Transient and Myeloid Leukemia in Children with Down Syndrome Mosaic

Transient leukemia (TL) occurs in 5 to 10% of newborns with Down syndrome (DS). In almost all cases it resolves spontaneously within 3 months, but 20–25% of the children develop myeloid leukemia (ML-DS) until the age of 4 years. TL and ML-DS can occur also in children without any clinical signs of Down syndrome, but with constitutional trisomy 21 due to mosaicism. It can be difficult to diagnose TL or ML-DS in these children and the treatment strategies have not been defined.

Between 1/2002 and 7/2011, 15 newborns and infants were diagnosed with DS mosaic. Nine of them presented with TL and 8 children suffered from ML-DS; 2 of them with a history of TL (table 1). In children without any stigmata the special morphology and immunophenotype of blasts triggered the screening for GATA1 mutation and trisomy 21 mosaic. Diagnostic work-up was performed according to standard guidelines: morphology, immunophenotyping (IP), cytogenetics and FISH (trisomy 21), molecular genetics (GATA 1 mutation screening). Screening of GATA1 mutations was done with direct sequencing of PCR product (Exon1, Exon2, and Exon3). For monitoring of GATA1 mutant clone qPCR have been used with patient specific TaqMan probes and primers. Mosaic was detected by cytogenetics or FISH in bone marrow, blood and/or fibroblasts.

All newborns with TL achieved complete remission (CR). Due to clinical symptoms caused by the leukemic blasts, in 3 children low-dose cytarabine was applied. One patient died due to cardiovascular failure. In all patients GATA 1 mutation was confirmed. Minimal residual disease by qPCR (mutation-specific probes) or immunophenotyping (IP) revealed negativity in 3 out of 3 children monitored (follow-up 2 to 10.1 yrs).

Two children with (unknown) trisomy 21 mosaic were diagnosed as acute megakaryoblastic leukemia (AMKL) and treated according the high risk arm of the AML-BFM 2004 including allogeneic stem cell transplantation (one child), GATA1 mutation was identified retrospectively. Both children are alive in CR. Six children with ML-DS were initially treated according the AML-BFM protocol. After ML-DS was confirmed, therapy was continued with the intensity reduced schedule according to the ML-DS 2006 protocol. All children are still in CR (follow-up 1.5 to 6.7 years, median 2.4 yrs). This was confirmed by MRD-monitoring, which achieved negativity after two treatment elements (qPCR <10⁻⁴ n=3; IP <10⁻³ n=6). In one child a distinct refractory myeloid leukemia population (GATA1mut negative/trisomy 21 negative) arose after the 1st induction. Due to treatment refractory, allogenic stem cell transplantation was applied.

GATA1 mutated leukemia has to be excluded in all young children with AMKL (<5years old) to prevent overtreatment. Treatment with reduced intensity protocol like ML-DS 2006 seems to be effective and sufficient in children with trisomy 21 mosaic and GATA1 mutated ML-DS.

No relevant conflicts of interest to declare.