RESUMEN
OBJECTIVES: To determine the incidence of out-of-range segesterone acetate (NES) concentrations in participants of a pharmacokinetic/pharmacodynamic trial of a continuous use contraceptive vaginal ring (CVR) releasing NES and estradiol (E2). We hypothesized that out-of-range concentrations reflect nonadherent ring use and predict ovulation risk. STUDY DESIGN: We conducted a secondary analysis of data from a prospective, multi-centered, randomized, Phase IIa dose-finding trial for a CVR releasing NES and E2. Our primary outcome was the risk of ovulation associated with out-of-range NES events. We calculated the 5th and 95th percentile NES concentrations of subjects at steady state to determine high and low cutoffs. We used a Fisher's exact test to determine group differences, and calculated the relative risk of ovulation for each group. RESULTS: We analyzed available serum NES data from cycles 2 (n = 172), 3 (n = 156) and 7 (n = 115) to determine the 5th and 95th percentile of all NES concentrations (64, 296 pg/mL). In the 443 cycles of observation, no ovulations occurred in participants with NES concentrations within the expected range. In contrast, we found ovulatory elevations of progesterone in 21 cycles with out-of-range values. Of these, 15 (71%) cycles had evidence of one or more nonadherent low and 6 (29%) one or more unexpected peak. The relative risk of ovulation increased with evidence of multiple non-adherent levels. CONCLUSIONS: We found out-of-range NES concentrations, suggestive of improper use of a CVR associated with an increased risk of ovulation, with a direct relationship between the number of out-of-range events and the relative risk. IMPLICATIONS: The results of this study support the use of out-of-range serum NES values as a marker of adherence in contraceptive clinical trials of continuous vaginal rings, and suggest that nonadherence occurs even in early phase clinical trials with close monitoring.