Myelodysplastic Syndrome (MDS) and Myeloproliferative Disease (MPD) in Children: A Prospective Registration of 222 Cases

MDS and MPD are rare in children, however, they are important because the diagnosis is difficult and the prognosis is generally very poor. We conducted a prospective registration of patients suspected of having MDS employing pathological central review through the MDS Committee of the Japanese Society of Pediatric Hematology. Of total, 222 children were enrolled between 1999 and 2006: 96 cases of primary MDS (RA 52, RAEB 25, RAEB-T 19, by FAB classification; RA 28, RCMD 24, RAEB-1 11, RAEB-2 28, AML 5 by WHO classification), 71 cases of JMML (excluding patients with Noonan syndrome and NF1), 1 case of CMML, 19 cases of therapy-related MDS (t-MDS), and 35 cases with constitutional predisposition to MDS. There were128 male and 94 female. The median age at diagnosis was 4.5 years (range, 0.0–19.2 years): primary MDS, 6.8y (0.3–17.4); JMML, 1.5y (0.1–7.1); t-MDS, 11.9y (2.8–19.2); Constitutional MDS, 2.1y (0.0–14.7).The median follow up for 144 surviving patients was 39 months (range 1 to 118 months) after diagnosis. The 3-year overall survival (OS) estimated by Kaplan-Meier analysis, for primary MDS, JMML, t-MDS and constitutional MDS was 73.8%, 62.4%, 69.7% and 65.1% respectively. In primary MDS classified by FAB, OS for RA, RAEB and RAEB-T was 95.9%, 44.7% and 39.5% respectively, while in primary MDS classified by WHO, OS for RA, RCMD, RAEB-1 and RAEB-2 was 100%, 91.0%, 61.4% and 35.0%. Thus, the WHO system could also be employed in children. The data of karyotype were available in 214 of 222 patients (96%). Cytogenetic abnormality was noticed in 55% of primary MDS (40% RA, 72% RAEB, 72% RAEB-T), 27% of JMML and 83% of t-MDS cases. Totally-7/7q-was the most frequent aberration, followed by complex (more than 3 aberrations) and trisomy 8. The del(5q), del(20q), and –Y, which were frequent chromosomal aberrations in adult MDS were rare in children. −7/7q-were observed in primary MDS (27% of abnormal karyotypes) and in JMML (53%). In primary MDS, OS for −7/7q-, complex, others and normal were 50.0%, 25.0%, 79.9% and 86.0%. Thus, the cytogenetic classification defined by the IPSS was useful to predict outcome in childhood primary MDS, however, it was not the case in patients with JMML. Regarding treatment, patients with RA or RCMD were divided into 3 categories: HSCT in 25 (including HSCT after the immunosuppressive therapy [IST] in 6), IST in 15 and no treatment in 10. OS for these categories were 91.6%, 100% and 100%. There were 8 cases of RA with −7/7q-. Six of the 8 cases underwent HSCT but 4 died. The other two received IST and they are long term survivors. The IST therapy could be an option for those with −7/7q-. Patients with RAEB/RAEB-T were divided into 3 categories: HSCT in 12, the AML-type chemotherapy followed by HSCT in 22, and the AML-type chemotherapy in 4. OS for these categories were 53.6%, 28.7% and 37.5%. In HSCT group (n=12), 4 had −7/7q- or complex karyotype. Regarding the donor type, related donors were used in 7 cases. In AML/HSCT group (n=22), 5 had −7/7q- or complex karyotype. Related donors were used only in 6 cases. Chemotherapy induced complete remission (CR) in 11 patients in this group. OS for the patients who attained CR was 40.5% while OS for the others was only 16.0%. Thus, AML-type chemotherapy seemed essential to cure a subset of patients with RAEB/RAEB-T although HSCT without AML-type chemotherapy maybe justified when related donors are available. Regarding JMML, OS (62.4%) was as good as that in Europe. Although we obtained important information, we could not define an optimal management for patients in each subtype of the diseases. Since MDS and MPD in childhood are very rare, an international cooperation is urgently needed to improve treatment outcome.