Incidence and Outcomes of Pediatric Myelodysplastic Syndrome in the US

Abstract

Background There is lack of epidemiological data on pediatric myelodysplastic syndrome (p-MDS) in the literature. MDS became reportable to the Surveillance, Epidemiology and End Results (SEER) Program in 2001, providing an opportunity to estimate the incidence and survival of pediatric patients with MDS in the United States. Methods We utilized data from the National Cancer Institute SEER-18 to determine the incidence and long term overall survival (OS) of pediatric patients (ages 0 to 20 years) diagnosed with de novo MDS or therapy-related MDS. Inclusion criteria was diagnosis of MDS (International Classification of Diseases-Oncology, Third Edition, ICD-O-4 codes 9980/3, 9991/3, 9992/3, 9982/3, 9985/3, 9983/3, 9986/3, 9986/3, 9989/3, 9985/3, 9975/3, and 9987/3) between 2001 and 2011. Follow up was updated through the end of 2011 (November 2013 submission). Overall survival was estimated using the method of Kaplan-Meier. A Cox proportional hazard model was used to compare the effects of age, race, gender, histological subtype, and etiology (de novovs. therapy-related) on survival. Results The incidence of p-MDS was 1.16 cases/1 million population*year. A greater incidence occurred in children younger than 1 year of age possibly reflecting congenital bone marrow failure syndromes (Figure 1). A total of 314 p-MDS cases were included in the analysis with median follow up of 31 months (range 0-131). Median age of patients was 9 years; 167 (53.3%) had MDS unclassifiable (NOS), 40 (12.7%) had therapy-related MDS (t-MDS), 44 (14%) had refractory anemia with excess blasts (RAEB), 32 (10.3%) had refractory anemia (RA), 17 (5.4%) had refractory cytopenia with multilineage dysplasia (RCMD), 6 (1.9%) had refractory anemia with ring sideroblasts (RARS), 5 (1.6%) had refractory anemia with excess blasts in transformation (RAEBT), and 3 (0.9%) had MDS associated with isolated del(5q). Male patients comprised 154 (49%) of cases. Racial groups included white (218, 69.4%), 52 (15.7%) black, 37 (11.8%) of other races, and 7 (2.3%) the race was unknown. The 5 year-OS for the entire cohort was 68% (95% C.I.=62.3-73.7). Patients with t-MDS had significantly worse 5 year-OS (41.2%; 95%C.I.=23.8-58.6) compared to those with de novo MDS (71.3%; 95%C.I.=65.3-77.2; P=0.004, Figure 2). In multivariate analysis of age, race, gender, histological subtype, and etiology (de novovs. therapy-related) utilizing Cox regression model, only t-MDS was associated with higher risk of death (HR=2.07, 95% C.I.=1.25-3.42, P=0.005). Conclusions Pediatric MDS is a rare disorder, with higher incidence among children younger than 1 year of age. Over two thirds of p-MDS patients will become long-term survivors, although significantly inferior outcome is seen in t-MDS. Figure 1 Figure 1. Figure 2 Figure 2. Disclosures No relevant conflicts of interest to declare.