Combined oral contraception: nomegestrol/estradiol (Zoely)

Table 1 Summary of efficacy

Mansour et al. (2011) and Westhoff et al. (2012)

Table 2 Summary of safety

Mansour et al. (2011) and Westhoff et al. (2012)

Clinical effectiveness

Safety and tolerability

The summary of product characteristics advises that, because no epidemiological data are currently available for combined oral contraceptives containing estradiol, the warnings of serious adverse events seen with other combined oral contraceptives (such as venous thromboembolism, other circulatory disorders, breast cancer and cervical cancer) are considered applicable to nomegestrol/estradiol.

According to the summary of product characteristics for Zoely, the most frequently reported adverse events in studies of nomegestrol/estradiol were acne and abnormal withdrawal bleeding (predominantly absence of withdrawal bleeding), which affected at least 1 in 10 women. Other common adverse events, affecting between 1 in 100 and 1 in 10 women, were decreased libido, depression or mood changes, headache or migraine, nausea, unscheduled bleeding or spotting, menorrhagia, breast or pelvic pain and weight increase.

In Mansour et al. (2011) and Westhoff et al. (2012) about 50% of women taking nomegestrol/estradiol had adverse events that were considered to be related to the contraceptive, compared with about 37% of women taking drospirenone/ethinylestradiol. The most frequently reported related adverse events (occurring in 5% or more women) were acne, irregular withdrawal bleeding and weight gain in both studies. According to the summary of product characteristics, in these 2 studies combined, acne was reported by 15.4% of women taking nomegestrol/estradiol compared with 7.9% of women taking drospirenone/ethinylestradiol; weight gain was reported by 8.6% of women taking nomegestrol/estradiol compared with 5.7% of women taking drospirenone/ethinylestradiol; and abnormal withdrawal bleeding (predominantly absence of withdrawal bleeding) was reported by 10.5% of women taking nomegestrol/estradiol compared with 0.5% of women taking drospirenone/ethinylestradiol. The significance of the differences between the groups was not reported. (See table 2 for data on adverse events in the individual studies.)

In both studies adverse events led to about 18% of women stopping nomegestrol/estradiol and 10% of women stopping drospirenone/ethinylestradiol. In Mansour et al. (2011) acne, irregular withdrawal bleeding and weight gain accounted for most of the differences in discontinuation rates between the 2 groups (see table 2 for more details). In Westhoff et al. (2012) statistically significantly more women taking nomegestrol/estradiol stopped treatment because of irregular withdrawal bleeding (3.8%, compared with 1.8% with drospirenone/ethinylestradiol; p<0.023). Differences between the contraceptives are not reported for acne and weight gain.

In both studies the presence of acne decreased over time with both contraceptives. The authors suggest that nomegestrol/estradiol may not cause acne, but acne may have been frequently reported because assessment for acne at each clinic visit had raised awareness. Also, differences between the 2 groups might be because drospirenone/ethinylestradiol has a positive effect on acne, rather than nomegestrol/estradiol having a negative effect.

The authors also suggest that reports of adverse effects and discontinuation because of irregular withdrawal bleeding in women taking nomegestrol/estradiol might have been reduced if healthcare professionals and women had been counselled that withdrawal bleeding might not occur.

In both studies the body weight of women taking nomegestrol/estradiol increased by 1 kg on average over 13 cycles. In comparison, women taking drospirenone/ethinylestradiol gained 0.35 kg in Mansour et al. (2011) (p<0.001) and 0.2 kg in Westhoff et al. (2012) (p value not reported). These results are consistent with a Cochrane review that concluded that combined hormonal contraceptives do not have a large effect on weight.

Venous thromboembolism was not reported by any women taking nomegestrol/estradiol in the studies. However, according to the summary of product characteristics, venous thromboembolic events have been reported during post-marketing use. A European Medicines Agency review of the risk of venous thromboembolism with combined hormonal contraceptives reports that available data for nomegestrol are insufficient for comparison with other progestogens, but further studies are ongoing or planned.

Evidence strengths and limitations

The 2 studies were well designed. However, although they were randomised, controlled and allocation was concealed, they do have limitations. Neither of the studies was blinded, which is a potential source of bias. An open-label design was chosen because the differences in the nomegestrol/estradiol and drospirenone/ethinylestradiol regimens (24/4 compared with 21/7) would lead to obvious differences in the timing of withdrawal bleeding and in the missed pill guidance given to the 2 groups. Both studies were funded by Merck Sharp & Dohme.

The inclusion and exclusion criteria of the studies were generally representative of the standard population used for evaluating contraceptives in healthy women of childbearing potential. However, no young women aged 12–17 years were included: both studies included women aged 18–50 years. The summary of product characteristics for Zoely states that no data on efficacy and safety are available for young women under 18.

The definition used for in-treatment pregnancies for the licensing of contraceptives in Europe is pregnancies with an estimated date of conception from the day of first intake of trial medication up to and including the day of last intake of trial medication, extended with a maximum of 2 days. In Westhoff et al. (2012) the extension period was 7 days rather than 2 days. However, in the European public assessment report for Zoely, the analyses for this study are based on a 2-day extension period and the results seen are similar to those reported in Westhoff et al. (2012).

The baseline characteristics of the women appear to be well balanced between the 2 groups in both studies. However, no statistical analysis was performed to determine the significance of any differences.

According to the European public assessment report for Zoely, baseline characteristics differed between the 2 studies with regards to race, ethnicity, weight, gynaecological and contraceptive history and socioeconomic background. This may be the reason for some of the differences between the study results, such as the Pearl Index (0.38 for nomegestrol/estradiol and 0.81 for drospirenone/ethinylestradiol in Mansour et al. 2011; 1.27 for nomegestrol/estradiol and 1.89 for drospirenone/ethinylestradiol in Westhoff et al. 2012) and discontinuation rates (28.2% for nomegestrol/estradiol and 23.4% for drospirenone/ethinylestradiol in Mansour et al. 2011; 40.7% for nomegestrol/estradiol and 37.9% for drospirenone/ethinylestradiol in Westhoff et al. 2012).

The total number of women who stopped taking the contraceptive was high in both groups in both studies. However, the European public assessment report for Zoely notes that the discontinuation rates are lower than planned in the sample size calculations. The difference between the discontinuation rates in the studies is mainly because of a higher percentage of 'withdrawal of informed consent' and 'lost to follow up' in Westhoff et al. (2012) than in Mansour et al. (2011).

The European public assessment report for Zoely suggests that women in Westhoff et al. (2012) adhered less to the contraceptive regimen than women in Mansour et al. (2011). In Westhoff et al. (2012) 27 women were excluded from the intention-to-treat analysis because of the limited credibility of their electronic data; in Mansour et al. (2011) only 2 women were excluded for this reason.

Mansour et al. (2011) was conducted in Europe, Asia and Australia and may be more relevant to UK practice than Westhoff et al. (2012), which was conducted in the United States, Canada, Argentina, Brazil, Chile and Mexico.