Stephan Moll, MD writes… A major international coagulation conference, the bi-annual meeting of the International Society for Thrombosis and Haemostasis (ISTH; www.isth.org), took place in Amsterdam, Holland, from June 29th to July 4th, 2013. The clinically relevant highlights about thrombosis and anticoagulation are summarized below.
None of the scientific-clinical data presented at the meeting are immediately practice changing – and probably shouldn’t anyway, as abstracts are never in detail peer reviewed. However, several presentations are noteworthy, as they indicate clinically relevant changes to come in the next few years. The most relevant and outstanding development for me was the publication of the apixaban VTE study, discussed below under “A. New Oral Anticoagulants”.
A. New Oral Anticoagulants (NOACs)
Eliquis (apixaban) for Acute DVT and PE
The major DVT and PE acute treatment trial (AMPLIFY trial) of the new oral anticoagulant Eliquis was presented at ISTH. The drug is presently FDA-approved only for the treatment of atrial fibrillation. On the same day of the oral presentation, the New England Journal of Medicine published the full study – amore detailed discussion of the study is discussed here on Clot Connect. Summary: Apixaban was as effective as warfarin and lead to less major and clinically relevant bleeding than warfarin. Hopefully, the drug will be FDA approved soon – in next 6 months? – and clinically available. We will then have two new oral anticoagulant drugs for the treatment of DVT and PE to choose from –Xarelto (rivaroxaban) and Eliquis (apixaban).
Both drugs (apixaban and rivaroxaban) were started in their respective trials without bridging therapy with low molecular weight heparin, making them very attractive and easy to use in the acute DVT and PE management. Both are to similar degrees cleared by the kidney (25 % for Eliquis, 33 % for Xarelto). Eliquis is dosed twice daily, Xarelto once daily. The impressive finding in the Eliquis (apixaban) study is that the drug was safer than warfarin, having lead to less major bleeding.
B. Major Bleeding on the new oral anticoagulants
No antidote is presently available to reverse the anticoagulant effect of NOACs and, thus, to treat major bleeding occurring on these drugs. Whether prothrombin complex concentrates (PCC; Kcentra, Bebulin, Profilnine) have any clinical benefit in this situation is unclear (further discussed below).
a) Pradaxa (dabigatran)
A reversal agent for Pradaxa (dabigatran) is in development. It is an antibody fragment directed against dabigatran, which can neutralize the drug’s anticoagulant effect in vitro, referred to as “BI 655075”. Detailed, thorough pharmacokinetic and pharmacodynamic data on this antidote, as studied in rats, were presented (abstract # OC 36.2). The reversal agent is now in a human volunteer phase 1 study (clinicaltrials.gov: NCT01688830).
b) Xarelto, Eliquis and Edoxaban
A different strategy is being used in the development of a drug that can neutralize the anti-Xa anticoagulants Xarelto, Eliquis and edoxaban. The drug in development is an inactivated factor X, that cannot work as a coagulation factor, but is available to bind, and, thus, remove the anti-Xa drugs [ref: Lu et al, Nature Medicine 2013;19:446-451]. It is designed to inactivate all 3 anti-Xa drugs, Xarelto, Eliquis and edoxaban. The drug is called Andexanet (PRT4445) and made by Portola. Data were presented at ISTH [abstract # AS 20.1] of a phase 2 human volunteer study who took the anti-Xa anticoagulant apixaban; the anticoagulation effect of apixaban, as measured ex vivo in the blood of the volunteers, was successfully and completely reversed by Andexanet , and Andexanet was well tolerated. Further human volunteer trials using Andexanet to reverse the effect of Xarelto and edoxaban are on the way by the same investigators (Crowther M et al). Portola’s press release is here.
c) Prothrombin Complex Concentrates (PCC)
It is not known whether PCCs have any benefit when used to treat major bleeding in patients on one of the NOACs. A number of studies have shown some benefit of PCCs (a) on reversal of prolonged/abnormal coagulation tests in in vitro mixing studies and plasma of human volunteers on NOACs who received a PCC, and (b) in rat and mouse bleeding models. However, other studies have shown no benefit. Whether any benefit demonstrated in such studies, i.e. the normalization of a prolonged clotting test in vitro, indicates any efficacy in a patient with major bleeding is not known. At ISTH no patient data were presented and no clinical trial is ongoing testing a potential benefit of PCCs in patients with major bleeding on NOACs.
