Apixaban causes a false increase in activated protein C resistance (APCR) ratios and possibly protein S activity.
To investigate whether this increase can mask a diagnosis of factor V Leiden (FVL) or protein S deficiency in an actual population of patients undergoing apixaban treatment and hypercoagulation testing.
During a 4.5-year period involving 58 patients, we compared the following 4 groups: heterozygous for FVL (FVL-HET)/taking apixaban, wild-type/taking apixaban, heterozygous for FVL/no apixaban, and normal APCR/no apixaban. Patients taking apixaban were also tested for protein S functional activity and free antigen (n = 40).
FVL-HET patients taking apixaban had lower APCR ratios than wild-type patients (P < .001). Activated protein C resistance in FVL-HET patients taking apixaban fell more than 3 SD below the cutoff of 2.2 at which the laboratory reflexes FVL DNA testing. No cases of FVL were missed despite apixaban. In contrast to rivaroxaban, apixaban did not interfere with the assessment of protein S activity (mean activity 93.9 IU/dL, free antigen 93.1 IU/dL, P = .39). A total of 3 of 40 patients (8%) had low free protein S antigen (30, 55, and 57 IU/dL), with correspondingly similar activity results (27, 59, and 52 IU/dL, respectively). Apixaban did not cause a missed diagnosis of protein S deficiency.
Despite apixaban treatment, APCR testing can distinguish FVL-HET from healthy patients, rendering indiscriminate FVL DNA testing of all patients on apixaban unnecessary. Apixaban did not affect protein S activity.
Apixaban is a direct, antithrombin-independent factor Xa inhibitor. It inhibits both free and clot-bound factor Xa and prothrombinase activity, thus, interfering with conversion of prothrombin to thrombin and preventing formation of cross-linked fibrin clots.1,2 Apixaban is used to treat or prevent deep venous thrombosis or pulmonary embolism and to prevent stroke in patients with atrial fibrillation.2 Apixaban is known to prolong clot-based assays, including prothrombin time and activated partial thromboplastin time (aPTT), depending on the reagent. Changes observed in the clotting assays are subject to a high degree of variability.2 Most of the available studies were conducted in vitro, on spiked samples from healthy volunteers.3 The in vivo performance profile of the clot-based assays in real-life thrombophilic patients taking apixaban still remains to be established.
Since its release on the market in December 2012, apixaban has quickly become one of the most commonly used oral anticoagulants in industrialized countries, accounting for more than 40% of direct-acting oral anticoagulant prescriptions a year.4–6 Almost a third of the special coagulation laboratories in the United States now offer an anti-Xa test to measure apixaban activity levels.7 Patients receiving apixaban for thrombotic events often undergo hypercoagulation testing to rule out common inherited or acquired causes of hypercoagulability, including assays for activated protein C resistance (APCR) to detect factor V Leiden (FVL) mutation, and tests for protein S deficiency. FVL is the most common inherited risk factor for venous thromboembolism. Heterozygosity for FVL confers a 3- to 7-fold increase in the risk of thromboembolic events, whereas homozygosity is associated with an 80-fold increase.8,9 FVL affects approximately 5% of the white population and is responsible for more than 95% of cases of APCR.8,10,11 Activated protein C is a plasma anticoagulant that cleaves factor Va at several conserved arginine residues. The cleaved factor Va, in turn, acts as a cofactor for activated protein C in degrading factor VIIIa.8,12,13 The molecular basis for FVL is a point mutation in the factor V gene at G1691A, resulting in an arginine to glycine substitution at amino acid 506, which makes activated factor V resistant to cleavage by activated protein C.8,12,13 The most commonly used APCR test is a clot-based assay that measures the aPTT before and after activated protein C is added to the sample, expressed as a ratio of these 2 values. Originally, the APCR assay was conducted on undiluted patient plasma, which made results susceptible to the effects of anticoagulants, factor deficiencies, factor elevations, and acute thrombosis.8,11,14 However, an improved, “modified” version soon became available, in which a patient's plasma was diluted 5-fold with factor V–deficient plasma containing a heparin neutralizer. This increased the selectivity of the test for FVL by normalizing the concentrations of other plasma proteins involved in the formation and regulation of thrombin.15,16 The dilution with factor V–deficient plasma also rendered the test suitable for use in patients on heparin (with the addition of a heparin neutralizer) and vitamin K antagonists (eg, warfarin), as well as in patients with acute thromboembolic events.9,11,14,16 The sensitivity and specificity of the APCR test in those patients approaches 100%.11,15–17 However, data concerning its sensitivity in patients taking factor Xa inhibitors, such as apixaban, are still scarce. With the increasing use of these medications, a new question has arisen about the value of the modified APCR assay for FVL mutation screening in patients on direct factor Xa inhibitors, such as apixaban.
