Children with cancer have an increased risk of developing thrombo-embolism because of the underlying disease and/or treatment, with prevalence ranging from 16 to 40% (Barg, 2020). Numerous risk factors for VTE in pediatric cancer patients have been identified, including cancer type, chemotherapy and central venous catheters. Children with lymphoma or acute lymphoblastic leukemia have a much higher risk than children with a solid tumor or brain tumor. Medication such as asparaginase and steroids increase the risk of thrombosis. The central venous catheter leads to an increased risk of thrombosis. Current guidelines do not advice standard thrombosis prophylaxis to children with cancer. The hypothesis is that thrombosis prophylaxis could be used in high-risk groups (teenagers, type of cancer, CVL, type of treatment).

General remarks

These studies included children with hematological malignancies. Therefore, the results might not be applicable to all children with malignancy who are initiating systemic anticancer treatment.

Venous thromboembolism

LMWH

DOAC

Major bleeding

Mortality

DOAC

LMWH

Catheter related thrombosis, post-thrombotic syndrome, quality of life, discontinuation of oncological treatment and postponing oncological treatment

Description of studies

O’Brien (2024) performed a phase 3, open-label, randomized, controlled trial which was conducted in 74 pediatric hospitals in nine countries. In total, 512 patients aged one to 18 years with newly diagnosed acute lymphoblastic leukemia (pre-B cell or T cell) or lymphoblastic lymphoma (B cell or T cell immunophenotype) and a central venous line in place throughout induction, were included in the study. The 512 patients were randomly assigned to receive either SOC (n=256), or SOC plus prophylactic apixaban (n=256), daily for 29 days. The prophylactic apixaban dose regimen was age- and weight-based. The two groups were similar in their proportion of gender and acute lymphoblastic leukemia risk category (standard or high risk). The median follow-up was 27 days (IQR 26-28). The following relevant outcomes were reported: VTE, major bleeding, and mortality.

Rutjes (2020) included several RCT’s in children, of which one fulfilled our inclusion criteria (Greiner, 2019). Greiner (2019) performed an open-label randomized controlled study which was conducted in 26 pediatric cancer centers in Germany and Switzerland. In total, 949 patients aged one to 18 years with newly diagnosed acute lymphoblastic leukemia were enrolled and randomized to receive low-dose unfractionated heparin (UFH, n=312), prophylactic low molecular weight heparin (enoxaparin) (n=317), or activity-adapted antithrombin (n=320) throughout induction therapy. The antithrombin group was not included in our literature analysis. Patients randomized in the UFH-group received standard of care, consisting of a low-dose UFH 2 IU/kg bodyweight/hour while the CVC was in use. Patients randomized in the enoxaparin group received enoxaparin at a dose of 80–100 IU/kg bodyweight once daily subcutaneously with a target anti-Xa level not exceeding 0.4 U/L, measured four hours after the third or fourth injection. The patients randomized in the UFH-group and the enoxaparin-group were comparable at baseline. The following outcomes were reported: symptomatic VTE, major bleeding, and survival.

Results

Venous thromboembolism

Total

O’Brien (2024) reported the outcome VTE, defined as the composite of non-fatal deep venous thrombosis (symptomatic or clinically unsuspected), pulmonary embolism, cerebral sinus venous thrombosis, and venous thromboembolism related death. VTE events were first confirmed by blinded adjudication solely based on imaging results and then further classified as symptomatic or clinically unsuspected based on the investigator’s report. Symptomatic VTE events were included in the analysis if they occurred during the intended treatment period whereas asymptomatic VTE events were included in the analysis when screening research imaging studies were completed by day 40. No VTE-related death or pulmonary embolism occurred and there was only one case of cerebral sinus venous thrombosis in the SOC-group. In total, VTE-events occurred in 31 (12%) of the 256 patients receiving apixaban compared to 45 (18%) of the 256 patients receiving SOC. The risk ratio (RR, 95%CI) was 0.69 (0.45 to 1.05). The risk difference (RD, 95%CI) was -0.04 (-0.10 to 0.01, which was in favor of the apixaban group and considered to be clinically relevant.

