Dabigatran in the treatment and secondary prophylaxis of venous thromboembolism in children with thrombophilia

In children, venous thromboembolism (VTE; including deep vein thrombosis [DVT] and pulmonary embolism [PE]) is a severe multifactorial disease. Common clinical risk factors include use of central venous catheters (CVCs), underlying disease, and thrombophilia.^1-5 For example, a significant association between inherited thrombophilic disorders and VTE onset, as well as recurrence, has been shown in a meta-analysis of observational studies in children⁶; therefore, screening for inherited thrombophilia defects, which typically include factor V Leiden (FVL), prothrombin (PT) mutations, and deficiencies of antithrombin, protein C, or protein S, could be of clinical value.^5,7 

Longer term complications of VTE include recurrence, postthrombotic syndrome (PTS), chronic thromboembolic pulmonary hypertension, and death.^4,8-10 The VTE recurrence rate has been reported to be ∼3% in newborns and 8% in older children,^6,11 with risk increasing to as high as 29% in children with certain inherited thrombophilia traits.¹² VTE-related death in children has been reported to be 0% to 3.7% and PTS has been reported with a frequency of 9.5% to 70%.^13-15 

Anticoagulation is the standard treatment for VTE.¹⁶ The exact duration for optimal anticoagulation therapy has yet to be established in children with acute VTE. Recent evidence-based recommendations suggest longer duration for unprovoked thromboembolic events (6-12 months), regardless of inherited thrombophilia markers, than provoked events (3 months).^16,17 Continuation of treatment is dependent on the benefits of maintaining a reduced risk of VTE recurrence vs the risk of bleeding.^18,19 

Standard of care (SOC) anticoagulation is typically low-molecular-weight heparin (LMWH) or vitamin K antagonists (VKAs) in pediatric patients with symptomatic VTE.¹⁶ Direct oral anticoagulants (DOACs) have shown superiority or noninferiority to SOC in lowering the prospect of thromboembolic complications, with comparable or diminished bleeding risk, and, therefore, current adult-based recommendations favor their use in the treatment of patients with proximal DVT and nonmassive PE, except for patients with antiphospholipid syndrome.²⁰ Currently, there is no preferred anticoagulation agent recommended for long-term use, particularly in thrombophilia subgroups; however, use of DOACs might be advantageous because of the requirement for less clinical monitoring and follow-up than SOC treatments, and also, fewer food and drug interactions.²⁰ The potential benefits of DOAC use, including stability and lower residual thrombus burden, along with a lower bleeding risk, have yet to be fully investigated in children with unprovoked VTE.

Two large international, multicenter, pediatric phase 2b/3 studies have examined the efficacy and safety of the DOAC dabigatran etexilate in the treatment of acute VTE and in long-term secondary VTE prevention in children.^21,22 The acute VTE treatment (DIVERSITY, #NCT01895777) open-label, randomized, phase 2b/3 study demonstrated the noninferiority of dabigatran compared with SOC in children from birth to the age of <18 years.²¹ The secondary VTE prevention (#NCT02197416) phase 3, single-arm, cohort study showed a favorable safety profile for dabigatran in secondary VTE prevention in children with persistent VTE risk factor(s) from birth to the age of <18 years.²² Based on the results of these 2 pivotal, global studies, dabigatran etexilate was approved by the European Medicines Agency and US Food and Drug Administration for pediatric indications in January and June 2021, respectively.^23,24 We present herein, a subgroup analysis from these studies of the efficacy and safety of dabigatran in children with thrombophilia vs those with negative/unknown thrombophilia status.

The findings presented here are from subgroup analyses of children with thrombophilia vs those with negative/unknown thrombophilia status from the acute VTE treatment (DIVERSITY) and secondary VTE prevention studies. Both study designs have been described previously and key exclusion criteria were: “conditions associated with increased bleeding risk, renal dysfunction, hepatic disease, active infective endocarditis, heart valve prosthesis requiring anticoagulation, and children aged 0 to <2 years with gestational age at birth <37 weeks or with bodyweight lower than the third percentile (according to World Health Organization standards).”^21,22,25,26 The studies were conducted in accordance with the Declaration of Helsinki and the principles of Good Clinical Practice and were approved by all investigational site ethics committees. Written informed consent was obtained before participation, according to the International Conference on Harmonisation Good Clinical Practice, and the regulatory and legal requirements of each participating country. Both studies were sponsored by Boehringer Ingelheim. As listed, both studies were registered at www.clinicaltrials.gov.

