Abstract
Objectives:
To evaluate the safety outcomes from clinical studies of a 12-month Contraceptive Vaginal System (CVS) releasing an average of segesterone acetate (SA) 150 mcg and ethinyl estradiol (EE) 13 mcg daily.
Study Design:
We integrated clinical safety data from nine studies of the CVS. Four studies used the final manufactured CVS, including a one-year pharmacokinetic study, two Phase 3 one-year trials, and a second year of treatment extension study. We assessed safety by evaluating adverse events reported by subjects on a daily diary and reviewed at scheduled study assessments. We also reviewed focused safety studies evaluating endometrial biopsies, vaginal microbiology and liver proteins.
Results:
The combined studies include 3052 women; 2308 women received the final manufactured CVS of whom 999 (43.3%) completed 13 cycles of use. Women using the final CVS most commonly reported adverse events of headache (n=601, 26%), nausea (n=420, 18%), vaginal discharge/vulvovaginal mycotic infection (n=242,10%), and abdominal pain (n=225, 10%). Less than 1.5% of women discontinued for each of these complaints. Four (0.2%) women experienced venous thromboembolism (VTE), three of whom had other risk factors for thrombosis (one with Factor V Leiden mutation and two with body mass index >29 kg/m2). During 21,842 treatment cycles evaluable for expulsion, women reported 4259 (19.8%) partial and 1509 (7%) complete expulsions, most frequently in the initial cycle (499/2050 [24.3%] and 190/2050 [9.3%], respectively). Focused safety studies revealed no safety concern.
Conclusion:
The one-year SA/EE CVS has an acceptable safety profile. Studies are warranted in obese women who are at higher risk of VTE.
Implications:
This one-year contraceptive vaginal system represents a new long-term user-controlled and procedure-free option with a similar safety profile to other combination hormonal contraceptives (CHC). The same precautions required for CHC prescription apply.
Keywords: Contraceptive Vaginal System, Nestorone, segesterone acetate, ethinyl estradiol, adverse events
1.0. INTRODUCTION
Segesterone acetate (SA), also known as Nestorone®, is a novel 19-nor-progesterone that is not active orally but highly effective when administered via non-oral routes such as implantable, vaginal and transdermal systems [1]. Like progesterone, SA binds specifically to progesterone receptors but not to androgen or estrogen receptors, thus limiting undesired androgen, antiandrogen or estrogenic effects [2]. Although SA binds to the glucocorticoid receptor no glucocorticoid-related effects appear at therapeutic doses (2). SA does not transactivate the androgen receptor and its metabolites do not interact with the GABA-A receptor [3]. This latter finding could theoretically avoid mood changes and sleepiness described with progesterone treatment related to allopregnanolone (the main progesterone metabolite) action on the GABA-A receptor [3].
The Population Council developed a segesterone acetate/ethinyl estradiol (SA/EE) Contraceptive Vaginal System (CVS) in a ring-shape with a final formulation delivering an average of SA 150 mcg and ethinyl estradiol (EE) 13 mcg daily. This one-year user-controlled ring has minimal to no need for daily attention and can be used without a trained health provider to insert or remove. Two Phase 3 studies and seven other clinical studies including a broad range of women at United States (US) and non-US sites included CVS safety data evaluations. This report summarizes the safety data from pooled analyses of adverse events (AEs) and clinical laboratory values and results from these focused safety studies.
2.0. MATERIAL AND METHODS
The overall safety evaluation plan for the SA/EE 150/13 CVS program includes Phase 1 and 2 studies of a pilot formulation of the CVS, with multiple doses manufactured by the Population Council (New York, NY), two Phase 3 studies with the final formulation manufactured by QPharma (Malmö, Sweden), and studies with a focused safety endpoint also using the final formulation. These focused evaluations include three sub-studies nested in the pivotal Phase 3 US trial at specific sites, which also assessed vaginal microbiology [4], endometrial changes, and coagulation factor changes [5] in addition to the primary efficacy and safety endpoints for the pivotal studies. A drug-drug interaction study tested the effects of miconazole formulations on EE and SA vaginal absorption [6]; a pharmacokinetic study compared EE exposure between the pilot SA/EE 150/13 CVS and an approved levonorgestrel (LNG) 150mcg /EE 30mcg oral contraceptive during one cycle of use; and a pharmacodynamic Phase 1, 3-month study compared EE impact on hepatic proteins from oral versus vaginal delivery [7]. Finally, a 13-cycle Phase 2 dose-finding study of the pilot SA/EE 150/13 CVS included colposcopic evaluation of vaginal and cervical epithelium [8,9]. Investigators conducted all studies with appropriate Institutional Review Board approval; and all subjects signed written informed consent prior to initiating any study procedures.
