Considerations of safety for women prescribed apixaban or rivaroxaban who breastfeed-a physiologically based pharmacokinetic analysis.

BACKGROUND

The risk of venous thromboembolism is significantly increased during the perinatal period. The safety of the direct oral anticoagulants, apixaban and rivaroxaban, has not been established in women who breastfeed. Physiologically based pharmacokinetic (PBPK) modeling provides a means to understand drug disposition and is increasingly being used to predict infant exposure to maternal medication via breastmilk.

OBJECTIVES

To utilize established PBPK models of apixaban and rivaroxaban to simulate infant exposure to maternal ingestion.

METHODS

Existing PBPK models for apixaban and rivaroxaban were adapted to the lactation setting. These models were verified using published reports of apixaban and rivaroxaban from breastfeeding women. A virtual postpartum population was then generated to simulate maternal and infant exposure to rivaroxaban and apixaban at different doses and time points during the postpartum period. Work was conducted on Simcyp Simulator (v23).

RESULTS

The adapted apixaban and rivaroxaban PBPK models captured the published breastfeeding clinical data well. Apixaban transferred into human breastmilk significantly more than rivaroxaban. Steady state infant plasma AUC24-48 to a fully breastfed infant following maternal dose of rivaroxaban 20 mg daily was 242 (±107) ng/mL·h in the immediate postpartum period and for apixaban 5 mg twice a day AUC24-48 was 2041 (±466) ng/mL. Maternal exposure to apixaban and rivaroxaban was lower immediately postdelivery and then normalized by 6 weeks postpartum.

CONCLUSION

Our work demonstrates that apixaban would not be a suitable agent for breastfeeding women during the postnatal period, while rivaroxaban shows significant promise due to low infant exposure and warrants further investigation.