Direct oral anticoagulants (DOAC) for prevention of recurrent arterial or venous thromboembolic events (ATE/VTE) in myeloproliferative neoplasms

In patients with BCR-ABL-negative myeloproliferative neoplasms (MPN), arterial or venous thromboembolic events (ATE/VTE) are a major burden. In order to control these complications, vitamin K antagonists (VKA) are widely used. There is no robust evidence supporting the use of direct oral anticoagulants (DOAC) in MPN patients. We therefore compared the efficacy and safety of both anticoagulants in 71 cases from a cohort of 782 MPN patients. Seventy-one of 782 MPN patients (9.1%) had ATE/VTE with nine ATE (12.7%) and 62 VTE (87.3%). Forty-five of 71 ATE/VTE (63.4%) were treated with VKA and 26 (36.6%) with DOAC. The duration of anticoagulation therapy (p = 0.984), the number of patients receiving additional aspirin (p = 1.0), and the proportion of patients receiving cytoreductive therapy (p = 0.807) did not differ significantly between the VKA and DOAC groups. During anticoagulation therapy, significantly more relapses occurred under VKA (n = 16) compared to DOAC treatment (n = 0, p = 0.0003). However, during the entire observation period of median 3.2 years (0.1–20.4), ATE/VTE relapse-free survival (p = 0.2) did not differ significantly between the two anticoagulants. For all bleeding events (p = 0.516) or major bleeding (p = 1.0), no significant differences were observed between VKA and DOAC. In our experience, the use of DOAC was as effective and safe as VKA, possibly even potentially beneficial with a lower number of recurrences and no increased risk for bleedings. However, further and larger studies are required before DOAC can be routinely used in MPN patients.

Curto-Garcia et al. [22] retrospectively reported on the results of 32 MPN patients with 38 venous thromboembolism and DOAC treatment. During a median follow-up period of 2.1 years, neither VTE recurrences nor major bleedings were observed.
Ianotto et al. [21] reported on the retrospective course of 25 MPN patients under DOAC treatment. During a median follow-up time of 2.1 years, two arterial thromboembolic and three bleeding events were observed. A case-control comparison of 25 MPN patients treated with low-dose aspirin (ASS) showed no difference in efficacy and safety [21].
Preliminary and retrospective data from Fedorov et al. [23] reported 22 DOAC-and 31 VKA-treated MPN patients with comparable incidences of recurrence and bleeding events.
However, there is no robust evidence supporting the use of DOAC in MPN, including a direct comparison with VKA treatment. Therefore, at our center, we retrospectively evaluated 71 MPN patients with 71 ATE/VTE treated with VKA (n = 45) or DOAC (n = 26) to compare the efficacy and safety of both anticoagulants.
The main objective of this retrospective, non-interventional, single-center study was to compare VKA with DOAC therapy in MPN patients. Of particular interest was the efficacy in ATE/VTE treatment, the prevention of ATE/VTE relapses, and the subsequent risk of bleeding complications under both anticoagulants. The data were collected in an electronic system. The Ethics Committee of our center approved the study. We focused on each patient with at least one MPNassociated arterial (ATE) or venous (VTE) thromboembolic event treated with VKA or DOAC. In line with previous studies, we defined an ATE or VTE associated with MPN if it occurred within 2 years prior to MPN diagnosis or after [27,28].
The follow-up time was defined as the time between the first occurrence of an ATE/VTE and last visit to our center. Treatment time was defined as the time between the start of anticoagulation (= after the first ATE/VTE) and the end of anticoagulation or the last visit to our center (if anticoagulation was not stopped) or first relapse (whichever came first).
For each MPN patient with an ATE/VTE and anticoagulation with VKA or DOAC, we collected demographic data, mutational profile, method of objective diagnosis of ATE/VTE, and presence of cardiovascular (CV) risk factors. In addition, further details on ATE/ VTE such as localization, total number, PT (prothrombin time)-INR (international normalized ratio), time of diagnosis and other cytoreductive or antiplatelet therapy were collected. Cytoreductive treatment was defined as the use of hydroxyurea, busulfan, anagrelide, interferon-alpha, ruxolitinib, and/or other JAK inhibitors at the time of ATE/VTE or within 6 months thereafter. Finally, time intervals regarding anticoagulation treatment, duration of treatment, occurrence of bleeding complications, and the number of relapses that occurred during or after completion of anticoagulation were recorded. The diagnosis of an ATE/VTE event required objective diagnostic procedures such as ultrasound, computed tomography, angiography, or scintigraphy.
The severity of bleeding complications was defined according to the criteria of the International Society on Thrombosis and Hemostasis [29]. According to these criteria, we considered a bleeding greater than II°( e.g., transfusion-related anemia, central nervous system involvement, or other life-threatening bleeding) to be clinically relevant.