Two pivotal Phase 3, 13-cycle studies provide most of the safety information: a U.S. only study and a multinational study including U.S. and non-U.S. sites. Subjects self-reported adverse events, concomitant medications and CVS expulsions (partial or complete) on diary cards (see definitions in Supplement). During the first 6 months of study enrollment, two venous thromboembolic events (VTEs) occurred in women in the U.S.-only study who had a pre-treatment BMI>29.0 kg/m2. The Data Safety Monitoring Board recommended limiting further enrollment to women with pre-treatment BMI <29.0 kg/m2 and to discontinue women in the two studies with a BMI>29.0 kg/m2. An additional open-label investigation collected pharmacokinetic (PK) and pharmacodynamic (PD) data in 39 subjects during Cycles 1, 3, and 13 of CVS use and safety data from all study cycles [See Supplement].
For the overall integrated safety summary analysis, we created three pooled datasets of the SA/EE CVS (Table 1). Safety evaluations included in the pooled analyses are based on variables measured in the same way for the study groupings. Investigators at study sites assessed relatedness of adverse events to CVS use and graded relatedness as possibly, probably, or definitely/highly probably according to the definitions recommended in the protocol (Table 1S).
Table 1.
Summary of SA/EE contraceptive vaginal system (CVS) Safety Analysis Studies
| Population / Analysis Set | Product Studied | Doses Studied (mcg/day) | Total Subjects |
|---|---|---|---|
| All-Dose SA/EE CVS | Phase 3a and Pilot SA/EE CVS | 50/20 150/13 150/20 200/13 |
2,843 |
| All SA/EE 150/13b CVS | Phase 3a and Pilot SA/EE CVS | 150/13 | 2,569 |
| Phase 3 Trials SA/EE 150/13b CVS | Phase 3 SA/EE CVSa | 150/13 | 2,308 |
SA = segesterone acetate
EE = ethinyl estradiol
CVS = Contraceptive vaginal system
Formulation intended for marketing
The in-vitro release data initially supported an average daily release of 150/15 micrograms per day for SA and EE, respectively. Estimates from an ex-vivo study results indicate 150/13 micrograms per day for SA and EE, respectively.
The ALL DOSE SA/EE CVS DATASET(N=2843) and the SA/EE 150/13 CVS DATASET (N=2,569) are used as the denominator for program-wide serious adverse event (SAE) assessments; and the PHASE-3 TRIALS SA/EE CVS DATASET (N=2,308) are used as the denominator for the pivotal trials adverse events and labs assessments.
We performed all statistical analyses using SAS® version 9.2 or higher. We used descriptive statistics to summarize the variables in the safety summary.
3.0. RESULTS
The All-Dose SA/EE CVS dataset includes 2,843 women who used one of five SA/EE CVS doses from 1 to 13 cycles duration. Subject demographics for the Phase 3 (150/13 dose; n=2,308) analysis set are summarized in Table 2. Most (n=1,638, 71%) subjects in the analysis set identified their race as Caucasian. A majority of subjects (n=1609, 70%) were 20–29 years although the entire range included 18.1 to 40.8 years. About 2/3 (n=1,584, 69%) of women had a BMI of 20.1 to 27.0 kg/m2 at enrollment. Demographics for the All-Dose dataset had similar characteristics.
Table 2.
Demographics and Baseline Characteristics of SA/EE CVS users in clinical trials
| Characteristic | All Dose SA/EE CVS (n=2,843) | Phase 3 Trials SA/EE 150/13 CVS (n=2,308) |
|---|---|---|
| Age (years) | 27.0 ± 5.2 | 26.7 ± 5.1 |
| Age category | ||
| 18–19 | 166 (6) | 144 (6) |
| 20–24 | 992 (35) | 852 (37) |
| 25–29 | 914 (32) | 757 (33) |
| 30–35 | 553 (19) | 393 (17) |
| ≥36 | 218 (8) | 162 (7) |
| Parity | ||
| 0 | 1670 (59) | 1478 (64) |
| ≥1 | 1173 (41) | 830 (36) |
| Region | ||
| USA | 1723 (61) | 1536 (67) |
| Europe | 411 (14) | 309 (13) |
| Non-USA and non-Europe | 709 (24.9) | 463 (20.1) |
| Weight (kg) | 64.5 ± 11.0 | 64.3 ± 10.6 |
| BMI (kg/m2) | 24.3 ± 3.9 | 24.1 ± 3.74 |
| < 20.0 | 317 (11) | 251 (11) |
| 20.1 to 25.0 | 1485 (52) | 1245 (54) |
| 25.1 to 27.0 | 414 (15) | 339 (15) |
| 27.1 to 29.0 | 320 (11) | 264 (11) |
| > 29.0 | 307 (11) | 209 (9) |
| Race | ||
| Asian | 82 (3) | 82 (4) |
| African Ancestry | 328 (12) | 328 (14) |
| Caucasian | 1670 (59) | 1638 (71) |
| Other | 250 (9) | 248 (11) |
| Unknown | 513 (18) | 12 (<1) |
| Ethnicitya | ||
| Hispanic or Latina | 701 (25) | 690 (30) |
| Not Hispanic or Latina | 1641 (58) | 1618 (70) |
| Unknown | 501 (18) | 0 |
SA = segesterone acetate; EE = ethinyl estradiol; CVS = Contraceptive vaginal system;
All data are presented as n (%) or mean ± standard deviation
The analysis sets are defined in Table 1’ All SA/EE 150/13 CVS Analysis Set includes all subjects in the Phase 3 Trial Analysis Set and is a subset of the All Dose SA/EE CVS Analysis Set.