Interesting, though: in a healthy human volunteer study (abstract # OC 36.5), individuals were given rivaroxaban 20 mg twice daily and then received the 3-factor PCC Profilnine (50 U/kg), the 4-factor PCC Kcentra (=Beriplex; 50 U/kg) or normal saline. Findings: (a) Kcentra shortened the prolonged prothrombin time (PT) some, Profilnine did not; (b) both PCCs improved the endogenous thrombin potential (ETP), both to a similar extent; (c) most noteworthy: anti-Xa levels did not change. Conclusion: Some laboratory parameters changes suggest that the PCCs might have a beneficial effect if given to a patient with a major bleed, but other data suggest they would not be beneficial. Thus, it remains unclear whether PCCs have any benefit in the patient on NOACs who has a major bleed. However, if one were to empirically use a PCC, a 3-factor PCC may just be as good as a 4-factor PCC, as both had a similar effect on ETP.
d) Major Bleeding ≠ Major Bleeding
The new oral anticoagulants have a different bleeding pattern compared to warfarin: (a) all three NOACs (dabigatran, rivaroxaban and apixaban) lead to less intracranial bleeds, and (b) apixaban lead to less major bleeding in the atrial fibrillation and in the VTE trials. Intracranial bleeds are, appropriately, the most feared of major bleeds. This is supported by a study of major bleeding on warfarin, which found that patients with intracranial bleeds had a high mortality, whereas patients with major gastrointestinal bleeds had a low mortality (abstract # OC 36.1).
C. Performance of Coagulation Tests on the NOACs
The performance of coagulation tests depends on (a) the sensitivity of the assay to be influenced by the NOAC and (b) when the last dose of drug was taken.
a) Tests that can be used to measure NOACs anticoagulant effect (summary table here)
A number of publications in the last 2-3 years have shown that:
- For dabigatran: (a) the PT and aPTT are insensitive and are often normal in spite of therapeutic dabigatran levels; (b) the tests of choice to accurately assess dabigatran’s anticoagulant effect are the ecarin clotting time (ECT), chromogenic ecarin time (ECA), or dilute thrombin time (dTT); (c) the thrombin time (TT) is very sensitive to dabigatran levels and, thus, if the TT is normal, no significant dabigatran is on board.
- For rivaroxaban: (a) the aPTT and typically also the PT are rather insensitive and are often normal in spite of therapeutic rivaroxaban levels; (b) the test of choice to accurately assess rivaroxaban’s anticoagulant effect is an anti-Xa assay (with rivaroxaban calibrators).
- For apixaban: a) the PT and aPTT are quite insensitive and are very often normal in spite of therapeutic apixaban levels; (b) the test of choice to accurately assess apixaban’s anticoagulant effect is an anti-Xa assay (with apixaban calibrators).
b) Interferences with thrombophilia labs
Many coagulation tests – thrombophilia labs and individual clotting factor assays – can be influenced by NOACs. Antigen levels (protein S, protein C, antithrombin) are reliable, but all clotting factor activity assays are potentially influenced by the NOACs, yielding false low or false high values. The topic has recently been reviewed in a full publication in respect to dabigatran [ref 1] and rivaroxaban [ref 2], and was presented at ISTH in a nice summarizing table by Dr. Lindhoff-Last.
D. D-dimer as a Predictor of Recurrent Venous Thromboembolism
In a number of studies over the last few years the D-Dimer (tested on [ref 3] or off warfarin [ref 4]) has been found to be a helpful predictor for recurrent VTE. Several abstracts from ISTH add to our knowledge of D-dimer’s role in the prediction of recurrent VTE. What is clear is that there are still several unknowns about the performance of D-dimer assays and how clinicians should use results for clinical decision-making.
a) What D-dimer assay does a lab use?
Different D-dimer assays have varying sensitivities to detect elevated D-dimer levels and, are, thus, referred to as highly sensitive, intermediate-sensitive, and low sensitive [ref 5,6]. Before making any decisions on the results of a D-dimer test, either (a) when ruling out VTE after a pre-test probability assessment as been done via a clinical prediction rule such as the Wells Score, or (b) when using the D-Dimer for clinical decision making regarding the length of anticoagulation therapy, a clinician should known what D-dimer assay his/her laboratory uses and what sensitivity the assay has.
b) Is the D-dimer useful in men as well as in women?
- Abstract # OC 12.3 (Rodger M et al): This summary of the REVERSE I and II studies confirmed that a negative D-dimer on warfarin is a predictor of a low risk of VTE recurrence in women. However, men have a higher risk of recurrence, no matter whether D-dimer is positive or negative.
- The DODS study (abstract # AS 18.3; Kearon C et al.) also showed that (a) the D-dimer (obtained 4 weeks after having stopped warfarin) is not helpful in men to predict a low risk of recurrent VTE: even men with negative D-dimer have a high risk of recurrence; (b) women with a negative D-dimer have quite a low risk of recurrent VTE, particularly if the first VTE episode was hormone-associated: of 58 women who had a hormone-associated VTE and who stopped anticoagulation, none had a VTE recurrence.
c) Does the degree of positivity predict anything more than a simple positive/negative result?