For the acute VTE treatment study (DIVERSITY), patients from birth to the age of <18 years with a diagnosis of VTE (ie, DVT, PE, or cerebral venous sinus thrombosis) objectively confirmed by compression ultrasound, computed tomography, or magnetic resonance imaging, and who were initially treated for 5 to 21 days with SOC (eg, unfractionated heparin or LMWH), with parenteral anticoagulation therapy expected to last for ≥3 months, were eligible for inclusion.^21,25 

For the secondary VTE prevention study, patients aged 3 months to <18 years with an objectively confirmed diagnosis of VTE, who had been treated for acute VTE with SOC for ≥3 months, or who had completed dabigatran or SOC treatment in the acute VTE treatment study and had an unresolved clinical thrombosis risk factor requiring further anticoagulation for secondary prevention of VTE, were eligible for inclusion.^22,26 

In the current analysis, patients were assessed for thrombophilia status and were classified as “thrombophilia positive” or “thrombophilia status negative/unknown.”

In the acute VTE treatment study (DIVERSITY), patients were randomized (2:1) to receive open-label dabigatran or SOC, treated for 3 months from randomization, and followed up for an additional month.^21,25 In the secondary VTE prevention study, patients were treated with dabigatran for up to 12 months.^22,26 Dabigatran was dosed according to an age- and weight-adjusted nomogram derived from estimated renal function to achieve exposure comparable to that of adult populations treated with dabigatran.^25-27 

Primary and secondary end points differed according to the individual study. In the acute VTE treatment study, the primary efficacy end point was a composite (centrally adjudicated by an independent, blinded committee) of the proportion of children with complete thrombus resolution, freedom from recurrent VTE (including symptomatic and asymptomatic, contiguous progression or noncontiguous new thrombus, DVT, PE, paradoxical embolism, and thrombus progression), and freedom from VTE-related death.^21,25 Secondary/other end points included residual thrombotic burden by the end of study treatment (defined as complete or partial thrombus resolution, stabilization of thrombus, or thrombus progression); incidence of bleeding events; major bleeding events (MBEs; defined as fatal bleeding, clinically overt bleeding [≥20 g/L decrease in hemoglobin over 24 hours], retroperitoneal, pulmonary, intracranial, central nervous system bleeding; bleeding requiring surgical intervention) clinically relevant nonmajor (CRNM) bleeding events (defined as overt bleeding for which a blood product was administered and that was not directly attributable to the patient’s underlying medical condition, or bleeding that required medical or surgical intervention to restore hemostasis, other than in an operating suite);²⁸ and other safety/tolerability/adherence outcomes.^21,25 

In the secondary VTE prevention study, all outcomes were considered safety related.²² Primary end points included recurrence of VTE, mortality, MBEs, CRNM bleeding events, and minor bleeding events; assessed at 3, 6, and 12 months after enrollment.²² Secondary end points included the occurrence of newly diagnosed or worsening of baseline PTS (per the modified Villalta scale)^29,30 at 3, 6, and 12 months after enrollment. PTS was assessed at 3 months to capture any patients enrolled with a history of VTE who might subsequently present with PTS.

At each on-treatment study visit in both studies, adherence was calculated as adherence (%) = (actual number of dabigatran doses taken since last count ÷ planned number of dabigatran doses that should have been taken in the same period) × 100. The overall average adherence was calculated as the average of adherence at each visit.

Thrombophilia screening was not a requirement of the clinical trials, therefore, patients’ requirement for thrombophilia screening was determined by the attending physician at each site. The study databases were reviewed, and patient thrombophilia status was determined and classified as “thrombophilia positive” or “thrombophilia status negative/unknown”; a further subset of patients with confirmed inherited thrombophilia³² (those with only FVL and/or PT [G20210A] mutation) was also established. For the thrombophilia-positive subgroup, thrombophilia was additionally categorized as “major” or “minor” based on predefined criteria. Major thrombophilia included patients with homozygous FVL, homozygous PT mutation, compound heterozygous FVL and PT mutations, antithrombin deficiency, protein C deficiency, protein S deficiency, or those who were antiphospholipid antibody (APLA)– and/or lupus anticoagulant (LA)–positive, as well as specified combinations of these or other thrombophilic conditions or mutations (listed in Table 3 footnotes).³¹ Minor thrombophilia included coagulation disorders not defined as major thrombophilia: heterozygous FVL, heterozygous PT gene mutation (including those whose zygosity was recorded as unknown), specified combinations of thrombophilic conditions, and specified other mutations and conditions with uncertain thrombophilic significance (Table 3 footnotes).³¹ 

