Estradiol and Efavirenz: Interaction Details

a) In a systematic review evaluating drug-drug interactions (DDIs) between hormonal contraceptives (HCs) and antiretrovirals (N=49 studies), increased pregnancy rates or presumptive ovulation rates (based on luteal progesterone levels), or decreased progestin pharmacokinetic parameters occurred in 15 of 18 studies describing DDIs between efavirenz and various HCs including vaginal rings (containing ethinyl estradiol/etonogestrel), implants (containing levonorgestrel or etonogestrel), and emergency contraception (with oral levonorgestrel). Additionally, the presence of CYP2B6 single nucleotide polymorphisms enhanced this DDI in 5 studies. Included studies were generally of female patients of reproductive age who were not currently pregnant or breastfeeding and wanting to use hormonal contraception to prevent pregnancy. The quality of included studies was generally moderate. Limitations included the inability to control for confounding factors, and direct comparisons or meta-analyses were not possible due to diversity in outcomes. Efficacy and safety of antiretroviral treatment with efavirenz is not expected to be compromised by this drug interaction ³.

b) Concomitant use of a CYP3A4 inducer with oral or implantable contraceptive products containing levonorgestrel (684 events) or etonogestrel/desogestrel (864 events) was associated with a disproportionately higher rate of unintended pregnancy compared to all other event types reported to the FDA Adverse Event Reporting System (FAERS) between 1971 and 2020 [levonorgestrel (14,504 total events); etonogestrel/desogestrel (9348 total events)]. When compared between CYP3A4 inducer exposure vs no exposure, cases of unintended pregnancy made up a significantly higher proportion of total events when levonorgestrel was administered orally (32.8% vs 10.5%) or as an implant (51.9% vs 11.9%), and a similar association was identified with implanted etonogestrel products (42.5% vs 13.1%). However, when contraceptives were administered as an intrauterine device (levonorgestrel, 10.3% vs 11.5%) or intravaginal ring (etonogestrel, 10.9% vs 8.7%), no significant associations were identified. Oral desogestrel (pro-drug of etonogestrel) in combination with ethynyl estradiol also was not significantly affected (11.8% vs 17.4%). Intrauterine and vaginal ring products may be preferred in lieu of oral and implantable contraceptive products in women concomitantly receiving CYP3A4 inducers ¹.

c) In a pharmacokinetic study in healthy HIV-negative women (n=28; mean age, 26 years), coadministration of efavirenz (600 mg daily) and an oral contraceptive containing ethinyl estradiol (EE) and norgestimate (NGM) resulted in similar exposure for EE to that seen when given alone; however, exposure to progestin components (norgestimate and levonorgestrel) was significantly decreased. In this open-label, 3-period, 4-treatment study, participants received in period 1 (treatment A) Ortho Tri-Cyclen Lo(R) (EE 0.025 mg plus NGM 0.18 mg on days 1 to 7 (phase 1); EE 0.025 mg plus NGM 0.215 mg on days 8 to 14 (phase 2); and EE 0.025 mg plus NGM 0.25 mg on days 15 to 21 (phase 3)). This was followed by period 2 (days 29 to 56) where participants with acceptable baseline safety assessments received a full cycle of Ortho-Cyclen(R) (EE 0.035 mg plus NGM 0.25 mg (phases 1 to 3; treatment B)). Participants with satisfactory safety assessments began a second cycle of Ortho-Cyclen(R) (days 57 to 77; period 3) coadministered with efavirenz 600 mg/day for 14 days (days 57 to 70; treatment C). Cmax, AUC, and Cmin for EE were not significantly different when efavirenz was given concurrently. However, norgestimate exposure was significantly decreased in the presence of efavirenz. The adjusted geometric means for Cmax, AUC, and Cmin were reduced by 46% (90% CI, 39% to 52%), 64% (90% CI, 62% to 67%), and 82% (90% CI, 79% to 85%), respectively. Post-hoc analysis also showed similar results with levonorgestrel exposure (adjusted geometric means for Cmax, AUC, and Cmin reduced by 80% to 86% in the presence of efavirenz) ⁴.

d) In a pharmacokinetic study, concurrent administration of efavirenz and ethinyl estradiol/norgestimate resulted in markedly decreased exposure and plasma concentrations of norelgestromin and levonorgestrel, both active metabolites of norgestimate. Twenty-one subjects were administered ethinyl estradiol 0.035 mg/norgestimate 0.25 mg orally once daily concurrently with efavirenz 600 mg orally once daily for 14 days. Results indicated a 46% decrease in norelgestromin Cmax (90% CI, 39% to 52%), a 64% decrease in norelgestromin AUC (90% CI, 62% to 67%), and an 82% decrease in norelgestromin Cmin (90% CI, 79% to 85%). In 6 subjects, there was an 80% decrease in levonorgestrel Cmax (90% CI, 77% to 83%), an 83% decrease in levonorgestrel AUC (90% CI, 79% to 87%), and an 86% decrease in levonorgestrel Cmin (90% CI, 80% to 90%). Efavirenz and ethinyl estradiol pharmacokinetics were not altered ⁵.