Ethnicity was captured on the CRF using the word ‘Latino’ for Phase 3 studies, only ‘not Hispanic’ or ‘Hispanic’ was captured on the CRF of Phase 2 studies. Ethnicity was not captured for any other study.
Subject disposition in the All-Dose and the Phase 3 trials analysis sets are shown in Table 3. Among the 2,308 women enrolled in the Phase 3 trials, 1332 (58%) women completed the study, and 999 of these women completed 13 cycles (one-year) of use. Of the 976 (42%) subjects who discontinued early, the most common reasons related to withdrawal of consent or lost to follow-up (n=411, 18%) and an AE (n=281, 12%). The change in eligibility related to BMI resulted in 147 (6%) women being discontinued early. Overall, investigators enrolled only 209 women with BMI>29.0 kg/m2 of whom 36 (17.2%) completed 13 treatment cycles. The disposition of subjects in the supportive analysis sets was generally similar.
Table 3:
Subject disposition in the Phase 3 trials (n=2,308) analysis set of women using the SA/EE contraceptive vaginal system, and Reasons for discontinuation.
| Disposition | All Dose SA/EE CVS (N=2,843) | Phase 3 SA/EE 150/13 CVS (N=2,308) | Phase 3 SA/EE 150/13 CVS BMI Subgroups | |
|---|---|---|---|---|
| <29.0 kg/m2 (N=2,099) | >29.0 kg/m2 (N=209) | |||
| Completed the Study | 1761 (62) | 1332 (58)* | 1296 (62) | 36 (17) |
|
Discontinued Prematurely |
1082 (38) | 976 (42) | 803 (38) | 173 (83) |
| Adverse Event** | 328 (12) | 281 (12) | 267 (13) | 14 (7) |
| Withdrew Consent | 228 (8) | 214 (9) | 191 (9) | 23 (11) |
| Lost to Follow-up | 220 (8) | 197 (9) | 184 (9) | 13 (6) |
| Eligibility*** | 147 (5) | 147 (6) | 31 (1) | 116 (56) |
| Pregnancy | 63 (2) | 54 (2) | 50 (2) | 4 (2) |
| Compliance | 62 (2) | 49 (2) | 47 (2) | 2 (<1) |
| Expulsions | 33 (1) | 33 (1) | 32 (2) | 1 (<1) |
| Other | 3 (<1) | 1 (<1) | 1 (<1) | 0 |
All data are presented as n (%).
SA = segesterone acetate; EE = ethinyl estradiol; CVS = Contraceptive vaginal system; BMI= Body Mass Index.
1332 completed the study and 999 completed the 13 cycles of CVS use. Not all completers who were enrolled late in the study completed 1 year due to the expiration of the drug
Most Common AE leading to discontinuation: headache 30(1.3%); nausea 28(1.2%); Vaginal discharge/vulvovaginal mycotic infection 30(1.3%); abdominal pain 30 (1.3%). Metrorrhagia/menorrhagia occurred in 7% of the subjects and 1.7% discontinued for this reason (n=39)
The Sponsor discontinued subjects with a BMI >29.0 kg/m2 during the study.
3.1. Overview of Adverse Events
In the Phase 3 trial analysis set, most (n=2008, 87%), reported at least one AE, the majority of which were either mild or moderate in severity based on protocol definitions (Table 1S). Table 4 presents a summary of the most common adverse reactions (occurring in >2% of subjects) and Table 2S (online supplement) presents AEs incidence based on BMI. Overall, 1602 (69%) subjects experienced AEs considered by the site investigator to be related to CVS use. Adverse events that resulted in study discontinuation occurred in 281 (12%) subjects.
Table 4.