- Abstract # OC 12.5 (Eichinger S et al.) showed that the degree of D-dimer positivity correlated slightly with the risk of VTE recurrence: a doubling of the D-dimer result lead to an approximately 1.35 fold higher risk of recurrent VTE.
d) Should we use one D-dimer range, or age and gender-specific D-dimer cut-offs?
- Abstract # OC 12.4 (Cosmi B et al): The authors used age and gender specific D-dimer cut-offs and showed that the D-dimer, if used in that manner, helped predict VTE recurrence. May be at some point in the future we should routinely use age- and gender-specific cut-offs.
E. IVC Filters in Patients with PE
One of the most relevant clinical studies (abstract # AS 18.2) presented at ISTH was the PREPIC II study. The study was done to assess whether placing an IVC filter in patients with acute PE has any benefit. Patients with acute PE (who also had a leg DVT) were randomized to receive an IVC filter or no filter. A total of 399 patients were enrolled, 50 % whom received a removable IVC filter, with the plan to remove it again within 3 months. Patients were routinely treated with anticoagulation. Findings: (a) the risk of recurrent PE was low and was no different between the two groups: 3% in IVC filter group vs. 1.5% in no-filter group at 3 months, and 3.5% versus 2.0 % at 6 months; (b) filter complications were relatively low, occurring in 2.1 % of IVC filter patients (2 asymptomatic perforations; 2 hematomas at puncture site); and (c) IVC filter removal was successful in 92.2 % and failed in 7.8 %. Conclusions: There is no benefit of routinely placing an IVC filter in patients with acute PE.
F. Bleeding risk scores
Several risk scores exist to assess a patient’s risk for bleeding on warfarin: (a) OBRI (Byeth 1998), (b) Hemorr2hages (Gage 2006), (c) Riete Score (Ruiz-Giminez 2008), (e) HAS-BLED (Pisters 2010), (f) Atria score (Fang 2011), (g) ACCP 2012 score (Kearon 2012). Using bleeding risk scores could potentially be a helpful tool when making decisions about length of anticoagulation therapy.
Most scores were developed from patients with atrial fibrillation (who are often older than patients with VTE). Comparison of these scores (abstracts # OC 42.1 and # OC 42.3) to predict major bleeding in VTE patients showed that the scores perform suboptimally. Further research is needed to optimize existing bleeding risk scores or create a new one that accurately assesses and predicts the risk of major bleeding in VTE patients on warfarin. Furthermore: as the pattern of bleeding is different with the NOACs, warfarin bleeding risk models may not accurately predict major bleeding on NOACs. A new bleeding prediction risk model specific for NOACs may have to be developed.
G. Interesting/Relevant Ongoing Clinical Studies
a) ALIFE-2 study (www.ALIFE2study.org)
Abstract SC11-3.05. Title: “Anticoagulants for living fetuses in women with recurrent miscarriage and inherited thrombophilia”. The primary objective of this randomized international, multi-center study is to evaluate the efficacy of low molecular weight heparin (LMWH) in women with recurrent miscarriage and inherited thrombophilia. Women will either receive once daily prophylactic-dose LMWH or no pharmacologic intervention. The study has been enrolling since Jan 2013.
b) HIGHLOW study (link)
Abstract SC11-3.05. Title: “Evaluation of efficacy and safety of intermediate dose LMWH versus fixed low dose LMWH in pregnant women with a history of previous VTE”. This is a randomized, open-label trial. Patients receive either intermediate dose LMWH (nadroparin; not routinely available in the U.S.), adjusted to actual body weight during pregnancy or fixed low-dose LMWH, i.e. nadroparin 2850 IU, the standard dose for prophylaxis of VTE. The study has been enrolling since April 2013.
- Adcock DM et al. The effect of dabigatran on select specialty coagulation assays. Am J Clin Pathol. 2013 Jan;139(1):102-9.
- Mani H et al. Ex vivo effects of low-dose rivaroxaban on specific coagulation assays and coagulation factor activities in patients under real life conditions. Thromb Haemost. 2013 Jan;109(1):127-36. Epub 2012 Nov 8.
- Rodger MA et al. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ. 2008 Aug 26;179(5):417-26.
- Douketis J et al. Patient-level meta-analysis: effect of measurement timing, threshold, and patient age on ability of D-dimer testing to assess recurrence risk after unprovoked venous thromboembolism. Ann Intern Med. 2010 Oct 19;153(8):523-31.
- Di Nisio M et al. Diagnostic accuracy of D-dimer test for exclusion of VTE: a systematic review. JTH 2006;5:296-304.
- Righini M et al. D-Dimer for venous thromboembolism diagnosis: 20 years later. J Thromb Haemost 2008;6:1059-71.
Disclosure: I have consulted for Janssen, Boehringer-Ingelheim, Daiichi.
Last updated: Aug 12th, 2013