For both studies, patient demographics, medical history, and baseline clinical characteristics were analyzed descriptively. Time-to-event analyses were summarized as Kaplan-Meier estimates; other end points, including safety and adverse events (AEs) were analyzed descriptively. In the acute VTE treatment study, noninferiority testing for the composite primary end point was performed on Mantel-Haenszel–weighted rate difference with a margin of 20% used for a 2-sided test at significance level of .05.³³ 

The acute VTE treatment study (DIVERSITY) took place across 65 centers in 26 countries. The analysis population comprised 267 patients; 62 patients with documented thrombophilia (32 with confirmed inherited thrombophilia [FVL and/or PT mutations]) and 205 patients with thrombophilia-negative/unknown status (Figure 1). Within the thrombophilia subgroup, 37 patients were categorized as having major thrombophilia and 25 with minor thrombophilia. A sensitivity analysis of the thrombophilia-negative (n = 145) and thrombophilia-unknown (ie, not tested; n = 60) subgroups found both groups to be largely similar in terms of safety and efficacy outcomes (supplemental Table 1). Therefore, these subgroups were merged into 1 “thrombophilia-negative/unknown” group.

The secondary VTE prevention study took place across 60 sites in 22 countries. The analysis population included 213 patients; 106 with documented thrombophilia (44 with confirmed inherited thrombophilia [FVL and/or PT mutations]) and 107 with thrombophilia-negative/unknown status (Figure 1). Within the thrombophilia subgroup, 73 patients were categorized as having major thrombophilia and 33 with minor thrombophilia. The electronic case report form captured data on thrombophilia and thrombophilia types, therefore, all patients for whom thrombophilia status was not captured were automatically considered “thrombophilia negative/unknown.” Of the patients included in the analyses, 91 were previously enrolled in the acute VTE treatment study (47 with thrombophilia and 44 as thrombophilia negative/unknown); supplemental Tables 2 and 3 detail the 35 patients with thrombophilia who rolled over from the DIVERSITY study to the secondary VTE prevention study.

Patient demographics and clinical characteristics at baseline in each study are described in Table 1 (by thrombophilia status) and supplemental Table 4 (by major or minor thrombophilia classification). In the acute VTE treatment and secondary VTE prevention studies, ∼23% and ∼50% of patients had thrombophilia diagnosed, respectively. In both studies, patients with documented thrombophilia were ∼2.5 years older, on an average, than patients with negative/unknown thrombophilia status.

Patients with thrombophilia, and in particular, major thrombophilia, were predominantly male. In the secondary VTE prevention trial, patients with thrombophilia were more likely to have PE as their index VTE event than those with negative/unknown thrombophilia status. In the acute VTE treatment study, a higher proportion of patients had previous VTE in the thrombophilia group than in the thrombophilia-negative/unknown group but rates were similar between groups in the secondary prevention study (Table 2).

Medical history and details of thrombophilia conditions are shown in Tables 2 and 3, respectively (by thrombophilia status), and supplemental Table 5 (by major or minor thrombophilia classification). Presence of a CVC was recorded for a greater proportion of the thrombophilia-negative/unknown groups. In the secondary VTE prevention study, a greater proportion of patients with thrombophilia than those negative/unknown had other conditions requiring secondary VTE prophylaxis (including recurrent unprovoked VTE or structural venous abnormality).

In the secondary VTE prevention study, 24 patients (23.1%) with thrombophilia had a history of PTS compared with 12 (11.3%) in the thrombophilia-negative/unknown group. Of the 73 patients with major thrombophilia and 33 patients with minor thrombophilia, 12 (16.9%) and 12 (36.4%) patients, respectively, had a history of PTS.

