Measuring direct oral anticoagulants—when, how

Amy Carpenter Aquino

January 2022—Laboratories don’t have to monitor direct oral anticoagulants, but they might want to measure DOAC drug levels in some situations in some patients, said Karen A. Moser, MD, associate professor of pathology, University of Utah Spencer Fox Eccles School of Medicine, and medical director of the hemostasis/thrombosis laboratory at ARUP Laboratories, in a CAP21 session. She and others reported what those situations are, 2021 recommendations, CAP proficiency testing findings, DOAC effect on coagulation assays, and cases from their practices.

“Clinical trials evaluating these drugs described expected on-therapy concentrations for each of the drugs—dabigatran, the direct thrombin inhibitor, and three direct Xa inhibitors,” she said, referring to rivaroxaban, apixaban, and edoxaban.

In the clinical trial cohorts, expected peak and trough ranges were described for each of the four DOACs in patients treated for deep vein thrombosis and pulmonary embolism, and collated in the 2018 International Council for Standardization in Haematology recommendations for laboratory measurement of DOACs, with an update to the recommendations published in 2021 (Gosselin RC, et al. Thromb Haemost. 2018;118[3]:437–450 and Douxfils J, et al. Thromb Haemost. 2021;121[8]:1008–1020). “That at least gives us something to compare with in the literature if we choose to make a measurement of any of the DOACs while a patient is on therapy,” Dr. Moser said. “If we make a random measurement, it’s hard to know what that number might mean. We can’t compare it with a peak or a trough” and are left wondering if the number means the patient is sufficiently anticoagulated.

DOACs have a yearly risk of one to three percent for major bleeding and one to two percent for thromboembolic events, said co-presenter Oksana Volod, MD, director of coagulation and associate professor of pathology and laboratory medicine at Cedars-Sinai Medical Center, Los Angeles.

“Several studies demonstrated there is a dose-response relationship between DOAC concentrations and those adverse events,” Dr. Volod said. “Thromboembolic events, as well as strokes, mainly occurred in patients with the lowest trough levels, whereas high trough levels were associated with a higher risk of major bleeding.” It is likely that patients could further benefit from tailored DOAC therapy, in particular special patient populations, such as those with extremely high or low body weight, impaired kidney function, or prior intestinal surgery, and in specific clinical situations—for example, patients who restart anticoagulation after a major bleeding or who experience a thrombotic event while on DOAC therapy (Toorop MMA, et al. J Thromb Haemost. 2020;18[12]:3163–3168).

Dr. Volod cited the key recommendations from the 2021 update of the International Council for Standardization in Haematology consensus on laboratory measurement of DOACs:

• There are insufficient data to date for providing dose-adjustment recommendations based on DOAC levels alone.
• Nevertheless, DOAC measurements may identify potential excessive clearance or drug accumulation and could be used in situations where the benefit of such measurement is likely to outweigh the risk, for example, in nonurgent situations.
• If a DOAC measurement has been requested for urgent purpose, results should be provided within 30 minutes to aid in acute clinical decision-making.

Meeting the 30-minute recommendation may not be feasible, Dr. Volod said. The Cedars-Sinai laboratory offers apixaban and rivaroxaban levels 24/7, but “specimen drawing and delivery especially can take several minutes, so results may not be available within 30 minutes.”

The possible indications for DOAC testing in nonurgent situations, according to the ICSH, are advanced age, severe renal failure and dialysis dependence, high bleeding risk interventions, and a BMI above 40 kg/m². For patients with a BMI less than 40 or weight less than 120 kg, any DOAC can be used, but for patients above those levels, rivaroxaban or apixaban are the preferred drugs. “And no monitoring is indicated because of the lack of available safety and efficacy data,” Dr. Volod said. Another nonurgent indication for DOAC testing would be in a case of a possible drug interaction.

The urgent situations are those of acute bleeding and where appropriate reversal strategies have to be determined.

In the perioperative setting, the American Society of Regional Anesthesia and Pain Medicine 2015 guidelines suggest that DOAC interruption be based not only on their respective half-lives but also on residual drug concentration. “In this situation,” Dr. Volod said, “to accurately measure low plasma drug concentrations, we recommend that laboratories have to have assays, Xa assays, calibrated for the assessment of low plasma concentrations,” for example, less than 50 ng/mL^-1 (Douketis JD, et al. Reg Anesth Pain Med. 2016;41[2]:127–129).

In emergency situations (bleeding, thrombosis, urgent invasive procedure, thrombolysis), “we potentially can assess if DOAC is present and how much,” Dr. Volod said. “The aim would be to identify a level within or above the on-therapy range to guide the potential use of specific reversal agents or at least identify if a drug is there.” DOAC antidote administration is warranted if the drug concentration is greater than 30 ng/mL^-1 in patients requiring an urgent intervention, according to International Society on Thrombosis and Haemostasis guidelines. In patients with serious bleeding, antidote administration should be considered if the drug concentration exceeds 50 ng/mL^-1 (Levy JH, et al. J Thromb Haemost. 2016;14[3]:623–627).

