Whole Exome Analysis Reveals Spectrum of Gene Mutations in Juvenile Myelomonocytic Leukemia

Juvenile myelomonocytic leukemia (JMML) is a rare pediatric myeloid neoplasm clinically characterized by excessive proliferation of myelomonocytic cells and hypersensitivity to granulocyte–macrophage colony-stimulating factor (GM-CSF). A cardinal genetic feature of JMML is frequent somatic and/or germline mutations of RAS pathway genes involved in GM-CSF signal transduction, such as NRAS, KRAS, PTPN11, NF1, and c-CBL, which are found in >70% affected children in a mutually exclusive manner. To define the molecular pathogenesis of JMML, we identified the full spectrum of gene mutations in 13 cases of JMML using whole exome sequencing of paired tumor-normal DNA. We also performed target-deep sequencing of relevant mutational targets in 92 cases of JMML.

We evaluated 92 children (61 boys and 31 girls) with JMML, including 7 with Noonan syndrome-associated myeloproliferative disorder, who were diagnosed at institutions throughout Japan. The median age at diagnosis was 19 months (range, 1–160 months). Karyotypic abnormalities were detected in 15 cases, including 8 with monosomy 7. Fifty-six of the 92 (61%) cases received allogeneic hematopoietic stem cell transplantation.

Exome capture from paired tumor-normal (CD3-positive T cell) DNA obtained from 13 cases of JMML was performed using SureSelect® Human All Exon V3 (Agilent Technologies, Santa Clara, CA, USA) covering 50 Mb of the coding exons, followed by massive parallel sequencing using HiSeq 2000 (Illumina, San Diego, CA, USA) according to the manufacturers' protocol. Candidate somatic mutations were detected through our pipeline for whole exome sequencing (genomon: http://genomon.hgc.jp/exome/index.html). All candidate germline and somatic nucleotide changes were validated by Sanger/deep sequencing. A total of 92 JMML tumor specimens were screened for mutations in RAS pathway genes (PTPN11, NRAS, KRAS, c-CBL, and NF1) and 3 newly identified genes using deep sequencing.

For the current exome study, 10 missense and 1 nonsense single nucleotide variations were confirmed as nonsilent somatic mutations. The average number of mutations per sample (0.79; range, 0–4) was surprisingly low compared with those reported in other human cancers. Among the 11 somatic mutations, 6 involved the known RAS pathway genes (1 NF1, 1 NRAS, 2 KRAS, and 2 PTNP11 mutations) and included 5 mutations/deletions of either NF1 (n = 2), c-CBL (n = 1), or PTPN11 (n = 2) as detected in the germline samples. Nonoverlapping RAS pathway mutations were confirmed in 11 of the 13 discovered cases of JMML (85%). Five of the 11 somatic mutations were observed in 3 non-RAS pathway genes that have never been reported in JMML cases.

Deep sequencing revealed RAS pathway mutations in 80 of 92 cases (87%) in a mutually exclusive manner; PTPN11 mutations were predominant (39/92 or 42%), followed by N/KRAS (24/92 or 26%), c-CBL (11/92 or 12%), and NF1 (6/92 or 6.5%) mutations. In agreement with previous reports, the majority of c-CBL (7/11) and NF1 (5/6) mutations were bi-allelic in the affected cases, showing compound heterozygous mutations or uniparental disomy of the mutant alleles, whereas most of the PTPN11 and N/KRAS mutations were heterozygous. In contrast, the remaining 12 (13%) cases were negative for known RAS pathway mutations. In addition, the 3 newly identified genes were recurrently in 18 cases (20%). Many of these mutations had lower allele frequencies compared to the accompanying RAS pathway mutations, indicating that they were responsible for disease progression rather than the establishment of JMML. The probability of 5-year transplantation-free survival (95% confidence interval) for the latter patients was significantly inferior to that of other cases (0% vs. 19% (8–34%), p < 0.001).

Whole exome sequencing revealed the spectrum of gene mutations in cases of JMML. Together with a very high frequency of RAS pathway mutations, the paucity of non-RAS pathway mutations is a prominent feature of JMML. Somatic mutations of 3 newly identified genes were common among recurrent secondary events presumed to be involved in tumor progression and associated with poor clinical outcomes. Our findings provide an important clue that aids in understanding the pathogenesis of JMML and will help in the development of novel diagnostics and therapeutics for this type of leukemia.

Maciejewski:NIH: Research Funding; Aplastic Anemia & MDS International Foundation: Research Funding.