By day 84 (or end of therapy [EOT]), few patients experienced progression of index thrombus. Thrombus progression was experienced by 8.1% of patients with documented thrombophilia (Table 4), mostly those with major thrombophilia (6.3% of patients with confirmed inherited thrombophilia [FVL and PT mutations]; Table 5; supplemental Table 6) and by 2.0% of patients with negative/unknown thrombophilia status (Table 4). The VTE recurrence rate was 12.9% among patients with thrombophilia (mostly those with major thrombophilia; supplemental Table 6) and 2.9% for those with negative/unknown thrombophilia status (Table 4; Figure 2A). For the thrombophilia group, 5.1% of patients treated with dabigatran and 13.0% treated with SOC experienced thrombus progression, and 7.7% and 21.7%, respectively, experienced VTE recurrence (Table 4). For these patients, partial thrombus resolution was achieved in 43.6% and 34.8% of patients treated with dabigatran and SOC, respectively, and complete thrombus resolution in 35.9% and 21.7% of patients, respectively. For the high-risk group of patients with major thrombophilia, the frequency of thrombus progression was 10.0% for those treated with dabigatran and 11.8% for those with SOC, and the VTE recurrence rate was 15.0% and 23.5%, respectively (supplemental Table 6). The proportion of patients with VTE outcomes was similar between treatment groups within the negative/unknown thrombophilia group.

Complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death (composite primary end point) was achieved in 19 out of 62 patients (30.6%) in the thrombophilia group (9/32 patients [28.1%] with confirmed inherited thrombophilia) compared with 100 out of 205 (48.8%) in the thrombophilia-negative/unknown group (Tables 4 and 5). There was no notable difference between major (10/37, 27.0%) and minor (9/25, 36.0%) thrombophilia groups (supplemental Table 6). In the thrombophilia subgroup, the treatment comparison showed noninferiority of dabigatran to SOC for this end point, and it was achieved by numerically more patients treated with dabigatran than those treated with SOC: 35.9% vs 21.7% (Mantel-Haenszel–weighted difference in rates for SOC minus dabigatran, −0.135; 95% confidence interval (CI), −0.36 to 0.08; P for noninferiority = .0014; Table 4; Figure 3). For patients with confirmed inherited thrombophilia (n = 32), 34.8% and 11.1% treated with dabigatran and SOC, respectively, met this end point (P for noninferiority = .0009; Table 5). For the subgroup of patients with major thrombophilia (n = 37), 25.0% treated with dabigatran and 29.4% treated with SOC met the primary end point (P for noninferiority = .1806; supplemental Table 6). In patients with minor thrombophilia (n = 25), 47.4% treated with dabigatran achieved the composite primary end point vs 0% treated with SOC (noninferiority P < .0001). Dabigatran was also noninferior to SOC in patients with thrombophilia-negative/unknown status: 48.6% vs 49.3% (Mantel-Haenszel–weighted difference, 0.002; 95% CI, ‒0.14 to 0.15; noninferiority P = .0033; Table 4; Figure 4).

In the acute VTE treatment study, within the treated set, PTS (newly identified or worsening from baseline) was reported as an AE in 2 out of 62 patients (3.2%) in the thrombophilia group (1 each with minor and major thrombophilia) and 5 out of 204 (2.4%) in the thrombophilia-negative/unknown group (Table 4).

Bleeding events were observed in 23.0% of patients with documented thrombophilia (31.3% with confirmed inherited thrombophilia) and in 21.0% of patients with negative/unknown thrombophilia status (Tables 4 and 5), and in 18.9% and 24.0% with major and minor thrombophilia, respectively. Within the thrombophilia subgroup, 17.9% of patients treated with dabigatran compared with 26.1% treated with SOC had bleeding events (Table 4; Figure 4A; by thrombophilia status), although the difference was not statistically significant (P = .66). For patients with confirmed inherited thrombophilia, 26.1% treated with dabigatran and 44.4% treated with SOC experienced bleeding events (Table 5). No patients with thrombophilia had an MBE over the course of the study. Among patients with negative/unknown thrombophilia status, bleeding events were similar between the SOC and dabigatran treatment groups (Table 4; Figure 4A).

VTE recurrence at 12 months was reported for 3 out of 106 patients (2.8%) with thrombophilia (2.3%, 1/44 patients with confirmed inherited thrombophilia) compared with 0 out of 107 (0%) with negative/unknown thrombophilia status, with no discernible difference between major and minor thrombophilia subgroups (Tables 6 and 7; Figure 2B). VTE recurrence at 12 months was reported for 2 out of 73 patients (2.7%) with major thrombophilia (Table 6).

In the secondary VTE prevention study, newly identified or worsening of baseline PTS was reported as a study outcome at 12 months by 3 out of 106 patients (2.8%) with thrombophilia and 0 out of 107 patients without thrombophilia (Table 5).