The CAP Hemostasis and Thrombosis Committee, of which Dr. Volod is a former member, recently reported 2013–2016 Survey data on the measurement of rivaroxaban and dabigatran (Volod O, et al. Arch Pathol Lab Med. Epub ahead of print June 16, 2021. doi:10.5858/arpa.2020-0633-CP), including the effect of those drugs on coagulation assays such as PT, aPTT, and thrombin time. “As published in the literature,” Dr. Volod said, “aPTT is not very sensitive to direct Xa inhibitors and it is insensitive to apixaban.”

“There is significant variability between reagents,” she said, noting that the three most common instrument-reagent combinations reported in the Survey were those of Diagnostica Stago, Instrumentation Laboratory (now Werfen), and Siemens Healthineers. “The most sensitive in terms of aPTT was the IL reagent showing abnormal results for aPTT and rivaroxaban,” she said. Rivaroxaban aPTT results were measured at concentrations of 50, 200, and 400 ng/mL, and the aPTT reagents were more responsive to rivaroxaban concentration in the 200–400 ng/mL range. aPTT was reported as prolonged by 92 percent of participants for 200 ng/mL and by 94 percent of participants for 400 ng/mL.

In an earlier study of patients treated with DOACs, Dr. Volod said, prothrombin time was somewhat more sensitive to direct Xa inhibitors (rivaroxaban, edoxaban, apixaban). Rivaroxaban showed the strongest effect on PT, followed by edoxaban and then apixaban (Douxfils J, et al. J Thromb Haemost. 2018;16[2]:209–219).

In the CAP Survey, the prothrombin time was more responsive to rivaroxaban. At the 50 ng/mL rivaroxaban concentration, 49.6 percent of participants reported normal prothrombin time results, but PT was reported as prolonged by 98.7 percent of participants for the 200 ng/mL rivaroxaban concentration and by 100 percent of participants for the 400 ng/mL rivaroxaban concentration. The Diagnostica Stago Neoplastin CI Plus reagent appeared to be the most sensitive to rivaroxaban across all three drug concentrations.

“There was insufficient sensitivity, even if paired with a normal aPTT, to completely exclude DOAC presence,” Dr. Volod said.

Table 1.
DOAC—assay choices for measurement

LC/MS

• Considered gold standard for quantification
• Reportable range 5–500 ng/mL
• Only available in small number of reference laboratories
• No international reference standard available
• Timing of measurement matters
• Prefer peak or trough instead of random

DOAC-specific anti-Xa

• Classified as research use only
• Classified as a laboratory-developed test for clinical use; requires local validation
• Timing of measurement matters
• Prefer peak or trough instead of random

DOAC-specific DTI

• Dilute thrombin time
• Ecarin-based assays

Other

• Anti-Xa level calibrated for UFH/LMWH
• PT/PTT
• Thrombin time
• TEG
• Use qualitative tests with caution

Douxfils J, Ageno W, Samama C-M, et al. J Thromb Haemost. 2018;16[2]:209–219.

Nearly 50 percent of participants reported that the prothrombin time was not prolonged with a 50 ng/mL rivaroxaban concentration, “again highlighting that with the normal PT, we cannot exclude drug presence,” Dr. Volod said.

“This supports the [2018] ICSH consensus screening recommendations that PT and aPTT may not be reliable to detect the presence of on-therapy concentration of all DOACs,” she said. They are not responsive to on-therapy apixaban level, and they should not be used to quantify DOAC concentration. In a patient with known DOAC exposure, a prolonged PT or aPTT should be considered secondary to drug effect until proven otherwise, according to the recommendations. The recommendations also say that in emergent or life-threatening conditions, tests for quantifying DOAC should be performed to aid in managing the patient. “A normal thrombin time will exclude the presence of significant dabigatran concentration,” Dr. Volod said, referring to the recommendations. “And nonspecific point-of-care testing methods, like PT, aPTT, and activated clotting time, may not have sufficient responsiveness to detect DOAC presence and should not be used for those purposes” (Gosselin RC, et al. Thromb Haemost. 2018;118[3]:437–450).

The use of viscoelastic assays in DOAC measurement does not hold much promise because they are not sensitive enough to DOAC effects, Dr. Volod said. “They are more sensitive to dabigatran because the activator is kaolin in the majority of those assays and they are least sensitive to rivaroxaban and apixaban.” The ICSH consensus recommendation in 2018 said that at that time, there was not enough clear data to support the use of TEG or ROTEM for detecting DOAC anticoagulant activity.

Amanda M. VanSandt, DO, medical director of the hemostasis and thrombosis service, Oregon Health and Science University, and associate professor of pathology and laboratory medicine, OHSU School of Medicine, set out the assay choices for DOAC measurement (Table 1). Laboratories with a more routine volume of cases, she said, should favor validating an on-site quantitative DOAC assay for accurate and timely results. “Actual measurements of concentration would be most useful, such as the LC/MS method or a DOAC-specific calibrator with an anti-Xa assay.” Laboratories that don’t have quantitative DOAC assays on site should consider send-out testing for quantitative measurements. However, laboratories that don’t have a validated quantitative assay locally and cannot await send-out testing, such as in emergencies, should consider the use of a qualitative assay to assess for DOACs, such as routine clotting time or perhaps a heparin-specific assay, she said.