Statistical methods
For continuous variables, the median and range are provided. The annual incidence of ATE/VTE recurrences was calculated by dividing the number of events by the sum of patient-years. Differences in the proportions were estimated using Fisher's exact test, Chi-square test, Mann-Whitney U test (statistical significance threshold set at p < 0.05), or log-rank test (Mantel-Haenszel test).

Results
Of the total 71 MPN-associated ATE/VTE, nine (12.7%) were ATE and 62 (87.3%) VTE. Table 1 provides an overview of demographic data and clinical characteristics of all 71 patients diagnosed with 71 initial ATE/VTE. Most ATE/VTE patients were female (n = 49, 69.0%) and were diagnosed as PV (n = 30, 42.3%), followed by MF (n = 20, 28.2%) or ET (n = 19, 26.8%). Two MPN patients with ATE/VTE were found to have a MPN at bone marrow biopsy, but both could not be further classified and were referred to as MPN unclassifiable. The JAK2 V617F mutation was the most frequent driver mutation (n = 63, 88.7%). The median age at ATE/VTE diagnosis was 54.0 years (22.0-82.0).
All 71 patients had a median total follow-up time of 3.2 years (range: 0.1-20.4) with an incidence rate for all 71 ATE/ VTE of 3.4% per patient/year. The corresponding rates for ATE and for VTE were 0.4% and 3.0% per patient/year, respectively.
Out  significant difference during the follow-up time (p = 0.0053) (Fig. 1). After comparing the absolute number of ATE/VTE recurrences, an analysis was performed that considered the probability of "recurrence-free" survival during the follow-up time. In this analysis, the difference between VKA-and DOAC-treated patients was not statistically different (Fig. 2, p = 0.2). The incidence rate of ATE/ VTE recurrences in VKA-treated patients was 8.1% per patient/year and 7.2% per patient/year in DOAC-treated patients. This difference was also not statistically different (alpha = 5%).

Bleedings
During anticoagulation with either VKA or DOAC, 10 of 71 patients (14.1%) experienced 11 bleeding complications over  In the VKA group, seven bleeding complications (63.6%), including four major bleeding complications, were recorded after a median time of 1.6 years (range: 0.1-6.8). Three out of four major bleedings (one esophageal varicose vein bleeding and two severe epistaxis episodes) occurred during VKA use alone (without low-dose acetylsalicylic acid). One patient underwent a combination therapy of VKA and low-dose acetylsalicylic acid and experienced major postoperative bleeding 1 day after total hip replacement implantation. During VKA treatment, three minor bleedings occurred (a menorrhagia, an episode with bloody semen, and an unspecified bleeding tendency).
During DOAC therapy, two minor and two major bleeding complications (n = 4, 36.4%) occurred after a median time of 0.5 years (range: 0.3-1.6). Both major bleeding episodes under DOAC anticoagulation (without low-dose acetylsalicylic acid) were gastrointestinal bleedings of unknown localization. The remaining two minor bleedings were epistaxis and petechial bleeding during DOAC therapy.
Overall, no significant differences were observed between DOAC and VKA anticoagulation therapy for both overall (p = 0.516) or major bleeding (p = 1.0). A comparison regarding different clinical and laboratory parameters between VKAand DOAC-treated patients is shown in Table 3. In the VKA group, only the total follow-up time (p = 0.0005) and number of ATE/VTE recurrences (p = 0.0053) were statistically different.

