A UHPLC-MS/MS method to simultaneously quantify apixaban, edoxaban and rivaroxaban in human plasma and breast milk: For emerging lactation studies

Highlights

• How much direct oral anticoagulants (DOACs) transfer into human breast milk is unknown.

• First report of an analytical method for quantification of DOACs in breast milk.

• The method and assay developed is precise and accurate in human plasma and milk.

• The assay can now be used for lactation studies involving DOACs.

Abstract

Clinical studies are needed to clarify the use of direct oral anticoagulants (DOACs) in breastfeeding women. To support emerging clinical studies on investigating DOAC’s transfer into breast milk, an ultra-high-performance liquid chromatography/tandem mass spectrometry (UHPLC‐MS/MS) method was developed and validated for quantifying three DOACs – apixaban, edoxaban and rivaroxaban in human plasma and breast milk. Protein precipitation with methanol was performed for sample preparation. Chromatographic analysis was performed using a C18 column. The MS detection was performed in MRM mode. The method was validated in accordance with the European Guideline (EMA). The calibration range was 5–500 ng/mL in plasma and 5–250 ng/mL in breast milk. The within-batch and between-batch variability remained <9%. Recoveries ranged from 106.13% to 109.05% in plasma and from 93.40% to 107.91% in breast milk. The lot-to-lot matrix variability was within ±15% among a range of samples originating from many different subjects. All analytes were stable when stored for 24 h at room temperature, 7 days at 2–8 °C, and at least 5 weeks at −20 °C in both plasma and breast milk. The developed method fulfilled the EMA bioanalytical method validation guideline and was shown to be simple, fast, accurate and will now be used in a clinical trial evaluating the transfer of apixaban and rivaroxaban into human breast milk.

Introduction

Venous thromboembolism (VTE) is a global disease with high incidence, prevalence, morbidity and mortality [1]. It is well established that the risk of VTE is elevated during pregnancy and postpartum period. Women’s relative risk of VTE increases by 5-fold during pregnancy, rising to as much as 84-fold during postpartum period [2], [3]. In the UK, pulmonary embolism (PE) has remained a leading direct cause of maternal death for over 20 years [4], [5]. Many international obstetric guidelines recommend appropriate thromboprophylaxis in women identified at risk during postpartum period [6], [7], [8], [9], [10]. Traditionally, vitamin K antagonist (VKA) such as warfarin and parental anticoagulants such as low molecular weight heparin (LMWH) have been the golden standards for the prophylaxis and treatment of VTE during the postnatal period, as both classes of agents are known to be safe to the infant, during breastfeeding [11]. However, VKA has limitations in that they require monitoring and dose adjustment, the practicalities of which can be challenging for a mother with a newborn [12]; LMWH has the disadvantage of having to be administrated as an injection, and in some women adherence to LMWH is reported to drop during the postpartum period relative to the antenatal period [13]. During the last decade, following their availability in clinical practice, direct oral anticoagulants (DOACs) have increasingly been prescribed in clinical practice because they have the advantages of being oral agents, a quick onset of action, and a predictable dose/response relationship and anticoagulant activity, negating the requirement for monitoring [14]. There are currently four DOACs approved by the European Medicines Agency (EMA): the direct thrombin inhibitor dabigatran and the factor Xa (FXa) inhibitors rivaroxaban, apixaban, and edoxaban. Currently, DOACs are contraindicated in lactating women due to limited data on the extent of their transfer into human breast milk, and the unknown clinical implications for nursing infants [15], [16].

The FDA has highlighted the need for clinical lactation studies [17], [18], [19], [20], [21]. A recent completed clinical trial involving 2 breastfeeding women in the UK, the Dalmation study, indicated that a small amount of dabigatran was excreted in breast milk (milk to plasma ratio (M/P): 0.02–0.1) [22]. For the other 3 DOACs – apixaban, edoxaban and rivaroxaban, the manufacturers found that they were excreted into rat milk, however, to the best of our knowledge no clinical trials has been done for the FXa inhibitors in breastfeeding women [23], [24], [25]. It is therefore necessary to quantify how much these 3 FXa inhibitors are excreted into human breast milk and their pharmacokinetic profiles in breastfeeding women through formal clinical lactation studies. To support this effort, a method for quantifying apixaban, edoxaban and rivaroxaban in plasma and breast milk is needed.

Human breast milk is a highly complex biological matrix which contains large amounts of lactose, fat and proteins, and the proportions of these are dynamic over the time of day, the day of lactation, from mother to mother, and among species [26], [27]. More specifically, the colostrum excreted at the very early stage of lactation contains more proteins and less fat compared to the mature milk excreted from 2 weeks after delivery; preterm milk contains more proteins and fat when compared to term milk; and the foremilk is lower in fat than towards the end of the feed [26]. These alterations can influence drugs’ passive diffusion to breast milk and ultimately affect their exposures to breastfed infants. To extract drugs and quantify their concentrations in breast milk with requisite precision and accuracy is thus an analytical challenge. A number of LC-MS methods for simultaneously quantifying specific several DOACs in human plasma have been reported [28], [29], [30], [31], but very few has been validated in human breast milk. One study utilised [¹⁴C]-apixaban to evaluate the excretion of apixaban into rat milk using a quantitative whole-body autoradiography method [32]. This method is normally used in laboratory animals but not in humans [33]. To date only 3 studies have reported bioanalytical methods for quantifying a DOAC in human breast milk [34], [35], [36]. However, the LC-MS/MS methods developed and validated in these studies were only applicable for quantification of rivaroxaban, and they did not comply with the formal guideline which requires tests for stability and matrix effects quantification [37]. So far, no study has reported the formal development and validation of a method for simultaneous quantification of all DOACs in human breast milk.

Our aim was to develop and validate an ultra-high-performance liquid chromatography and tandem mass spectrometry (UHPLC-MS/MS) assay for the simultaneous quantification of apixaban, edoxaban and rivaroxaban in human plasma and breast milk so that the extent of a DOAC’s excretion into breast milk compared with plasma in a proposed phase IV clinical trial could be undertaken [38].