Greiner (2019) reported on the outcome symptomatic VTEs, defined as symptomatic VTE and diagnosed based on clinical suspicion and confirmed by one or more suitable imaging methods within a routine diagnostic work-up.

Symptomatic VTE

O’Brien (2024) reported 4 (2%) non-fatal symptomatic VTE’s in the apixaban group (N=256), compared to 6 (2%) VTEs in the SOC-group. The RR (95%CI) was 0.67 (0.19 to 2.33). Corresponding RD (95%CI) was -0.01 (-0.03 to 0.02), which was in favor of the apixaban group and not considered to be clinically relevant. Besides, one of the patients in the SOC-group had a fatal cerebral sinus venous thrombosis.

In the study of Greiner (2019) symptomatic VTE occurred in 11 (3.5%) of the 317 patients in the enoxaparin group and in 25 (8.0%) of the 312 patients that received SOC. The RR (95%CI) was 0.43 (0.22 to 0.86). Corresponding RD (95%CI) was -0.05 (-0.08 to -0.01), which was in favor of the enoxaparin group and considered to be clinically relevant.

Asymptomatic VTE

O’Brien (2024) reported 27 (11%) non-fatal clinically unsuspected VTE’s in the apixaban group (N=256), compared to 38 (15%) VTEs in the SOC-group. The RR (95%CI) was 0.71 (0.45 to 1.13). Corresponding RD (95%CI) was -0.04 (-0.10 to 0.01), which was in favor of the apixaban group and considered to be clinically relevant.

Major bleeding

O’Brien (2014) reported on the outcome major bleeding, defined as fatal bleeding; clinically overt bleeding associated with a decrease in haemoglobin of at least 20 g/L (ie, 2 g/dL) in 24 hours; bleeding that is retroperitoneal, pulmonary, intracranial, or otherwise involves the central nervous system; and bleeding that requires surgical intervention in an operating suite, including interventional radiology. Two major bleeding events occurred in each group (RR (95%CI) 1.0 (0.14 to 7.01) and RD (95%CI) 0.00 (-0.02 to 0.02)).

Greiner (2019) also reported on the outcome major bleeding, defined as hemorrhage meeting one or more of the following criteria: clinically evident, fatal, requiring erythrocyte replacement (10-20 ml/kg body weight), located within the cranium/spine, eye,

retroperitoneum (diagnosis using magnetic resonance imaging, computed tomography and/or ultrasound) or severe or life-threatening event resulting from the hemorrhage, therefore requiring intensive care. Major bleeding occurred in one (0.5%) of the 317 patients that received enoxaparin and in four (1.1%) of the 312 patients that received SOC. The RR (95%CI) was 0.25 (0.03 to 2.19). Corresponding RD (95%CI) was -0.01 (-0.02 to 0.00), which was in favor of the enoxaparin group and was considered to be not clinically relevant. 

Mortality

O’Brien (2014) reported on the outcome mortality, with one death occurring in the apixaban group and four deaths in the SOC-group. The RR (95%CI) was 0.25 (0.03 to 2.22). Corresponding RD (95%CI) was -0.01 (-0.03 to 0.01), which was in favor of the apixaban group and considered to be not clinically relevant. In the SOC-group, three participants died due to events that occurred during the treatment-emergent period ((infection [n=1], intracranial hemorrhage [n=1], and cardiac arrest [n=1]), and one due to an infection that occurred outside the treatment-emergent period. In the apixaban group there was one apixaban-related death (coagulopathy and hemorrhage after cardiac arrest of unknown cause).

Greiner (2019) reported on the outcome survival defined as time from diagnosis to the date of last follow-up or death from any cause – only exploratory analysis were available. The probability of overall survival at five years was similar in treatment arms (enoxaparin-group : 90.9±1.6%, SOC-group: 92.4±1.5%).

Catheter related thrombosis, post-thrombotic syndrome, quality of life, discontinuation of oncological treatment and postponing oncological treatment

None of the studies described these outcomes.

Level of evidence of the literature

Venous thromboembolism

Symptomatic

LMWH

Evidence regarding the outcome symptomatic VTE is derived from a RCT and therefore the level of evidence started as high. The level of evidence was downgraded to low. There was risk of bias (open label design; downgraded one level). Furthermore, the 95%CI of the effect estimate crossed one of the thresholds for clinical relevance (imprecision, downgraded one level).