Summary of Adverse Reactions Occurring in >2% of Subjects - Phase 3 trials SA/EE contraceptive vaginal system Analysis Set (N=2,308)
| Adverse Reaction | Subjects (%) with Adverse Reactions |
|---|---|
| Any event | 1,602 (69) |
| Headache | 601 (26) |
| Nausea | 420 (18) |
| Vaginal discharge | 242 (10) |
| Uterine spasm | 225 (10) |
| Vulvovaginal mycotic infection | 173 (7) |
| Metrorrhagiaa | 160 (7) |
| Breast tenderness | 134 (6) |
| Vomiting | 108 (5) |
| Genital pruritus female | 92 (4) |
| Urinary tract infection | 92 (4) |
| Migraine | 77 (3) |
| Vaginal candidiasis | 70 (3) |
| Libido decreased | 69 (3) |
| Acne | 66 (3) |
| Mood swings | 60 (3) |
| Dizziness | 58 (3) |
| Dyspareunia | 56 (2) |
| Vulvovaginal discomfort | 56 (2) |
| Withdrawal bleedb | 56 (2) |
| Breast pain | 49 (2) |
All data are presented as n (%).
Discontinuation rates for the common adverse reactions were: Headache including migraine n= 30 (1.3%); Vaginal discharge/vulvovaginal mycotic infections n=30 (1.3%); Nausea/vomiting n=28 (1.2%) and metrorrhagia/menorrhagia n=39 (1.7%)
SA = segesterone acetate; EE = ethinyl estradiol; CVS = Contraceptive vaginal system
Metrorrhagia includes verbatim terms of: irregular bleeding, irregular spotting, breakthrough bleeding, breakthrough spotting, spotting, frequent spotting, prolonged spotting.
Withdrawal bleed includes verbatim terms of: heavy bleeding with menses, bleeding very much, prolonged bleeding with menses, frequent bleeding with menses.
Among women in the All Dose dataset (N=2,843), 51 (2%) women experienced 55 Serious Adverse Events (SAEs). Sixteen SAEs occurred in more than one subject, including four VTEs, two allergic reactions and ten spontaneous abortions. In the Phase 3 trials, 47 SAEs occurred in 43 (2%) subjects. Investigators assessed 15 of these SAEs as possibly or probably related to the CVS. No SAEs occurred in more than one subject among 209 SA-only CVS participants. No deaths occurred in any study.
3.2. Venous Thromboembolic Events
Four non-fatal VTEs occurred at US study sites across the clinical program of the SA/EE CVS with a total exposure of 2,021 woman-years (WY); all four subjects recovered. These events included one pulmonary embolism, two deep vein thromboses, and one cerebral venous thrombosis that occurred in cycles two, three, six and seven of treatment, respectively. Three of these cases occurred in women with risk factors for VTEs: two subjects had high pre-treatment BMI (29.1 and 30.8 kg/m2) and one 39-year-old woman had Factor V Leiden heterozygous mutation discovered after the event. The fourth case occurred in a 28-year-old woman with a BMI of 25.2 kg/m2 who withdrew from the study before a clotting evaluation could be conducted. She reported smoking but <10 cigarettes daily. The VTE rate for women with a BMI of <29.0 kg/m2 results in 10.8/10,000 women-years (95% CI 8.9, 13.1).
No VTE cases occurred among the 1,120 (48%) of subjects enrolled in non-US sites. Comparing weight and BMI between non-US and US sites, subjects had a mean weight of 62.2 and 65.4 kg, respectively and a mean BMI of 23.6 and 24.4 kg/m2, respectively.
3.3. CVS Expulsions
Of the 2,308 women in the Phase 3 studies, 2,096 (90.8%) provided evaluable diary responses regarding expulsion. Overall, 1107 (52.8%) subjects reported at least one complete (24.6%) or partial expulsion (44.0%) during CVS use. Users reported CVS expulsion most frequently in the first cycle of use; among 2,050 women who documented whether expulsion occurred during the first cycle, 190 (9.3%) and 499 (24.3%) experienced complete and partial expulsions, respectively. When including all 21,842 treatment cycles eligible for analysis of expulsions, complete expulsion occurred in 1509 (7.0%) and partial expulsion in 4259 (19.8%) women. Only 5.5% of women required a replacement system due to a lost CVS.
3.4. Clinical Laboratory Values
No changes in the Clinical Laboratory values from Baseline to End-of-treatment occurred in 57% to 80% of the subjects according to the variables measured. Minimal, not clinically relevant changes in BUN, creatinine, hemoglobin and lipids occurred.
Thirteen (0.6%) subjects had elevations of a single liver function test above normal at baseline, 11 of which improved during CVS use. Fifteen (0.6%) other subjects developed non-clinically significant elevations during therapy for at least one liver function test; none resulted in discontinuation.