Discussion
Myeloproliferative neoplasm (MPN) patients have an increased risk of arterial and venous thromboembolic events (ATE/VTE). In larger MPN cohorts, the proportion of patients suffering from ATE/VTE is reported to be 10 to 30% [22]. Accordingly, vascular events occurred in 9.1% (71/782) of our 782 MPN patients. The incidence rate for the first 71 ATE/VTE was 3.4% per patient/year with a VTE rate of 3.0% per patient/year. Prospective studies in MPN observed comparable VTE rates of 0.5-3.7% [6,7]. The ATE incidence rate of 0.4% per patient/year in our MPN patients was also similar to the reported ATE rates of 0.2 to 1.5% [5,11].
In recent decades, anticoagulation with vitamin K antagonists (VKA) has been the treatment of choice to prevent ATE/ VTE recurrences in MPN patients. Hernández-Boluda et al. [14] reported a 2.8-fold risk reduction for recurrence in 150 ET and PV patients with ATE/VTE and VKA treatment. In 206 MPN patients, De Stefano et al. [15] also found a reduction in the recurrence rate of VTE with VKA. The incidence rate of recurrent VTE was 5.3% per patient/year among patients with long-term VKA and 12.8% per patient/year after VKA discontinuation (p = 0.008). The VTE recurrence rate in our cohort was comparable at 8.0% per patient/year.
As far as anticoagulation with direct oral anticoagulants (DOAC) is concerned, there are few studies in MPN that indicate good efficacy with sufficient safety. In a retrospective study by Curto-Garcia et al. [22] in 32 MPN patients receiving DOAC for MPN-associated venous thromboembolism treated with DOAC, no VTE relapse but one ATE occurred. No major and three minor bleedings were reported. Ianotto et al. [21] retrospectively reported two ATE and no VTE recurrences in a cohort of 25 DOAC-treated MPN patients. Three major and two minor bleedings were observed. Curto-Garcia et al. [22] reported a median age of 49.9 years and a median follow-up of 2.1 years in their publication. The median follow-up time in the study of Ianotto et al. [21] was quite similar with 2.1 years. However, both studies did not compare DOAC treatment with VKA in their cohort [21,22]. Fedorov et al. [23] reported preliminary data on recurrence rates and bleeding complications in 22 DOAC-and 31 VKA-treated MPN patients. During a short follow-up of 8 months, the number of ATE/ VTE recurrences (DOAC, n = 5 versus VKA, n = 6) and of all bleeding complications (DOAC, n = 5 versus VKA, n = 11) were not significantly different. The median age of our 71 MPN patients at the time of ATE/VTE was 54 years and was comparable to the studies of Curto-Garcia et al. and Ianotto et al. [21,22]. The median duration of anticoagulation was lower at 1.0 years for VKA and 1.3 for DOAC. During anticoagulation therapy, significantly more relapses occurred under VKA (n = 16) compared to DOAC treatment (n = 0, p = 0.0003). However, during the entire observation period of median 3.2 years (0.1-20.4), ATE/VTE relapse-free survival (p = 0.2) did not differ significantly between the two anticoagulants. This is mainly due to the significantly longer follow-up time for VKA patients (p = 0.0005). The corresponding recurrence rates for VKA and DOAC treatment (during and after discontinuation of anticoagulation) did not differ significantly either.
During anticoagulation with VKA, 53% and at the time of relapse, 57% of patients treated with VKA were in the therapeutic range with PT-INR. A major disadvantage of the VKA is the narrow therapeutic range and the time patients spend in the therapeutic range (TTR, "time in therapeutic range"). Even in well-conducted comparative studies of VKA and DOAK, the TTR was on average only between 55 and 65% [17,20,30,31].
As in the studies mentioned above, we have not observed any increased bleeding propensity under DOAC. In particular, the rate of major bleeding was not higher under DOAC compared to VKA. Regarding bleeding complications with anticoagulation, the German MPN Registry of the Leukemia Study Alliance [32] reported bleeding in 437 MPN patients, including eight with DOAC (rivaroxaban) treatment. In a multivariate analysis, the risk of bleeding during DOAC treatment was slightly reduced compared to VKA.
In summary, our results complement the currently limited literature [21][22][23]32] on the efficacy and safety of DOACtreated MPN patients. Despite the limitations-small number of patients, retrospective analysis, and short treatment timeour data suggest that the use of DOAC was as effective and safe as VKA. However, further and larger studies are required before DOAC can be routinely used in MPN patients.
Authors' contributions All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by Karlo Huenerbein. The first draft of the manuscript was written by Karlo Huenerbein and Dr. Kai Wille, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding Open Access funding enabled and organized by Projekt DEAL.
Data availability The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
Ethics approval Ethical approval was waived by the local Ethics Committee of the Ruhr University Bochum in view of the retrospective nature of the study, and all the procedures being performed were part of the routine care.
Consent to participate/for publication Informed consent was obtained from all individual participants included in the study. Patients signed informed consent regarding publishing their data and photographs.
Code availability Not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.