Patients with JMML have a variable clinical course. Many have relentless myeloproliferation, irrespective of the initial therapy, and die rapidly if not successfully given a bone marrow transplant (BMT). Others partially respond to therapy with reduction in their leukocytosis and organomegaly, but with continued cytopenias. Finally, some children have a much more indolent clinical course, requiring little or even no therapy. A number of cytogenetic abnormalities are found in JMML, particularly monosomy 7. Molecular abnormalities have also been described including mutations in PTPN11, NRAS and KRAS, but it is not clear whether the karyotype and genotype are related to the phenotype. To address this issue, we studied 33 cases diagnosed between 1992 to 2007. All fulfilled the minimal diagnostic criteria for JMML. Three had Noonan syndrome (NS), and 1 a constitutional trisomy 8 mosaicism (CT8M). The remaining 29 had no known genetic syndrome. The median age at presentation was 15 months (0–96 months) and 22 (69%) were under 2 years of age. Patients were retrospectively allocated to 1 of 3 clinical groups. Group I (18 patients, 55%) had aggressive disease with minimal or no response to initial therapy. Group II (8 patients, 24%) had a partial response to therapy but persistent severe cytopenias or persistence of a monosomy 7 clone. All patients in Group I and II were scheduled to receive a BMT; this was carried out in 16 of 18 Group I patients and 7 of 8 Group II patients. A further 7 cases (Group III, 21%) had indolent disease with a median follow-up of 5 years (1–12 years). It is noteworthy that 2 of the 3 patients with NS were in this category. There was no difference at presentation in the clinical or diagnostic parameters between the 3 clinical groups. Cytogenetic analysis was available in 32 patients with 20 (63%) having a normal karyotype, 9 (28%) monosomy 7, 2 complex abnormalities and 1 CT8M. The presence of monosomy 7 was restricted to Groups I and II, but as this cytogenetic finding was used as an indication for BMT, limited conclusions can be drawn about the natural history of this subset. All Group III patients had a normal karyotype. Mutation screening using WAVE DHPLC analysis revealed 12 patients (36%), including the 3 NS patients, with PTPN11 mutations, 6 (18%) with NRAS and 8 (24%) with KRAS mutations. The total incidence of mutations was similar in the 3 clinical groups being 78%, 88% and 71% in Groups I, II and III respectively. KRAS mutations were not present in Group III patients, but this was not significant. Six of the 33 patients in this series, all in Group I, have transformed to an acute leukemia. Two had cytogenetic abnormalities although none had monosomy 7, 1 had a PTPN11 mutation, 2 an NRAS and 2 a KRAS mutation. Four out of 14 patients with a RAS mutation underwent transformation compared to 1 out of 12 of those with a PTPN11 mutation (p = 0.18). Of note, 1 patient with a KRAS G12S mutation was in Group II, which differs from the indolent course reported for this mutation by others. These results indicate that neither cytogenetic abnormalities nor specific mutations in PTPN11, NRAS or KRAS predict the clinical course of JMML. This study also raises the possibility that the clinical course is determined, in part, by cooperating mutations or polymorphisms that have not yet been identified.

Author notes

Disclosure:Research Funding: Leukemia Research Fund, UK.

Sign in via your Institution