PTPN11 Mutational Spectrum in Juvenile Myelomonocytic Leukemia and Noonan Syndrome.

Somatic, heterozygous missense mutations in the PTPN11 proto-oncogene encoding SHP-2 are identified in 35% of patients with juvenile myelomonocytic leukemia (JMML). Other non-syndromic hematologic malignancies in which somatic PTPN11 mutations have been detected are pediatric myelodysplastic syndrome, acute monocytic leukemia (FAB-M5 AML) and common or B-cell precursor acute lymphoblastic leukemia. Germline PTPN11 mutations are found in 50% of patients with Noonan syndrome (NS), an autosomal dominant disorder characterized by facial anomalies, short stature and congenital heart defects. Infants with NS are predisposed to developing JMML (NS/JMML); however, the course of NS/JMML tends to be milder and self-resolving. JMML that is not associated with NS have a poor prognosis and are currently being treated with intensive regimens such stem cell transplantation. Differentiating JMML from NS/JMML is of critical clinical relevance and also provides interesting questions about the pathogenesis of these diseases. To that end, we have compared the spectrum of mutations in patients with isolated JMML, NS/JMML and NS alone. The assembly of all known published and unpublished germline and somatic exon 3 and 13 PTPN11 mutations detected in ours and other laboratories (78 pts with PTPN11 mutation positive isolated JMML; 18 pts with PTPN11 mutation positive NS/JMML) reveal that the identity of the affected residues or the type of substitution differ between NS and JMML, even though the resulting molecular defects appear to be functionally similar. In NS defects in exons 4, 7 and 8 account for approximately one-half of cases. On the contrary, mutations affecting these exons are rarely identified in JMML. A few germline NS-causative mutations affect the same residues of SHP-2 that are also altered by somatic mutations in non-syndromic JMML. In almost all of the cases, the germline and somatic mutations affecting identical residues differ with respect to the amino acid substitution. There are 2 major hot spots: 7 out of 18 patients (39%) with NS/JMML carry the T73I substitution. In isolated JMML the E76K mutation is detected most often (18 out of 78 patients (23%)). We describe 2 novel JMML mutations (E76M, G503V) and 2 novel NS/JMML mutations (R598W, S502A). Six mutations associated with isolated NS are also observed in NS/JMML. These findings imply the presence of a germline mutation needs to be excluded in all mutation positive neonates with presumed isolated JMML. In addition, our findings raise a number of research questions: First, are somatic PTPN11 mutations alone sufficient to initate leukemia and what are the molecular factors influencing the consequences of a PTPN11 mutation in hematopoietic cells? Second, do identical mutations have different consequences on cell fate of hematopoetic cells depending on whether they occur as germline or somatic events? Do some patients with isolated NS and PTPN11 mutation develop transient myeloproliferation of hematopoetic cells which may be subtle and unrecognised?