Subjects had a median change in total cholesterol from Baseline to End-of-study of +0.23 mmol/L. The median changes from Baseline to End-of-study are +0.16 mmol/L for High Density Lipoprotein (HDL), 0 mmol/L for Low Density Lipoprotein (LDL) and +0.15 mmol/L for triglycerides.
Mean hemoglobin and hematocrit values remained stable during the Phase 3 clinical studies. The mean change in hemoglobin from Baseline to End-of-study was −2.7 ± 8.3 g/L.
3.5. Vital Signs and Physical Exam
Table 5 summarizes vital signs for the Phase 3 clinical studies. Most subjects had normal vital signs at both assessments. Mean weight change from Baseline to End-of-study was 0.4 ± 3.3 kg. The mean change from baseline in BMI was 0.1 ± 1.2 kg/m2. The investigators did not report any weight changes as clinically relevant, but 35 women reported weight increase as an AE. Eight women reported hypertension as an AE and one of those discontinued. Investigators also reported no clinically relevant changes in physical exam data from Baseline to the End-of-Study.
Table 5.
Change from Baseline of Vital Signs Data- SA/EE 150/13 CVS Analysis Set (N = 2,308).
| Vital Signs | Baseline Mean (SD) | Median Change (Min/Max) |
|---|---|---|
| Diastolic BP (mmHg) | 68.7 (7.7) | 0.5 (−28/30a) |
| Systolic BP (mmHg | 108.4 (10.1) | 1.0 (−37/40a) |
| Pulse (bpm) | 73.9 (10.3) | 2.0 (−42/47) |
| Weight (kg) | 64.3 (10.6) | 0.2 (−17b/17.7c) |
| Body Mass Index (kg/m2) | 24.1 (3.7) | 0.09 (−6.6/6.6) |
SA = segesterone acetate; EE = ethinyl estradiol; CVS = contraceptive vaginal system
BP = blood pressure; Min= minimum; Max=maximum
Eight subjects reported hypertension as an AE and four of them received short term antihypertensive treatment. Only one of these discontinued for hypertension of 132/102
Three (3) subjects reported weight decrease as an AE
Thirty-five (35) subjects reported weight increase as an AE, 9 of these discontinued for the weight increase of 4–10 kg and 1 of these was discontinued for BMI >29.0
3.6. Endometrium
A multicenter endometrial sub-study enrolled 156 women during the Phase 3 study to evaluate endometrial safety of the SA/EE 150/13 CVS with 13 cycles of use. Subjects had a baseline endometrial biopsy between Days 7 to 14 of the screening cycle for women not currently using CHC or at any time prior to initiating the CVS for women using CHC. The investigators excluded women whose screening biopsies showed hyperplasia (two) or other anomalies (one endometritis and four polyps). In total, 83 subjects had follow-up biopsies at the end of CVS treatment, with 27 subjects sampled at Cycles 5 to 6; 25 subjects at Cycles 7 to 11; and 31 subjects after 12 to 13 cycles of therapy. Three, blinded independent Board-Certified pathologists read each of the biopsies and reported no cases of endometrial hyperplasia or carcinoma. Secretory and Atrophic/Inactive tissue were the most common diagnoses.
DISCUSSION
The large database collated from the SA/EE CVS clinical trials demonstrate a safety profile as expected for a CHC containing EE. We found no obvious safety signals in the data for standard laboratory chemistry, hematology, vital signs or physical exams.
Among the most common adverse reactions reported, headache and nausea are also reported with other CHCs. Whereas mood disorders are reported with CHC use up to 24% [10], only 3% of CVS users complained of mood swings.
In addition to the data presented here, three other safety studies have also been published. Results of the colposcopic examinations from a 13-cycle, multicenter, randomized, dose-finding study of the pilot rings did not raise safety issues. In that study, 150 healthy, reproductive-age women had cervical and vaginal colposcopic examinations conducted pre-treatment and during cycles three, thirteen, and at the end of study participation (including early termination), with similar findings across dosing groups and no safety concerns [8,9].
A vaginal drug-drug interaction study demonstrated that co-administration of miconazole suppositories with the CVS led to higher systemic exposure of both SA and EE, while co-administration with miconazole cream did not affect hormone levels [6]. Women using the SA/EE CVS may be advised to use an oral formulation or miconazole cream rather than suppository to treat vaginal candidiasis.
A single-center microbiology sub-study of vaginal flora and the risk of vaginal infection associated with use of the SA/EE 150/13 CVS showed that women using the CVS on a 21-day in and 7-day out schedule for up to 13 consecutive cycles did not have any increase in vaginitis or clinically significant change in vaginal flora [4]. Cultures of the used CVS yielded a similar prevalence and concentration of microorganisms as observed in the cultures from vaginal fluid.
