgenicity, is being studied by the Population Council. Core design rings releasing 0.1 mg daily of nestorone resulted in a 47% amenorrhea rate. More than half (53%) of participants noted oligomenorrhea after 3 months of use. 17 No changes in total cholesterol, high-density lipoprotein cholesterol, or triglycerides were noted at 12 weeks. Because of a number of ovulations when used intermittently, use was subsequently studied continuously at doses of 50, 75, and 100 mcg. 18 Because the progestin is largely inactivated in the gastrointestinal tract, it is an attractive medication for use in lactating women. NORGESTRIENONE The chemical name of the progestin, norgestrienone (R-2010), is 17α-ethinyl-17β-hydroxy-estra-4,9,11-trien-3-one. Four women used a ring releasing this compound for 12 intermittent cycles with 3 weeks of use, followed-up by 1 week without the device. One-fourth of the cycles showed ovulation, and bleeding patterns were irregular. 19 Combination Estrogen and Progestin Early attempts at combining estrogen and progestin were logical extensions of oral contraceptive development in which hormone combinations were found to provide better cycle control at lower doses. Åhrens et al 20 reported on a comparative series of rings releasing 0.2 mg of estradiol with either 0.29 mg levongestrel, 0.85 mg norgestrel, 0.7 mg medroxyprogesterone acetate, or 1.9 mg megestrol acetate. Only the levonorgestrel-releasing vaginal ring suppressed ovarian function well and maintained a good bleeding pattern. The lipoprotein profile was unfavorable, with high-density lipoprotein cholesterol decreasing 32%. With a more potent estrogen, ethinyl estradiol (EE), an increase of sex hormone-binding globulin was noted, unlike estradiol, but a less negative effect on lipids was identified. Bioavailability of EE was high: up to 100% in ETG combination products, and in vitro release rates correlated well with calculated in vivo release. 21 The next paragraphs describe combinations that were studied more extensively and warrant discussion. LEVONGESTREL/NORGESTREL COMBINATIONS The earliest extensively studied contraceptive combination vaginal ring released 150 to 180 mcg of estradiol and 250 to 290 mcg of levongestrel daily. The ring was used cyclically to mimic oral delivery. The multicenter trial was conducted in 10 centers worldwide and included a comparison group with an oral contraceptive with 30 mcg of EE and 150 mcg of levongestrel. 22,23 The device was as effective as oral contraceptives, with a pregnancy rate less than 2 of 100 woman-years. The 50% continuation rates compared well with the 38% continuation rate with the oral route. The vaginal device resulted in fewer symptoms of headache or nausea. Development was shelved when macaque studies showed increased coronary artery atherosclerosis compared with a 50-mcg EE plus 0.5 mg norgestrel oral contraceptive. ETONOGESTREL COMBINATIONS Olsson and Odlind 24 reported on three consecutive 28-day cycles of intermittent use of rings releasing 15 mcg of EE and 150 mcg of ETG in 1990. Three of 11 women discontinued the ring early because of side effects: two because of the sensation of a foreign body and mood changes and one because of increased vaginal discharge. All symptoms subsided within a few days of the ring discontinuation. Bleeding pattern was excellent, with no breakthrough bleeding or spotting in 22 of 29 completed cycles (76%). Two cycles had bleeding during week 3 and were without withdrawal bleeding during the ring-free interval. Other cycles had breakthrough bleeding or spotting during the active treatment. No ovulatory cycle was noted while the drug was administered. Estradiol levels were 90 to 275 pmol/L during treatment, increasing during the ring-free interval. Apter et al 25 performed dose-finding studies using 15 mcg of EE while varying the ETG dose. One hundred micrograms or more of ETG provided good bleeding patterns and excellent suppression of follicular function. All detected follicles were less than 1 cm in diameter and serum progesterone levels remained less than 2.0 ng/dl. A multicenter trial in Germany of manufactured rings releasing 15 mcg of EE and 120 mcg of ETG in 50 volunteers over the course of 405 treatment cycles recorded no pregnancy. 26 Acceptability of the ring was excellent, with 95% of couples finding it easy to insert and not bothersome during intercourse. Irregular bleeding and lack of withdrawal bleeding was reported in 5% of cycles. A phase III US study with the 15-mcg EE, 120-mcg ETG ring of ethylene vinyl acetate in a homogeneous design has lead to a New Drug Application submission to the Food and Drug Administration. Similar acceptability and bleeding control as in the published European clinical trial data are expected. Product registration in the US is anticipated soon by Organon, with plans to market the product using the trademark, NuvaRing. NESTORONE COMBINATIONS Nestorone, a progestin in development by the Population Council, is chemically 17-acetoxy-16methylene-19 norprogesterone. It has been studied with 60 mg in combination with 122 mg of estradiol with no effect on low-density lipoprotein cholesterol, but a 40% increase in high-density lipoprotein and total cholesterol. 27 The dose of 60 mg of nestorone was thought to be higher than actually required for ovulation inhibition. Combinations with EE were also studied, showing no significant change in total or high-density lipoprotein cholesterol, 28 good bleeding, and ovulation control using 30 mcg of EE and 100 mcg of nestorone. 17 Commercial production of a nestorone/EE ring is anticipated with a phase III trial projected to begin in 2001 (Personal communication: Theodore M. Jackanicz; November 6, 2000). NORETHINDRONE COMBINATIONS Norethindrone was substituted for levonorgestrel in early combination rings releasing estradiol, with improvement in serum cholesterol level but inadequate cycle control. 