Barrier contraception should always be used in combination with other methods of contraception (for example, oral or other hormonal contraceptives). Because of the long half-life of efavirenz, use of adequate contraceptive measures for 12 weeks after discontinuation of efavirenz is recommended.

Females of reproductive potential should use effective contraception during treatment with efavirenz and for 12 weeks after discontinuing efavirenz due to the long half-life of efavirenz. Barrier contraception should always be used in combination with other methods of contraception. Hormonal methods that contain progesterone may have decreased effectiveness.

A pharmacokinetic study evaluated whether doubling the dose of levonorgestrel for emergency contraception (EC) to 3 mg could compensate the reduction in levonorgestrel Cmax and AUC when coadministered with efavirenz or rifampicin. A phase 2, open-label, multicentre, partially randomized, four parallel group trial was conducted in pre-menopausal females. Participants on dolutegravir (no inducing effects) received levonorgestrel 1.5 mg (control group); those on rifampicin-containing tuberculosis therapy received levonorgestrel 3 mg; those on efavirenz-based antiretroviral therapy were randomized 1:2 to levonorgestrel 1.5 mg or 3 mg. Plasma was collected through 48h post-dose to assess levonorgestrel pharmacokinetics. The area under the concentration-time curve (AUC) over 8 hours was the primary outcome. Levonorgestrel pharmacokinetic parameters were compared between groups using geometric mean ratios (GMR) with 90% confidence intervals. The study showed that participants receiving levonorgestrel 1.5 mg plus efavirenz (n=17) had 50% lower AUC0-8h compared to the control group (n=32) [0.50 (0.40, 0.62)]. Participants receiving levonorgestrel 3 mg had a similar AUC0-8h when receiving either efavirenz (n=35) [0.99 (0.81, 1.20)] or rifampicin (n=34) [1.16 (0.99, 1.36)] compared to control. Levonorgestrel 3 mg resulted in similar or higher maximum concentration with either efavirenz [1.17 (0.96, 1.41)] or rifampicin [1.27 (1.09, 1.49)] compared to the control group. Thus, this study demonstrates that doubling the levonorgestrel-EC dose allows to compensate both the inducing effect of efavirenz and rifampicin.

Levonorgestrel is primarily metabolized by CYP3A4. Efavirenz and rifampicin induce CYP3A4 while isoniazid inhibits CYP3A4. A previous PK study showed that a double-dose of levonorgestrel (3 mg vs 1.5 mg) offsets the effects of efavirenz and rifampicin on levonorgestrel pharmacokinetics over the first 8 hours after a single dose for emergency contraception. This follow-up study aimed at assessing the impact of pharmacogenetics on these drug interactions. Women living with HIV on efavirenz based antiretroviral treatment were randomized 1:2 to levonorgestrel 1.5 mg (group A) or 3 mg (group B). Of 122 cisgender women, 118 (97%) were valuable for genetic associations: 73 (62%) identified as Black, 33 (28%) Asian, and 10 (8%) Latina. CYP2B6 genotypes associated with higher efavirenz exposure were associated with more rapid levonorgestrel clearance (B=0.29, p= 0.025). NAT2 genotypes that increase plasma isoniazid exposure had slower levonorgestrel clearance (B=0.31, p= 0.022). In summary, the pharmacogenetics of efavirenz and isoniazid was shown to affect interactions with levonorgestrel emergency contraception. CYP2B6 slow metabolizer genotypes worsen the efavirenz-levonorgestrel interaction, likely due to increased CYP3A4 induction with higher efavirenz exposure, making this interaction more difficult to overcome. NAT2 slow acetylator genotypes were shown to attenuate the rifampicin-levonorgestrel interaction, likely due to increased CYP3A inhibition as a result of higher isoniazid exposure.

Coadministration of efavirenz (600 mg once daily) and a single oral dose of levonorgestrel (0.75 mg) to 21 HIV-negative women significantly reduced levonorgestrel AUC, Cmax and Cmin by 58%, 45% and 69%, respectively.