The proportion of patients with any bleeding while on treatment was numerically higher for patients with thrombophilia (29/106, 27.4%; and 16/44, 36.4% with confirmed inherited thrombophilia) compared with those with negative/unknown thrombophilia status (19/107, 17.8%; Figure 4B; Tables 6 and 7). In the thrombophilia group, this included 2 patients (1.9%) with CRNM bleeding and 1 (0.9%) with major bleeding. In the thrombophilia-negative/unknown group, 1 patient (0.9%) had CRNM bleeding, and 2 (1.9%) had major bleeding. No patients died from bleeding during the study (supplemental Table 8). Bleeding incidence was similar between major and minor thrombophilia subgroups (19/73, 26.0%; and 10/33, 30.0% at 12 months, respectively; Table 6).

None of the patients with thrombophilia who rolled over from the acute VTE treatment study (22 treated with dabigatran and 13 with SOC) and continued anticoagulation by dabigatran in the long-term secondary VTE prevention study experienced MBEs (supplemental Table 3).

In the acute VTE treatment study, rates of AEs, serious AEs, and AEs leading to treatment discontinuation were similar across subgroups (supplemental Table 7). The most common AEs included headache, nasopharyngitis, alopecia, and epistaxis.

In the secondary VTE prevention study, rates of AEs, serious AEs, and AEs leading to treatment discontinuation were higher among patients with thrombophilia (supplemental Table 8). The most common AEs included headache, nasopharyngitis, dyspepsia, and upper respiratory tract infection.

Adherence with study medication was routinely high in both studies. In the acute VTE treatment study, adherence exceeded 98% in all subgroups of thrombophilia status, and in the secondary VTE prevention study, adherence exceeded 96% in all subgroups.

Two large studies, 1 phase 2b/3 and 1 phase 3, have demonstrated the effectiveness and safety of dabigatran for the treatment of acute VTE and for secondary prevention of VTE in children.^21,22 This analysis from the same studies examined patient characteristics and outcomes in the subgroup of children with thrombophilia. It should be noted that testing for thrombophilia was not a protocol-specified requirement for either study. However, for those children with known thrombophilia, and also for those newly tested for thrombophilia markers, the assignment for thrombophilia tests and the interpretation of results was the responsibility of the treating pediatrician, who, in most cases, specialized in pediatric hematology and, therefore, was privy to international definitions on inherited and acquired thrombophilia in children.^34-36 Furthermore, collected and interpreted thrombophilia test results were verified with patients’ source data during regular monitoring procedures; 100% source data verification should be conducted according to the monitoring plan of both trials. Therefore, we have a reasonable degree of certainty that the subgroup classified as thrombophilia positive has been correctly identified.

Although the frequency of unfavorable VTE outcomes such as recurrence and progression of index thrombus were low in the acute VTE treatment study, the current analysis of data showed a higher proportion of patients with thrombophilia compared with those with negative/unknown thrombophilia status experienced these outcomes. Consequently, complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death (primary end point) was achieved by a lower proportion of patients in the thrombophilia group (but similarly for major and minor subgroups) than those in the thrombophilia-negative/unknown status group. For the thrombophilia-negative/unknown status group, there was no obvious difference in VTE outcomes between dabigatran and SOC treatment groups. Of note, we observed that a numerically higher proportion of patients who were thrombophilia positive, and particularly those with confirmed inherited thrombophilia with FVL and/or PT mutations, experienced complete thrombus resolution, freedom from recurrent VTE, and freedom from VTE-related death with dabigatran vs SOC. Similarly, numerically lower rates of VTE recurrence and thrombus progression, and higher rates of partial thrombus resolution and complete thrombus resolution were observed in patients from the thrombophilia-positive group treated with dabigatran vs SOC. Although differences between treatment cohorts did not reach statistical significance, this is not unexpected because of the small sample sizes. Irrespective of thrombophilia status, consistency of effect with the overall study results,²¹ in terms of noninferiority of dabigatran to SOC, was observed for the primary end point. Similar proportions of patients with and without thrombophilia experienced bleeding events. For the thrombophilia group, numerically fewer patients treated with dabigatran experienced bleeding events, including major and CRNM bleeding events, compared with those treated with SOC, although no statistical differences were observed between treatments. In the negative/unknown thrombophilia group, frequency of bleeding events was similar between treatment groups.