A previous study reported apixaban greater than 471 ng/mL was found to falsely elevate protein S activity tested with the Staclot Protein S assay (Stago, Parsippany, New Jersey); however, the study used normal samples spiked with apixaban, and it was not clear whether the extent of elevation would cause a missed diagnosis of protein S deficiency.18 Another study found a decreased APCR ratio that did not reach statistical significance; however, a Russell Viper Venom method, rather than an aPTT-based method, was used, and the presence or absence of FVL was not known in that study.4 Therefore, we sought to study the effect of apixaban on commonly used protein S and APCR tests in actual patients taking apixaban to determine if apixaban can cause misdiagnosis.
We previously showed that although a different factor Xa inhibitor, rivaroxaban, can artifactually slightly elevate the APCR ratio, it does not do so to the extent of masking FVL, even when levels of rivaroxaban were supratherapeutic.19 On the other hand, rivaroxaban caused a false elevation of protein S activity, and could cause a missed diagnosis of protein S deficiency if only a clot-based activity assay was used.19 In the current study, we sought to determine the effect of apixaban interference in APCR and protein S activity testing in an actual patient population by analyzing hypercoagulation panels received at our high-volume special coagulation laboratory. To our knowledge, this is the first study comparing the effect of apixaban on APCR and protein S activity in real patients with FVL undergoing apixaban treatment, rather than in spiked normal samples.
MATERIALS AND METHODS
We prospectively investigated all APCR test results performed by the Special Coagulation Laboratory at Massachusetts General Hospital between March 12, 2014 and November 15, 2018. Results from consecutive patients testing heterozygous for FVL while taking apixaban were recorded in group 1. For each patient in group 1, three matching patients were randomly selected among the patients tested on the same day for each of the three following control groups (groups 2–4). Group 2 consisted of patients who were also heterozygous for FVL, but were not taking apixaban. Group 3 included patients taking apixaban who tested wild type (WT) for FVL by DNA testing. Group 4 comprised patients who were not taking apixaban and had a normal APCR ratio. If control patients tested on the same date as a group 1 patient were not available, control patients were selected from the nearest dates that used the same equipment and reagent lot as in group 1, thus reducing bias that could potentially arise from operational differences. We also identified all patients taking apixaban who underwent protein S testing during a 6-month period (January–June 2019). All patients taking apixaban were tested for protein S with both an activity assay and a free-antigen assay.
The patients were evaluated using the standard FVL testing protocol adopted in our institution. The protocol included an aPTT-based APCR assay with dilution in factor V–deficient plasma (Coatest APC Resistance V assay; Chromogenix, West Chester, Ohio), on a Star Evolution analyzer, and an FVL DNA assay (Invader assay; Hologic, Boston, Massachusetts). APCR ratios of 2.0 and below were considered abnormal, and to ensure detection of FVL, values of 2.2 or lower underwent further workup by DNA testing. Activated protein C resistance and DNA analysis was performed for all patients taking apixaban.
