Discussion
A global CSVT rate of 2.9% was reported in a meta-analysis of 17 prospective studies that included 1752 pediatric patients with childhood ALL [7]. Two recent studies have reported incidences of 1.4% and 1.9% from the United Kingdom and the Nordic Society of Pediatric Hematology and Oncology (NOPHO) [2, 8]. However, we report an incidence of 10.5% for CSVT during treatment of our ALL patients which is much higher than that of the literature. This may be related to the treatment protocol that we use, in addition to our patients’ characteristics [9]. Carouso et al. found that a combination of ASP and steroids in the presence of other prothrombotic factors like inherited thrombophilia were significantly associated with thrombosis (OR: 34.5; 95% CI: 4.39–271.42; P = 0.0008) [7]. These studies have reported the incidence of thrombosis during ALL treatment to be highest during the induction phase, whereas in our studies, most CSVT events occurred during early continuation and reinductions I and II. This might be explained by the lower prednisone dose that we use during the induction phase compared to the BFM protocols reported by Nowak-Gottl et al. [10] (40 mg/m2/day vs. 60 mg/m2/day). However, we use higher doses of dexamethasone during early continuation and reinduction (12 mg/m2/day vs. 10 mg/m2/day). Moreover, the duration of the combined treatment (steroids + ASP) given was longer in the post-induction phase compared to the induction phase.
In Lebanon the carrier rate of factor V Leiden is 14.4% compared to 3–8% in Europe and the United States [11]. Similarly, the carrier rate of Prothrombin G20210A in Lebanon is 3% similar to that in Europe but higher than what is reported in the United States [9]. There is no solid evidence for screening all ALL children during treatment for heritable thrombophilia [12, 13]. Since such screening is of a high financial burden in a middle-income country like Lebanon, we screened only patients who developed CSVT. The lack of thrombophilia status in patients who did not develop CSVT prevented us from investigating thrombophilia as a risk factor and integrating it into our analysis.
Nonetheless, due to the higher incidence of CSVT in our study combined with the high thrombophilia rate in our country, we suggest revisiting the role of thrombophilia screening in regions with high prevalence of thrombophilia, using well-designed studies, especially if their treatment protocol relies on intensive combined ASP and steroids therapy [4].
The strong association between CSVT during childhood ALL treatment and older age at diagnosis, T-cell immunophenotype, and an intermediate/higher risk disease that is observed in our univariate analysis has been well described previously [2, 10]. In our study, T cell phenotype was only significant in the univariate analysis, whereas intermediate/high risk disease and mediastinal mass retained significance in multivariate analysis.
Studies have linked asparaginase and corticosteroids treatment to the increased TG level in ALL patients [14, 15]. Hypertriglyceridemia induced by asparaginase was recently reported to be significantly associated with increased thrombosis in childhood ALL [16]. We describe a maximum TG level of more than 615mg/dL (3x the upper normal limit) to be a statistically significant risk factor for the development of CSVT during treatment for childhood ALL. Our study was the first to report a TG level of more than 615 mg/dL to be a significant cutoff for the development of CSVT. Further studies with larger sample sizes are recommended to confirm this cutoff.
Our study was the first to describe a significant association between BSA and CSVT. BMI is a measure used to determine a person’s degree of overweight. BSA measures the total surface area of the body and is used to calculate drug dosages and medical indicators or assessments. When height value is fixed, a strong correlation exists between BSA and BMI as only weight values vary. However, when the two parameters change, no correlation exists between BSA and BMI [17] which explains why BSA and not BMI was significant in our study. Larger BSA requires higher doses of therapeutic drugs and hence possibility of more frequent side effects. Similar findings were observed in a previous study where moderate dose methotrexate caused toxicity in patients with larger body size upon treating non-Hodgkin’s lymphoma [18]. Further studies on larger patient samples are recommended to find an appropriate cutoff for BSA. This may help in considering prophylactic anticoagulation and/or capping of chemotherapy doses.
A previous study described the link between the presence of mediastinal mass in Nordic children with ALL and overall thrombotic events in general but not CSVT specifically [10]. To the best of our knowledge we are the first to describe the presence of mediastinal mass as a significant risk factor for the development of CSVT in children treated for ALL.
Initial blast count and BMI were not significant as risk factors for CSVT in our study. These results are supported by the Nordic study while exploring their relation to any thrombotic event [10].
Non-O blood group has been reported to be an important genetic risk factor for venous thromboembolism [19]. In our previous report, we reported lack of significance of non-O blood group for the development of CSVT in our patient population [4]. This factor was reinvestigated in our population and found to be not significant.
Given the significant delay that CSVT has on treatment, and the higher chances of disease relapse from early discontinuation of ASP as reported from Dana–Farber Institution [3], we believe that this issue is of significant importance and needs further investigation and future guideline modifications in areas where thrombotic/CSVT incidence is high. The incidence of CSVT is protocol specific such as use of prednisone or dexamethasone and their respective doses, the intensity and duration of ASP therapy and concomitant (Asparaginase plus steroid) therapy, as well as ethnic specificities. In a patient population similar to ours that is being treated with a protocol similar to St. Jude Total XV, we encourage considering the use of prophylactic anticoagulation in the presence of the risk factors that we highlighted. Studies with larger sample sizes are needed to confirm our recommendation. Furthermore, we recommend additional studies in different patient populations to tailor recommendations based on the population characteristics.