In the analysis of data from the secondary VTE prevention study, higher rates of VTE recurrence and any bleeding events were observed among children with thrombophilia compared with children in whom thrombophilia status was negative/unknown. However, rates of major or CRNM bleeding were low, with no clear difference in rates according to thrombophilia status.

The findings of greater VTE risk associated with presence of thrombophilia are broadly in line with previous observational studies of VTE recurrence in pediatric populations with so-called major thrombophilia, particularly in non–CVC-related events. Of note, only ∼15% (dabigatran arm) and ∼22% (SOC arm) of patients in the DIVERSITY study, and only ∼3% of patients enrolled in the secondary prophylaxis study, had central line–related events.^21,22 For instance, in a previous German-wide national pediatric study, the presence of single vs combined prothrombotic defects was both associated with higher odds of VTE recurrence (single defect: odds ratio, −4.6; 95% CI, −2.3 to 9.0; P < .0001; combined defects: odds ratio, −24.0; 95% CI, −5.3 to 108.7; P < .0001).⁵ In terms of adult data, a post hoc analysis of dabigatran studies in adults (ie, phase 3 studies comparing dabigatran and warfarin, RE-COVER [#NCT00291330]³⁷ and RE-COVER II [#NCT00680186]³⁸ for acute symptomatic VTE, and RE-MEDY [#NCT00329238]³⁹ for secondary VTE prevention) revealed no difference in symptomatic VTE recurrence/VTE-related deaths between patients with thrombophilia treated with dabigatran or warfarin, with a similar safety profile.⁴⁰ More recent adult reports, including a systematic review, confirmed a similar efficacy of DOACs, including dabigatran, as an alternative anticoagulation strategy for patients with thrombophilia, except for cases of heparin-induced thrombocytopenia, APLA (eg, triple positive), and paroxysmal nocturnal hemoglobinuria, not included in our report because of their rarity in children.^41-43 To date, comparable evaluations of the performance of other DOACs in children with venous thrombotic events associated with thrombophilia are scant but are starting to emerge. A previous randomized trial on rivaroxaban in children, evaluating its efficacy and safety for acute VTE treatment, included 32 patients with inherited thrombophilia (intervention arm, 27 patients [8%]; SOC, 5 patients [3%]). However, no sensitivity analysis could be conducted because of its limited sample size.⁴⁴ 

If the signal for a differential effect of dabigatran vs SOC on residual thrombus burden translates into meaningful differences in outcomes, this would be expected to confer benefits in reducing VTE recurrences and PTS. In adults, persistence of thrombosis despite a course of anticoagulation is a predictor of VTE recurrence and PTS;^45,46 for example, a lack of thrombus resolution was associated with a statistically significant fourfold increase in the odds for PTS in the original pediatric cohort reporting the modified Villalta scale, 1 of the 2 pediatric PTS scales accepted by the International Society on Thrombosis and Haemostasis.^30,47 Nonetheless, a recent cross-sectional follow-up study of adults diagnosed with VTE randomized to receive either dabigatran or warfarin did not identify a difference in prevalence of PTS in either treatment group.⁴⁸ The generalizability of this finding to the pediatric population remains to be proven. Consistent with our results, pediatric studies have shown patients with CVC-related DVT to be younger than those with non–CVC-related DVT.⁴⁹ PTS predictors included increased residual DVT burden, which is relevant for our study findings.⁵⁰ For non-catheter–related DVT, like our findings, patients had thrombophilia more often, along with DVT recurrences. Furthermore, studies in adults have shown an increased risk of persistence of residual thrombus in patients with thrombophilia, highlighting the need to further investigate the role of DOACs in children with thrombophilia to prevent unfavorable VTE-related outcomes.⁵¹ 