Protein S functional activity was measured using STACLOT Protein S assay (Stago), which is based on prolongation of the aPTT with increasing concentrations of protein S, and uses bovine factor Va as a substrate for activated protein C, performed on a Star Evolution analyzer. Free protein S antigen was measured using the Asserachrom Free Protein S assay (Stago), a one-step ELISA assay that uses 2 monoclonal antibodies specific for distinct epitopes of the free form of protein S to directly measure free protein S in plasma. Free protein S antigen levels served as control to compare with the functional activity of protein S in patients on apixaban, because apixaban does not affect free protein S antigen results. Apixaban concentration for each specimen was determined using an anti-Xa assay (Stachrom; Stago). The results were provided in units per milliliter using a low–molecular weight heparin calibrator and converted into apixaban concentration using an apixaban calibration curve (using apixaban calibrators from Stago).
The geometric mean APCR ratios of the 4 groups were analyzed using Student t test, after performing analysis of variances. The mean protein S activity and free protein S antigen were analyzed using Student t test, after conducting an analysis for variances. P values < .05 were considered statistically significant. Excel software by Microsoft (Redmond, Washington) was used to perform the calculations.
Four groups of patients were identified for the APCR analysis as follows: group 1 (patients heterozygous for FVL taking apixaban [n = 14]), group 2 (patients heterozygous for FVL not taking apixaban [n = 14]), group 3 (WT patients taking apixaban [n = 14]), and group 4 (patients with normal APCR not taking apixaban [n = 14]). The patients included 29 males and 27 females between the ages of 3 and 85 years. In addition, one patient homozygous for FVL and taking apixaban was identified, and matched with a control patient homozygous for FVL and not taking apixaban (total n = 58 patients for the APCR study). Because the APCR ratios of patients homozygous for FVL are much lower than those of the heterozygous patients, regardless of the apixaban use, these 2 patients were excluded from the main analysis in order to avoid intentionally skewing the results toward statistical significance. Their APCR ratios were 1.28 on apixaban and 1.23 not taking apixaban, which was well below the APCR ratios in heterozygous patients as well as patients without FVL (shown below), indicating that the APCR assay was able to identify the homozygous FVL patient despite apixaban.
The geometric means (SD) of the APCR ratios in the 4 groups were as follows: group 1 (1.77 [0.07]), group 2 (1.79 [0.06]), group 3 (2.45 [0.11]), and group 4 (2.36 [0.10]). Corresponding mean apixaban concentrations are presented in Table 1. Figure 1 shows clear discrimination between FVL heterozygous and healthy patients, regardless of apixaban use. Individual APCR ratios of all FVL heterozygous patients, with or without apixaban, as well as the value 3 SD above the mean APCR in both groups of FVL patients (groups 1 and 2), fell below the cutoff of 2.0 used in our institution as the cutoff that defines normal versus abnormal APCR. Although apixaban slightly elevated APCR in WT patients (group 3 versus 4, P = .02), there was no statistically significant difference in the mean APCR of patients heterozygous for FVL taking versus not taking apixaban (P = .27). Both groups of FVL heterozygous patients had a significantly smaller APCR ratio than the WT patients (P < .001).
In the APCR part of the study, plasma apixaban concentrations ranged from 31.6 to 603.0 ng/mL (mean 123.4 ng/mL, SD 117.3 ng/mL) (Figure 2). The apixaban dosage was 5 mg twice a day in 25 patients (12 FVL heterozygous, 13 WT) and 2.5 mg twice a day in 3 patients (2 FVL heterozygous, 1 WT). In the protein S study, 37 of 40 patients (93%) took 5 mg of apixaban twice a day, and 3 of 40 patients (7%) took 2.5 mg of apixaban twice a day. The manufacturer-recommended apixaban dose is 2.5 to 5 mg twice a day, and for venous thromboembolism treatment, a 7-day induction period of up to 10 mg twice a day; the majority of patients qualify for 5 mg twice a day.2
For the protein S study, 40 patients taking apixaban were identified. The functional protein S activity was not significantly different than the free protein S antigen (Figure 3). The mean free protein S antigen was 93.9% (SD 21.1%) in patients on apixaban, and the mean protein S functional activity was 93.1% (SD 19.5%). The difference between the two means was not statistically significant. A total of 3 of 40 patients (8%) had low free protein S antigen (30, 55, and 57 IU/dL), with correspondingly similar activity results (27, 59, and 52 IU/dL, respectively). Therefore, unlike rivaroxaban, apixaban would not have caused a missed diagnosis of protein S deficiency if using a protein S activity assay alone. Overall, in the group of patients with low free protein S, the difference in the mean free protein S and the functional protein S was not statistically significant. The results are presented in Table 2.