With any CHC containing EE, one of the most concerning serious adverse events is VTE. In our safety database of 2,843 women who used the SA/EE CVS in clinical trials, we observed four cases of VTE in cycles two, three, six and seven of use. Two events occurred in women with a BMI >29.0 kg/m2 and two in women with a lower BMI. For women with BMI ≤29.0 kg/m2, the estimated VTE risk with SA/EE CVS use is 10.8/10,000 women-years. This rate is similar to other non-oral CHC including vaginal and transdermal products (3–12 cases per 10,000 WY) [11].
Pregnancy increases the risk of VTE as much or more than combined oral contraceptives; this finding is in accordance with current FDA-approved CHC labeling [15]. Indeed, the overall incidence of pregnancy-associated VTE is 5 to 65 cases per 10,000 WY [14]. Thus, the balance between the benefits and risks of contraceptive steroids is generally positive, particularly when compared with pregnancy’s risks [14]. Data are conflicting, with prospective active surveillance studies reporting no difference in VTE risk among different CHC formulations [11] while observational and database studies suggest an increase in risk with some combinations [14]. The discrepancy may be explained by different study designs and the fact that some observational studies do not adjust for important risk factors such as overweight, family history of thrombosis, and smoking. Indeed height, weight, and obesity are independent risk factors for VTE and this risk is further enhanced in obese women using CHCs [13].
No VTEs occurred in the non-US subgroup, representing 48% of the total population with a lower mean weight and BMI. It is noteworthy that many previous Phase 3 trials of approved CHCs excluded women with BMI>29 kg/m2. A trend for increased BMI has occurred over the recent decades, mostly in the US [12]. Also, we had no VTEs in our studies of SA-only formulations including 1,463 subjects with 1,558 WY of exposure, suggesting no risk attribution to the progestin SA. As a result of discontinuing enrollment of women with BMI >29.0 kg/m2, we significantly reduced the sample size of obese subjects completing 13 cycles thereby limiting the certainty of the safety conclusions in this subgroup. The estimated rate of VTEs among women with a BMI >29.0 kg/m2 cannot be established due to the small number of women in this group.
Three safety studies evaluated the effect of the SA/EE CVS and of the vaginal delivery of EE on hepatic protein synthesis. In one already published sub-study [5], we found that CVS use for up to 13 cycles resulted in small changes from baseline in plasma levels of factor VIII, fibrinogen and Protein S (PS), all within the normal range. However, SHBG levels increased above the normal range by Cycle 6. Women not using CHCs before CVS showed wider changes in values versus recent CHC users whose baseline values were increased by previous EE exposure [5].
In another comparative 3-month study of the SA/EE 150/13 CVS versus the LNG/EE 150/30 OC some measured hemostasis variables and SHBG were affected differently between groups, most likely because of difference in androgenicity of the progestins. A higher increase of SHBG occurred in users of SA/EE than in users of LNG, a progestin transactivating the AR while SA does not [3,16,17]. SHBG is a marker sensitive to both estrogen and androgen action but its serum levels are not causally related to venous thrombosis risk in women not using hormonal contraceptives, hence is not a VTE risk marker [18,19]. For the eight coagulation parameters, the mean values for the pilot SA/EE CVS and LNG/EE OC treatment groups were within the reference range for the laboratory with the exceptions of Factor VII-t and VIII, which were slightly elevated in the CVS treatment group. It is noteworthy that the overall measures of coagulation activity (clot formation and degradation), Prothrombin Fragment 1+2, and D-dimer were within their reference ranges for both the CVS and the OC treatments. The most sensitive factor to the isolated effects of EE was angiotensinogen and there was no difference between oral and vaginal treatments [17].
In a third randomized, cross-over, single-center study, 14 postmenopausal women used a pilot EE-only vaginal ring and 13mcg of oral EE. There was no difference observed between the systemic impact of 21 days of exposure from oral versus vaginal delivery [7]. Similarly, there were no differences between the routes of EE administration and values of angiotensinogen, SHBG or a panel of clotting factors. These data indicate that the well-established effects of EE-containing CHCs on hemostatic variables and estrogen-sensitive liver proteins are largely related to EE and independent of delivery route during short-term treatment. The precautions observed when prescribing an oral contraceptive containing EE would also apply to the CVS releasing EE.
Another estrogen-related issue of note is gallbladder disease. In the overall CVS analysis set (n=2,843) five subjects reported 6 events of gallbladder disease. Four of these events occurred during treatment, one of which included treatment with cholecystectomy. Female sex hormones, and specifically hormones in CHC products, are established risk factors for gallbladder disease. Estrogens increase cholesterol secretion and reduce bile salt secretion and progestins reduce bile salt secretion and impair gallbladder emptying, leading to bile stasis thereby promoting gallstone formation [20].