Considering the limited data available on the topic of DOACs in children with thrombophilia, we believe that the data summarized herein provides unique hypothesis-generating findings. We acknowledge that this analysis has some limitations. There is no generally accepted classification for categorizing thrombophilia in major and minor subgroups, but similar approaches have been used previously.^11,31,52 Major thrombophilia includes inherited and acquired coagulopathies that increase coagulability and frequency of thrombotic events. Minor thrombophilia includes other mutations and conditions with less clear clinical consequences. We have provided a detailed listing of coagulopathies and categorization into major/minor in the manuscript, as well as of those patients with confirmed inherited thrombophilia with FVL and/or PT mutations. Thrombophilic coagulopathies were prespecified and captured on the medical history page of the electronic case report forms, but there were no requirements to perform any tests concerning thrombophilia (including repeat, genetic, and/or family investigations) after a patient had been enrolled. However, FVL/PT mutations were confirmed by genetic tests (source-verified data were entered into the electronic case report forms) and, therefore, we were able to identify patients with confirmed inherited thrombophilia. Rare coagulopathies or mutations and conditions with indefinite thrombophilic significance were entered in the case report form as free text. Concentrations of antithrombin, protein C/S, and APLA/LA were not captured in a standardized manner, nor were types of APLA (immunoglobulin, subclasses G or M), or whether triple-positive status was present. As mentioned, we derive confidence in the thrombophilia diagnoses because these were based on the judgment of the pediatrician, who, in most cases, specialized in pediatric hematology and was familiar with the concept of developmental hemostasis and age-specific levels of thrombophilic markers. A sample selection bias might be present because testing for thrombophilia status was not protocol mandated. To mitigate this issue, a sensitivity analysis comparing patients with negative vs unknown thrombophilia status did not identify differences on the patient characteristics of both groups. Another limitation is the small number of patients in some of the subgroups analyzed herein; hence, our findings should be considered exploratory and be interpreted with caution.

Given that thrombophilia is associated with a higher risk of long-term complications of VTE, it is important to establish appropriate and stable treatments for prolonged duration in this patient group. The exploratory findings of this study suggest that dabigatran could potentially have improved efficacy in these higher risk patients with thrombophilia compared with those without thrombophilia, who are at lower risk. Taken together, these data suggest that dabigatran could be appropriate for long-term anticoagulation in pediatric patients with thrombophilia. Future studies might add to the evidence on the treatment effects of DOACs, such as dabigatran vs SOC.

The authors thank Ingrid Gergei and Martina Brueckmann for their contributions to this study; Sven Wichmann, Carolyn Cook, and Abu Sami for programming support; and Alison Monckton for data management support. Medical writing assistance was provided by Keith Day and Emma Beddie of Parexel. The authors would also like to thank the patients who participated in the studies who contributed data to this analysis and to their families, along with the investigators, study coordinators, study teams, and nurses.

The study was supported financially by Boehringer Ingelheim International GmbH.

Contribution: L.R.B., I.T., M.A., L.B., E.C., L.G.M., M.L., M.S., Z.S., M.B., and J.H. contributed to the concept, design and/or analysis of the data; L.R.B., M.A., L.B., E.C., L.G.M., M.L., P.S., J.F., O.L., and J.H. were involved in data collection; and all authors contributed to interpretation, critical writing, or revising of the intellectual content, and final approval of the version to be published.

Conflict-of-interest disclosure: L.R.B. is a member of a pediatric advisory board for Boehringer Ingelheim, and reports fees to his institution from Boehringer Ingelheim. I.T., M.S., and Z.S. are employees of Boehringer Ingelheim. M.A. is a member of a pediatric advisory board for Boehringer Ingelheim and has received advisory board fees from Daiichi Sankyo. J.H. is a member of a pediatric advisory board for Boehringer Ingelheim and has received honoraria from Boehringer Ingelheim for congress presentation. L.B. is a member of a pediatric advisory board for Boehringer Ingelheim and reports fees to her institution from Janssen Pharmaceuticals. E.C. is a member of a pediatric advisory board for Boehringer Ingelheim and reports personal fees from Roche, Sobi, Bristol Myers Squibb, CSL Behring, and Shire/Takeda. M.L. is a member of a pediatric advisory board for Boehringer Ingelheim. P.S. and J.F. significantly contributed to the enrollment of patients with thrombophilia in the program in their role as investigators (pediatric hematologists). O.L. significantly contributed to the enrollment of patients with thrombophilia in the program (investigator; pediatric hematologist) and has an Investigator Agreement with Bristol Myers Squibb, Novartis, and UCB. L.G.M. is a member of a pediatric advisory board for Boehringer Ingelheim and has received a research grant from Bristol Myers Squibb.

A complete list of the DIVERSITY and Secondary VTE Prevention Study Group investigators appears in the supplemental data.

Correspondence: Leonardo R. Brandão, Division of Haematology/Oncology, Department of Paediatrics, The Hospital for Sick Children, University of Toronto, 555 University Ave, Toronto, ON M5G 1X8, Canada; e-mail: leonardo.brandao@sickkids.ca.