Plasma apixaban concentrations in protein S part of the study ranged from 27.5 to 652.0 ng/mL (mean 139.8 ng/mL, SD 135.5 ng/mL) (Figure 4). The difference between the patients' protein S activity and free protein S did not correlate with their apixaban levels, attesting to the lack of significant false elevation in protein S activity even at higher apixaban concentrations (Figure 5).
The results of this study indicate although apixaban causes a slight rise in the APCR ratio (using an aPTT-based method) in WT patients taking apixaban, there is no statistically significant difference in patients heterozygous for FVL mutation taking versus not taking apixaban, and the APCR assay was able to accurately discriminate heterozygotes (and homozygotes) from healthy patients. Overall, the APCR ratio of patients heterozygous for FVL falls much below the WT patients, regardless of apixaban use. Therefore, taking apixaban is not expected to mask a diagnosis of FVL when using this APCR assay. There was a clear separation between the WT and the FVL patients taking apixaban (P < .001) and a much lesser difference between the WT patients taking versus not taking apixaban (P = .02) (Figure 1). One explanation for this phenomenon could be that diluting the patient's plasma 1:5 is sufficient to reduce the apixaban concentration to a level that causes only a minor interference. In contrast, the effect of FVL is significantly more pronounced, and could be detected whether or not the patient is taking apixaban. These results are reassuring, and suggest that if specimens are submitted for APCR testing while patients are taking apixaban, the laboratory will provide the correct answer even if the clinician does not inform the laboratory about apixaban, or if DNA testing is not available. The results also suggest that laboratories could proceed with this APCR test despite knowing apixaban is present. After the study period closed, two heterozygous FVL patients with supratherapeutic apixaban of up to 594.8 ng/mL were identified and their APCR results were still unaffected by apixaban.
During the investigation period, we also encountered one patient homozygous for FVL taking apixaban, and matched this patient with a control homozygous for FVL but not taking apixaban. Patients homozygous for FVL exhibit an even greater degree of activated protein C resistance than heterozygous patients. Therefore, their APCR ratios are even lower than those of heterozygous patients, and fall well below the normal cutoff point.
Plasma apixaban concentrations in our patients ranged from 27.5 to 652.0 ng/mL (mean 132.6 ng/mL, SD 128.5 ng/mL). According to one large multicenter study, the majority of real-life patients taking 2.5 to 5 mg of apixaban twice a day have plasma apixaban levels of 30 to 350 ng/mL.20 Ninety percent (62 of 69 patients) of our study population showed apixaban levels within this range, 1% (1 patient) had apixaban levels below this range, and the remaining 9% (6 patients) were above this range (Figures 2 and 4). By another report, the peak and trough concentrations for various doses are as follows (5th and 95th percentile with venous thromboembolism treatment): 2.5-mg bid dose: peak 30 to 153 ng/mL, trough 11 to 90 ng/mL; 5-mg bid dose: peak 59 to 302 ng/mL, trough 22 to 177 ng/mL; and 10-mg bid: peak 111 and 572 ng/mL, trough 41 and 335 ng/mL.2,21,22 Therefore, apixaban concentrations detected in this study are congruent with the concentrations typically seen in patients taking apixaban,2,23,24 and can be considered representative of a typical patient population.
In a similar study previously conducted by our group, argatroban (a direct thrombin inhibitor) was able to mask the diagnosis of FVL by falsely raising the APCR ratio into the normal range,25 and dabigatran (direct thrombin inhibitor) also substantially raised the APCR ratio.26 In contrast to these direct thrombin inhibitors, rivaroxaban (an antithrombin independent factor Xa inhibitor) raised the APCR ratio only slightly, and not into the normal range in any of our cases,19 highlighting that different direct oral anticoagulants can have dissimilar effects on the APCR ratio.