Qualitative assays, such as anti-Xa activity calibrated for unfractionated or low-molecular-weight heparin, could be considered useful in an emergent situation, Dr. VanSandt said, “though these have to be used with caution because heparin and the anti-Xa DOAC act in the same mechanism or at the same time in the coagulation cascade. This means that anti-Xa assays calibrated for unfractionated or low-molecular-weight heparin will likely turn positive in the setting of a DOAC once the medication reaches a certain concentration within the sample.”

At least one study suggests that using anti-Xa assays calibrated for UFH and LMWH could detect significant levels of DOAC (Gosselin R, et al. Int J Lab Hematol. 2016;38[5]:505–513), though others suggest response may vary by kit (Sabor L, et al. Thromb Res. 2017;156:36–38).

“If you’re using this type of method in your hospital or laboratory, we want you to make sure you understand how these DOACs are affecting the assay in your hand,” she said.

Laboratories would welcome anti-Xa chromogenic assays with DXa-inhibitor-specific calibrators, Dr. VanSandt said, and a limited number of FDA-approved calibrator materials are now available. These assays can measure a wide range of concentration of the DOACs, and just as the DOACs can make the heparin-calibrated assays positive, heparin can make the DOAC-calibrated assays positive. “So it would be useful to optimize your laboratory information system to detail what anticoagulant is being used by the patient so the lab can use the correct calibrator on the assay.”

A proposed novel testing strategy using a urine dipstick assay (DOAC Dipstick, Doasense GmbH, Germany) for DOAC detection has been tested in small trials, Dr. VanSandt said, explaining, “Because these medications are renally cleared, a urine assay is possible to detect the different DOACs.” Results from an interlaboratory study of seven participants showed the dipstick is fairly sensitive for apixaban, rivaroxaban, and dabigatran and “pretty specific for the drug involved,” she said (Harenberg J, et al. Semin Thromb Hemost. 2019;45[3]:275–284).

Drs. Volod and Moser presented clinical cases they encounter in practice, ones in which DOAC measurement could be helpful even though lab monitoring isn’t required. The first was of an aPTT and heparin assay discrepancy in a patient admitted to the hospital while on apixaban and then switched to unfractionated heparin.

At Cedars-Sinai, the method of measuring heparin effect is a heparin level by anti-Xa assay, Dr. Volod said. When the patient was switched to unfractionated heparin, the anti-Xa assay was showing a supratherapeutic heparin level. The value of the therapeutic level is 0.3 to 0.7 IU/mL, and the patient’s values ranged from 1.94 to 2.20 IU/mL. The clinicians stopped the heparin, and the case was brought to Dr. Volod’s attention. The patient was stable. She added to the workup, in addition to the anti-Xa assay, aPTT (normal, 35 sec), PT (elevated, 20.7 sec), and fibrinogen (elevated, 522 mg/dL). Since PT reagents have heparin neutralizers in them, PT usually will not be normal if a patient is on even a therapeutic heparin dose, she said, so the high PT value of 20.7 seconds “was another clue that the patient was still having a residual effect of apixaban.”

Based on those results, “it immediately came to my attention that most likely we still experienced the DOAC Xa-inhibitor apixaban effect on coagulation assays, such as unfractionated heparin as well as PT,” Dr. Volod said. She recommended continued patient monitoring with aPTT rather than an anti-Xa level. The heparin was restarted, and aPTT, which was steadily rising, was used for monitoring.

Depending on renal function, in the first 24 to 36 hours—and in Dr. Volod’s experience, even after several days up to one week—DOAC still can have an additive effect on the screening assays, she said. “Therefore, laboratories should be able to provide alternative strategies for assessing heparin anticoagulation.”

Dr. Moser presented a case similar to though not the same as Dr. Volod’s case, and it’s representative of the following situation she’s encountered: A physician calls the lab because they have a patient switching from rivaroxaban (or another direct Xa inhibitor) to LMWH. They want to use the lab’s anti-Xa assay calibrated for LMWH to monitor LMWH during the transition. They ask, “Will rivaroxaban interfere with the test?”

Just as in Dr. Volod’s case, the anti-Xa activity assay—the underlying parent assay—is the same, whether the laboratory calibrates it to measure UFH, LMWH, or rivaroxaban, Dr. Moser said. And it can pick up the effects of any or all of those. “So this is a sticky situation. It’s challenging to provide laboratory monitoring when we’re transitioning from these DOACs to one of the heparins if we want to make laboratory measurements.”

It’s easier with dabigatran, she said, because dabigatran is a direct thrombin and does not interfere with anti-Xa activity assays. “The direct Xa inhibitors cause the issue.”

Another case centered on the potential utility of viscoelastic assays. In this case, a patient with a fall was transferred to Cedars-Sinai for a higher level of care. “We knew the patient took rivaroxaban within 24 hours of admission and received a dose of Kcentra [prothrombin complex concentrate],” Dr. Volod said.