In conclusion, the overall safety database related to the SA/EE CVS indicates a safety profile that we would expect based on what is established with existing CHCs, but limited data are available in obese women. The marketed product Annovera™ (TherapeuticsMD, Boca Raton, FL) will have the same precautions as required for other EE-containing CHC. Women with known risk factors for VTE, such as high BMI, first degree relative with VTE or thrombophilia are at risk of thrombosis that may be enhanced by any method containing EE.
Supplementary Material
ACKNOWLEDGMENTS
The authors acknowledge the major contribution of Dr Daniel R Mishell Jr, (deceased), from the Department of Obstetrics and Gynecology, University of Southern California, Keck School of Medicine, Los Angeles, CA 90033, USA, who invented the concept of the vaginal system to deliver contraceptive steroids and conducted several of the studies analyzed here while he was a member of the International Committee for Contraceptive Research (ICCR) of the Population Council. The authors also acknowledge the contribution of Dr Horacio B Croxatto, from the University of Chile, who established the clinical center in Chile and participated in all pivotal clinical studies while he was a member of the ICCR.
The authors acknowledge the other principal investigators in the Phase 3 clinical trials: David F Archer, Norfolk, VA (USA); Luis Bahamondes Campinas Univ (Brazil); Carolyn Westhoff Columbia University, NY (USA); Dan Apter (Finland); Philip Darney, UCSF, CA (USA); Jeffrey Jensen OHSU, OR (USA); Anita Nelson UCLA-Harbor, CA (USA); Vivian Brache, Profamilia, (Dominican Republic); George Bartfai, Szeged University (Hungary); Erika Banks, Albert Einstein Medical Center, NY (USA); Livia Wan, New York University, NY, (USA); David. Portman and Lisa Keder Ohio State University, Columbus, OH (USA); William Schlaff, University of Colorado, Aurora, CO (USA); Ronald Burkman Bay State Medical Center, Springfield, MA (USA); Ken Muse, University of Kentucky, Lexington, KY (USA).
The authors would like to thank the women who volunteered for these studies, the participating study investigators and coordinators at the 27 clinical sites for conduct of the Phase 3 clinical studies. The Population Council and the NICHD co-sponsored the Phase 3 clinical studies; and collaborated in the study design together with FDA reviewers when submitted to the Population Council IND (and NDA 209,627). Both Sponsors collaborated in the collection, analysis and interpretation of data and in the writing of the report. Both conferred and agreed to submit this article for publication.
FUNDING
This work was supported by the National Institute of Child Health and Human Development of the National Institutes of Health Contract Numbers HHSN27500403366, HHSN27500403371, HHSN27500403372, HHSN27500403373, HHSN27500403374, HHSN27500403375, HHSN27500403376, HHSN27500403377, HHSN27500403378, HHSN27500403379, HHSN27500403380, HHSN27500403381, HHSN27500403382), and the United States Agency for International Development (Grant Number GPO-A-00-04-00019-00), as well as financial support from WHO’s RHRP Division for three clinical sites in Europe, and the Population Council.
REFERENCES
- [1].Sitruk-Ware R, Small M, Kumar N, Tsong YY, Sundaram K, Jackanicz T. Nestorone®: Clinical applications for contraception and HRT. Steroids 2003;68:907–13. [DOI] [PubMed] [Google Scholar]
- [2].Kumar N, Koide SS, Tsong Y, Sundaram K. Nestorone: a progestin with a unique pharmacological profile. Steroids 2000;65:629–36 [DOI] [PubMed] [Google Scholar]
- [3].Kumar N, Fagart J, Liere P, Mitchell SJ, Knibb AR, Petit-Topin I, et al. Nestorone (NES) a Novel Progestin for Nonoral Contraception: Structure-activity Relationships and Brain Metabolism Studies. Endocrinology 2017;158:170–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Huang T, Merkatz RB, Hillier SL, Roberts K, Blithe DL, Sitruk-Ware R, et al. Effects of a one-year reusable contraceptive vaginal ring on vaginal microflora and the risk of vaginal infection: an open-label prospective evaluation. PLOS ONE August 12, 2015. doi: 10.1371/journal.pone.0134460 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Archer DF, Thomas MA, Conard J, Merkatz RB, Creasy GW, Roberts K, et al. Impact on hepatic estrogen-sensitive proteins by a 1-year contraceptive vaginal ring delivering Nestorone® and ethinyl estradiol. Contraception 2016;93:58–64. [DOI] [PubMed] [Google Scholar]