Simultaneous testing of functional protein S activity and free protein S antigen in patients on apixaban showed that functional protein S activity was not significantly different than free protein S antigen. Furthermore, low free protein S antigen occurred in 3 of 40 patients (8%) in this study, and in all 3 cases protein S activity was also low during apixaban treatment. These results are in contrast to our previous study with rivaroxaban, which falsely elevated protein S activity and caused missed diagnosis of protein S deficiency.19
With rivaroxaban, the false protein S elevation is likely due to the fact that the protein S activity assay in this study is aPTT based, and protein S prolongs the aPTT by serving as a cofactor for activated protein C–mediated cleavage of factors V and VIII. The degree of aPTT prolongation is proportional to the amount of protein S in the specimen. While apixaban also prolongs the aPTT, the degree of interference is not as pronounced as with rivaroxaban, and the modest elevation is not sufficient to mask a diagnosis of protein S deficiency. Of interest, patients' apixaban levels did not correlate with a greater delta between the protein S activity and the free protein S, further suggesting a lack of interference in the protein S activity assay.
FVL can cause falsely low protein S activity in some assays. Those assays rely on the patient as the only source of factor V. When the patient has FVL, the patient's factor V resists degradation by the activated protein C/protein S complex, making it seem like there is less protein S present, whereas in reality this apparent decrease in protein S function is caused by the resistance of abnormal factor V to degradation by the activated protein C/protein S complex. The present study avoids such consequences by using a protein S activity assay that supplies exogenous normal factor V, thus minimizing this interference.
This study showed that apixaban did not cause false-negative APCR/FVL or protein S activity results, but it is worth noting that false-positive results were also not found. False-positive APCR results due to apixaban are not expected, because anticoagulants that can prolong the aPTT in the APCR assay, including argatroban, bivalirudin, or heparin (when heparin levels exceed the heparin neutralizer), increase the APCR ratio rather than decrease the ratio. Nevertheless, for completeness, no cases of false-positive APCR were found in this study. All patients with positive APCR results, with or without apixaban, were confirmed to have FVL by DNA testing. Thus, all APCR positives were true FVL positives.
Similarly, false-positive protein S activity results are not expected because when the aPTT in the protein S assay is prolonged by an anticoagulant, such as heparin, argatroban or bivalirudin, the assay is designed to report a higher, not lower, result. In the assay, a prolongation is attributed to protein C, with protein S and cleaving factors V and VIII, thus the more prolonged, the higher the level of protein S activity is reported. Nevertheless, for completeness, no cases of false-positive (abnormal low) protein S activity were found in the present study. All cases of low protein S activity also had low free protein S antigen (not affected by apixaban), and all cases of normal protein S activity also had normal free protein S antigen.
The findings of this study should not be extrapolated to assume that other types of assays for activated protein C resistance (FVL) or protein S activity are also not affected by apixaban. For example, further study is needed to determine if prothrombin time– or Russell Viper Venom–based assays, or even an aPTT-based assay without dilution into factor V–deficient plasma, are affected.
The data obtained in this study suggest the Coatest APCR V assay can be used to test patients anticoagulated with apixaban for the FVL mutation. Because FVL specimens are generally reliably detected at an APCR cutoff of 2.0 or lower, in an effort to ensure 100% sensitivity in detecting FVL, our Special Coagulation Laboratory currently uses a protocol in which all specimens with APCR ratios of 2.2 or lower undergo DNA testing. In the present study, this protocol detected all patients heterozygous or homozygous for FVL, suggesting that DNA testing of patients with APCR values above 2.2 is unnecessary regardless of apixaban use. Protein S activity in this study was not affected by apixaban, therefore apixaban is not expected to cause a missed diagnosis of protein S deficiency if a protein S activity test is used instead of a free protein S antigen test.
The authors have no relevant financial interest in the products or companies described in this article.