Impact of Thromboembolism on Overall and Event-Free Survival in Children with Acute Lymphoblastic

Introduction: Thromboembolism (TE) is a well-known complication of cancer and its treatments. The impact of TE on survival outcomes remains unclear, especially in children. We assessed whether TE development was associated with overall survival (OS) and event-free survival (EFS) in children with acute lymphoblastic leukemia (ALL).

Methods: We performed a population-based retrospective cohort study using the national registry Cancer in Young People Canada (CYP-C). Children 0-<15 years of age diagnosed with ALL (2000-2018) and treated at one of 12 Canadian pediatric centers outside of Ontario were included. OS was defined as the time between the date of cancer diagnosis and death and, EFS, as the time between the date of cancer diagnosis and the date of relapse, subsequent malignancy or death (whichever came first). Patients were categorized as to whether they experienced a radiologically-confirmed TE during treatment graded 3, 4 or 5 as per the Common Terminology Criteria for Adverse Events v.4 (i.e. requiring medical treatment, life-threatening or fatal). Only TEs that occurred before relapse or subsequent malignancy were considered. The Kaplan-Meier survival method estimated the 5-year OS and EFS of children with TE compared to those without TE. Univariate and multivariable Cox regression models were used to estimate the hazard ratio (HR) and 95% confidence interval (CI) of death or an event between groups, adjusted for age, sex, and leukemia risk group. A sub-analysis stratified the analysis by leukemia risk group.

Results: The study included 2,208 children (median age: 4 years [interquartile range: 2-7 years], 54.9% male). Precursor B-cell ALL was the most common diagnosis (1,789, 89.1%). Patients were stratified as standard/low risk ALL in 58.0% of cases, and high/very high risk ALL in 42.0%. Of these, 121 (6.0%) developed a TE, at a median time of 100 days (interquartile range: 30-183 days) after cancer diagnosis. Eight patients (0.4%) had a life-threatening or fatal TE. Patients with TE were more likely to be aged 10 years or older, to present with T-cell ALL, and to have high risk leukemia.

The 5-year OS (95% CI) of patients with and without TE was 80.2% (72.9-87.5%) and 93.7% (92.5-94.9%) respectively (log-rank test: p<0.001, Figure 1). The adjusted HR (95% CI) of death in children with TE was 2.09 (1.33-3.27, p=0.001).

Similarly, as shown in Figure 2, the 5-year EFS (95% CI) of patients with and without TE was 68.7% (59.7-77.7%) and 88.6% (87.1-90.1%), respectively (log-rank test: p<0.001). The adjusted HR (95% CI) of an event was 2.01 (1.39-2.90, p<0.001).

When stratified by leukemia risk group, no statistically significant difference was seen in standard/low risk ALL for both OS and EFS but TE was associated with a significantly lower OS and EFS in children with high/very high risk ALL (Table 1). In this group, the increased risk of death was attributable to both deaths following relapsed disease (HR [95% CI]: 2.37 [1.39-4.04]) and death not following relapse (HR [95% CI]: 2.93 [1.35-6.35]). Sensitivity analyses in which 1) patients with very high risk ALL were removed and 2) only grade 3 or 4 TE were considered showed similar results.

Conclusions: Clinically relevant TE led to a statistically significant reduction in OS and EFS in children with high risk/very high risk leukemia. Further research is needed to assess whether TE prevention may improve anti-cancer outcomes.