27 Next, 600 to 1,000 mg of norethindrone acetate were paired with 15 to 65 mcg of EE. Certain combinations provided excellent cycle control and ovulation suppression with minimal lipid changes. 29,30 Sixty women were enrolled in a 1-year trial of a Population Council ring releasing 20 mcg of EE and 1 mg of norethindrone acetate. The ring provided excellent blood hormone levels for nine cycles but showed 10% to 21% reduction in serum hormone levels between cycles nine and 13. Whereas no pregnancy occurred, ovulation (progesterone > 32 nmol/L) was noted in 18 cycles. Nausea was reported by half of the patients, and vomiting was reported by 20% in the first cycle of use. 31 A study soaking the device overnight, or early evening insertion with temporary removal, demonstrated that initial nausea could be reduced substantially. 32 An attempt at lowering the dose of EE to 15 mcg with the 1 mg of norethindrone acetate over the course of the four treatment cycles showed no clear difference in serum hormone level. Progesterone suggestive of ovulation was noted in 1% of treatment cycles. A trend toward a better bleeding pattern with less luteal activity was seen with the 20-mcg EE/norethindrone acetate ring compared with the 15-mcg EE/norethindrone acetate ring. 31 This hormone combination was also tested in vaginal rings produced commercially by Parke Davis (Morris Plains, NJ), but data from this multicenter trial have not been published. Rings for Menopause Vaginal continuous-release technology has also been used for menopausal symptoms. A core design estradiol-releasing ring measuring 9 mm x 55 mm is currently marketed in the US by Pharmacia (Pharmacia AB, Uppsala, Sweden) as Estring. It is indicated for urogenital symptoms and does not provide full estrogen replacement. The ring releases 7 to 8 μg/day. Serum estrogen is slightly increased but usually remains in the menopausal range. 33 However, women using the product have shown bone density increases at the wrist, suggesting some additional benefit to this low-dose therapy. Compared with estradiol (E2) pessaries, the ring is equally efficacious in alleviating dysuria (76%), urge incontinence (58%), and stress incontinence (53%), but was much more acceptable (60% vs 14%) in a Scandinavian study. 34 Patients also preferred the vaginal ring to estradiol cream. 35 The device delayed the recurrence of urinary tract infections in postmenopausal women with three or more infections during the previous year. 36 The Population Council researchers 37 have evaluated higher-dose vaginal delivery rings providing 60 or 140 mcg of estradiol per day. Hot flashes were reduced 80% and mood and urogenital symptoms were improved. Fourteen of 70 women discontinued before 6 months of use because of expulsion (5 women) or other reasons. Mean estradiol levels were 123 pmol/L for the 60-mcg E ring and 307 pmol/L for the 140-mcg E-releasing ring. An Irish company, Galen (Craigavon, United Kingdom), is developing a vaginal ring that will provide full estrogen replacement. The ring is currently undergoing phase III trials. They are also developing a combination product for complete hormone replacement. Phase III trials are expected in early 2001. Clinical Concerns With Vaginal Rings SIZE Expulsion is rare, but is occasionally seen with bowel movements, straining, or tampon removal. The size is a compromise. When a 50-mm ring was compared with a 58-mm ring, women with a uterus experienced more expulsions with the smaller ring. 38 After hysterectomy, a small percent of women have a vagina that is too small to accommodate a 58-mm diameter ring. REMOVAL FOR INTERCOURSE The vast majority of women do not remove the ring for intercourse. In a Latin American study, only 10% of women removed the ring for every sex act. 39 Fewer than 1% to 2% experienced discomfort or reported discomfort from their partners. Some partners seemed unaware of the device altogether. ODOR AND DISCOLORATION Less than 1% of women reported difficulties with odor that they attributed to the ring. The polysiloxane products do, however, show color changes in some users. VAGINAL DISCHARGE All rings are associated with increased vaginal secretions, which are a clear goal in the case of the Estring (Pharmacia), or a side effect for contraceptive products. In a contraceptive multicenter study, 23% of the ring users reported discharge, compared with 14% of the oral contraceptive users. 22 The mechanical reaction stimulates some discharge because unmedicated rings show similar rates of discharge. 7 The vaginal flora of the posterior vaginal fornix was maintained with the contraceptive product by culture evaluation. 40 The vaginal bacteriology of device users was similar to oral contraceptive users. A number of women have had monilial vaginitis or bacterial vaginosis develop during use. These infections have been treated successfully while they continued to use the product. Benefits of Vaginal Delivery For both contraception and menopause therapy, the constant hormone release rates after the first 24 to 48 hours provide excellent patient compliance with therapy. A number of women have worn devices beyond the prescribed time, proving that efficacy continues for some time. One study suggested wearing contraceptive rings up to the point of unacceptable breakthrough bleeding. 41 This would be especially feasible if the estrogen release decreased sooner than did the progestin component producing estrogen-withdrawal bleeding, while contraceptive levels of progestin are maintained. Acknowledgments The author is indebted to Dr. Theodore Jackanicz for his insightful comments on the manuscript. I am grateful to Ms. Rebecca Marquez for her help in the preparation of the manuscript. Support for this project was provided by the Contraceptive Research and Development (CONRAD) Program, Eastern Virginia Medical School, under a cooperative agreement with the US Agency for International Development (USAID). The views expressed by the author do not necessarily reflect the views of USAID or CONRAD.