Highlights Improvements in childhood cancer survival led to increasing numbers of survivors  Childhood cancer is treated within multiinstitutional clinical trials  Chemotherapy is the main element of therapy but irradiation is still needed in some  Survivors are at longstanding risk of severe somatic late effects  Survivors may face various social and socioeconomic difficulties in adulthood

We reviewed the clinical features, treatment, and outcome of 100 children with myelodysplastic syndrome (MDS), juvenile myelomonocytic leukemia (JMML), and acute myeloid leukemia (AML) associated with complete monosomy 7 (−7) or deletion of the long arm of chromosome 7 (7q−). Patients with therapyinduced disease were excluded. The morphologic diagnoses according to modified FAB criteria were: MDS in 72 (refractory anemia (RA) in 11, RA with excess of blasts (RAEB) in eight, RAEB in transformation (RAEB-T) in 10, JMML in 43), and AML in 28. The median age at presentation was 2.8 years (range 2 months to 15 years), being lowest in JMML (1.1 year). Loss of chromosome 7 as the sole cytogenetic abnormality was observed in 75% of those with MDS compared with 32% of those with AML. Predisposing conditions (including familial MDS/AML) were found in 20%. Three-year survival was 82% in RA, 63% in RAEB, 45% in JMML, 34% in AML, and 8% in RAEB-T. Children with −7 alone had a superior survival than those with other cytogenetic abnormalities: this was solely due to a better survival in MDS (3-year survival 56 vs 24%). The reverse was found in AML (3-year survival 13% in −7 alone vs 44% in other cytogenetic groups). Stable disease for several years was documented in more than half the patients with RA or RAEB. Patients with RA, RAEB or JMML treated with bone marrow transplantation (BMT) without prior chemotherapy had a 3-year survival of 73%. The morphologic diagnosis was the strongest prognostic factor. Only patients with a diagnosis of JMML fitted what has previously been referred to as the monosomy 7 syndrome. Our data give no support to the concept of monosomy 7 as a distinct syndrome.

Sterile alpha motif domain protein 9 (SAMD9) and its paralogue SAMD9-like (SAMD9L) are cytoplasmic proteins encoded by two juxtaposed single-exon genes on chromosome 7q21. They share a 60% amino acid sequence identity and likely originated from a duplication of a common ancestral gene 1 . Their function remains enigmatic; they have been linked to tumor suppression 2 , inflammation 3 , stress response 4 , development 4 , endosomal fusion 5,6 and protein translation 7,8 . Both proteins were also shown to function as restriction factors forming a cross-species barrier for poxvirus infection [9][10][11][12] . Structural analysis of these large proteins has been limited to homology modeling, which predicted identical domains in either protein (SAM, ALBA2, SIR2, P-loop/ NTPase and OB-fold) 13 . Moreover, these genes exhibit tight regulation during embryonic development and transition to ubiquitous expression levels in adult tissues 14,15 .Notably, Samd9l-haploinsufficient mice develop myeloid neoplasia mimicking human MDS with monosomy 7 5 . Several groups reported germline SAMD9 or SAMD9L mutations (SAMD9/9L mut ) underlying new human syndromes with a propensity for cytopenia, bone marrow failure (BMF) and MDS with non-random monosomy 7 or deletion of 7q 6,16-28 . SAMD9 mutations (SAMD9 mut ) were initially linked to a fatal, early-onset MIRAGE syndrome (myelodysplasia, infections, restriction of growth, adrenal hypoplasia, genital phenotypes and enteropathy) 6,29 . In contrast, SAMD9L mutations (SAMD9L mut ) were originally described in families with a progressive neurological phenotype, multi-lineage cytopenia and bone marrow hypoplasia (ataxia-pancytopenia syndrome) 16,17 . Recent reports broadened this spectrum and found missense SAMD9/9L mut in non-syndromic familial MDS [30][31][32][33] , truncating SAMD9L mut in children with autoinflammatory panniculitis resembling CANDLE

Juvenile myelomonocytic leukemia (JMML) is a malignant hematopoietic disorder of early childhood with excessive proliferation of the myeloid and monocytic lineage. Deregulation of the RAS signal transduction pathway is thought to play a key role in its pathogenesis. We examined peripheral blood or bone marrow cells of 36 children with JMML for activating point mutations in codons 12, 13 and 61 of the NRAS and KRAS proto-oncogenes by allele-specific restriction assay, singlestrand conformation polymorphism and/or direct sequencing. Codons 12, 13 and 61 of HRAS were examined in 26 of these patients. We detected RAS mutations in six cases (17%) located at N12 (n = 2), N13 (n = 3) and K13 (n = 1). In addition, we performed clonality studies on different cell lineages in four of these patients applying the RAS mutation, the karyotype and X-chromosome inactivation patterns as clonal markers. Erythroid cells carried mutant RAS, indicating clonal origin. In EBV B cell lines, one of three patients studied harbored a RAS mutation, while the other two patients had polyclonal B cells with wildtype RAS. T lymphocytes were examined in one patient; they were polyclonal and had wild-type RAS. It is likely that JMML is a heterogeneous disease with respect to clonal involvement of different lineages.

Early response after induction is a prognostic factor for disease outcome in childhood acute myeloid leukaemia (AML). Residual disease (RD) detection by multiparameter flow cytometry (MFC) was performed at day 15 and before consolidation therapy in 101 patients enrolled in the Nordic Society of Paediatric Haemato-Oncology AML 2004 study. A multicentre laboratory approach to RD analysis was used. Event-free survival (EFS) and overall survival (OS) was significantly different in patients with and without RD at both time points, using a 0·1% RD cut-off level. RD-negative and -positive patients after first induction showed a 5-year EFS of 65 ± 7% and 22 ± 7%, respectively (P < 0·001) and an OS of 77 ± 6% (P = 0·025) and 51 ± 8%. RD-negative and -positive patients at start of consolidation therapy had a 5-year EFS of 57 ± 7% and 11 ± 7%, respectively (P < 0·001) and an OS of 78 ± 6% and 28 ± 11%) (P < 0·001). In multivariate analysis only RD was significantly correlated with survival. RD before consolidation therapy was the strongest independent prognostic factor for EFS [hazard ratio (HR):5·0; 95% confidence interval (CI):1·9-13·3] and OS (HR:7·0; 95%CI:2·0-24·5). In conclusion, RD before consolidation therapy identifies patients at high risk of relapse in need of intensified treatment. In addition, RD detection can be performed in a multicentre setting and can be implemented in future trials.