DOAC

Evidence regarding the outcome symptomatic VTE is derived from a RCT and therefore the level of evidence started as high. The level of evidence was downgraded to very low. There was risk of bias (open label design; downgraded one level). Furthermore, confidence interval was very broad (serious imprecision, downgraded three levels).

Asymptomatic

LMWH

Since the included study did not report on the outcome asymptomatic VTE for thromboprophylaxis with LMWH, no GRADE conclusions could be drawn for this outcome.

DOAC

Evidence regarding the outcome asymptomatic VTE is derived from a RCT and therefore the level of evidence started as high. The level of evidence was downgraded to low. There was risk of bias (open label design; downgraded one level). Furthermore, the 95%CI of the effect estimate crossed one of the thresholds for clinical relevance (imprecision, downgraded one level).

Major bleeding

Evidence regarding the outcome major bleeding is derived from RCTs and therefore the level of evidence started as high. The level of evidence was downgraded to very low. There was risk of bias (open label design; downgraded one level) and the confidence interval was very broad (serious imprecision, downgraded three levels).

Mortality

LMWH

Since the included study did not report on the outcome mortality for thromboprophylaxis with LMWH, no GRADE conclusions could be drawn for this outcome.

DOAC

Evidence regarding the outcome mortality is derived from a RCT and therefore the level of evidence started as high. The level of evidence was downgraded to very low, because of a very broad confidence interval (serious imprecision, downgraded three levels).

Catheter related thrombosis, post-thrombotic syndrome, quality of life, discontinuation of oncological treatment and postponing oncological treatment

Since none of the included studies reported on the outcomes catheter related thrombosis, post-thrombotic syndrome, quality of life, discontinuation of oncological treatment and postponing oncological treatment related thrombosis, no GRADE conclusions could be drawn for those outcomes.

A systematic review of the literature was performed to answer the following question: what are the (un)desirable effects of thromboprophylaxis with Direct Oral Anticoagulants (DOAC) or Low-Molecular-Weight Heparin (LMWH) in children with malignancy in whom systemic anticancer treatment was initiated, compared to placebo or standard of care?

Relevant outcome measures

The guideline development group considered mortality, VTE, major bleeding, quality of life and discontinuation of oncological treatment as critical outcome measures for decision making; and catheter related thrombosis, PTS and postponing oncological treatment as an important outcome measure for decision making.

The working group defined the outcome measures as follows:

• Major bleeding: fatal bleeding, and/or symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intra-articular or pericardial, or intramuscular with compartment syndrome, and/or bleeding causing a fall in hemoglobin levels of 1.24 mmol/L (20 g/L or greater) or more, or leading to a transfusion of red cells;

A priori, the working group did not define the other outcome measures listed above but used the definitions used in the studies.

The working group defined the following as a minimal clinically (patient) important difference:

• VTE, major bleeding and mortality: risk difference of 3%*
• For all other outcome measures, the default thresholds proposed by the international GRADE working group were used as a threshold for clinically relevant differences: a 25% difference in relative risk (RR) for dichotomous outcomes (RR <0.8 or RR >1.25), and 0.5 standard deviations (SD) for continuous outcomes.

*Based on the differences applied in the guidelines on thromboprophylaxis in patients with COVID-19. This working group derived the minimal clinically (patient) important differences from the ACCP (2012).

Search and select (Methods)

The databases Medline (via OVID) and Embase (via Embase.com) were searched with relevant search terms from 2013 until 7^th March 2024. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 463 hits. Studies were selected based on the following criteria: (systematic reviews of) RCTs evaluating the effect of thromboprophylaxis with DOAC or LMWH in children with malignancy in whom systemic anticancer treatment was initiated compared to placebo or standard of care. 11 studies were initially selected based on title and abstract screening. After reading the full text, nine studies were excluded (see the table with reasons for exclusion under the tab Methods), and two studies were included (Rutjes, 2020 and O’Brien, 2024).

Results

Two studies were included in the analysis of the literature. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.