- [6].Simmons KB, Kumar N, Plagianos M, Roberts K, Hoskin E, Han L, et al. Effects of concurrent vaginal miconazole treatment on the absorption and exposure of Nestorone® (segesterone acetate) and ethinyl estradiol delivered from a contraceptive vaginal ring: a randomized, crossover drug-drug interaction study. Contraception 2018;97:270–6. [DOI] [PubMed] [Google Scholar]
- [7].Sitruk-Ware R, Plu-Bureau G, Menard J, Conard J, Kumar S, Thalabard JC, et al. Effects of oral and transvaginal ethinyl estradiol on hemostatic factors and hepatic proteins in a randomized, crossover study. J Clin Endocrinol Metab 2007;92:2074–9. [DOI] [PubMed] [Google Scholar]
- [8].Sivin I, Mishell DR Jr, Alvarez F, Brache V, Elomaa K, Lahteenmaki P, et al. Contraceptive vaginal rings releasing Nestorone and ethinylestradiol: a 1-year dose-finding trial. Contraception. 2005;71(2):122–9. [DOI] [PubMed] [Google Scholar]
- [9].Fraser IS, Lacarra M, Mishell DR, Alvarez F, Brache V, Lähteenmäki P, et al. Vaginal epithelial surface appearances in women using vaginal rings for contraception. Contraception 2000;61:131–8. [DOI] [PubMed] [Google Scholar]
- [10].Lundin C, Danielsson KG, Bixo M, Moby L, Bengtsdotter H, Jawad I, Marions L, Brynhildsen J, Malmborg A, et al. Combined oral contraceptive use is associated with both improvement and worsening of mood in the different phases of the treatment cycle-A double-blind, placebo-controlled randomized trial. Psychoneuroendocrinology. 2017;76:135–143. [DOI] [PubMed] [Google Scholar]
- [11].Dinger J, Möhner S, Heinemann K. Cardiovascular risk associated with the use of an etonogestrel-containing vaginal ring. Obstet Gynecol. 2013;122:800–8. [DOI] [PubMed] [Google Scholar]
- [12].May AL, Freedman D, Sherry B, Blanck HM; Centers for Disease Control and Prevention (CDC). Obesity - United States, 1999–2010. MMWR Suppl 2013;62:120–8. [PubMed] [Google Scholar]
- [13].Pomp ER, le Cessie S, Rosendaal FR, Doggen CJ. Risk of venous thrombosis: obesity and its joint effect with oral contraceptive use and prothrombotic mutations. Br J Haematol 2007;139:289–96. [DOI] [PubMed] [Google Scholar]
- [14].Heit JA, Spencer FA, White RH. The epidemiology of venous thromboembolism. J Thromb Thrombolysis 2016;41:3–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].FDA Medical Review of Quartette, page 60–62 at: https://www.accessdata.fda.gov/drugsatfdadocs/nda/2013/204061quartettetoc.cfm
- [16].Rad M, Kluft C, Menard J, Burggraaf J, de Kam ML, Meijer P,et al. Comparative effects of a contraceptive vaginal ring delivering a nonandrogenic progestin and continuous ethinyl estradiol and a combined oral contraceptive containing levonorgestrel on hemostasis variables. Am J Obstet Gynecol 2006;195:72–7. [DOI] [PubMed] [Google Scholar]
- [17].Sitruk-Ware RL, Menard J, Rad M, Burggraaf J, de Kam ML, Tokay BA, et. l. Comparison of the impact of vaginal and oral administration of combined hormonal contraceptives on hepatic proteins sensitive to estrogen. Contraception 2007;75:430–7 [DOI] [PubMed] [Google Scholar]
- [18].Stegeman BH, Helmerhorst FM, Vos HL, Rosendaal FR, Van Hylckama Vlieg A. Sex hormone–binding globulin levels are not causally related to venous thrombosis risk in women not using hormonal contraceptives. J Thromb Haemost 2012;10:2061–7. [DOI] [PubMed] [Google Scholar]
- [19].Sitruk-Ware R. Hormonal contraception and thrombosis. Fertil Steril 2016;106:1289–94. [DOI] [PubMed] [Google Scholar]
- [20].Tierney S, Nakeeb A, Wong O, Lipsett PA, Sostre S, Pitt HA, et al. Progesterone alters biliary flow dynamics. Ann Surg 1999;229:205–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].van den Heuvel MW, van Bragt AJ, Alnabawy AK, Kaptein MC. Comparison of ethinylestradiol pharmacokinetics in three hormonal contraceptive formulations: the vaginal ring, the transdermal patch and an oral contraceptive. Contraception 2005;72:168–74. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.