Long-term survival after nonintensive chemotherapy in some juvenile myelomonocytic leukemia patients with CBL mutations, and the possible presence of healthy persons with the mutations

To the editor:

Recently, mutations in the CBL gene have been demonstrated in patients with juvenile myelomonocytic leukemia (JMML).¹  Screening for CBL mutation using DNA from peripheral blood mononuclear cells (PBMCs) taken after informed consent of the patients' guardians revealed the presence of heterozygous CBL mutations in 4 (case nos. 1 and 4, in remission; case nos. 2 and 3, at diagnosis) of 12 JMML patients who possessed wild-type NRAS, KRAS, and PTPN11 genes and did not exhibit the clinical features of neurofibromatosis type I (Table 1). This study was approved by the Institutional Review Board of Shinshu University. To elucidate the possibility of the coexistence of cells harboring homozygous mutations, those with heterozygous mutations, and normal clones, we performed genetic analyses using DNA from granulocyte-macrophage colony constituent cells, erythroid colony constituent cells, or nails.³  The 3 JMML patients (case nos. 1-3) had heterozygous and/or homozygous CBL mutations. Taken together with the data obtained by microsatellite analysis for the detection of a loss of heterozygosity and those obtained by multiplex ligation probe amplification analysis for the determination of copy numbers, homozygous CBL mutations in 2 patients appeared to result from acquired uniparental disomy (UPD).^4,5  DNA obtained from the nails of case nos. 1 and 2 had heterozygous CBL mutations identical to blood samples, suggesting that the patients had germline heterozygous CBL mutations. These patients did not exhibit the neurologic abnormalities reported by Niemeyer et al.⁶  It is of interest that case nos. 1 and 2 have achieved hematologic improvement after nonintensive chemotherapy and approximately 15- to 28-year survival. Although Muramatsu et al⁷  reported no difference in the probability of 2-year overall survival between JMML patients with and those without CBL mutations, some JMML patients with the mutations may not always show a poor prognosis.

Because colonies of case no. 1 were not obtained at diagnosis, we cannot exclude the possibility that the patient had lost her homozygous clone when she went into remission. Among 34 JMML patients with CBL mutations reported by us and other investigators,^1,7  31 patients had homozygous mutations at onset, whereas the remaining 3 patients showed heterozygous mutations. Thus, loss of the wild-type allele may play an important role in the disease occurrence of JMML patients with CBL mutations.

Case no. 4 had mild hematologic abnormalities at diagnosis, which improved spontaneously. Sequence analyses of genomic DNA and transcribed mRNA from colony constituent cells at 16 years of age revealed a heterozygous CBL mutation (1096-1G>T) and the use of novel splice acceptor sites resulting in splice variants. The mutation was present in his lymphocytes, whereas the nails had the wild-type of the gene. None of 100 healthy controls nor the parents of case no. 4 possessed this mutation. Thus, the patient may continue to carry a somatic mutation without disease at present. Although T cells generally do not belong to the leukemic clone in most JMML patients, RAS gene mutations were detected in CD3⁺ cells of 3 JMML patients.³  It should be elucidated whether the heterozygous 1096-1G>T transversion contributes to spontaneous hematologic remission and a somatic mutation in lymphocytes. The father of case no. 1, who has been healthy and has shown normal hematologic findings, harbored the same heterozygous CBL mutation as his daughter in his nails as well as PBMCs, indicating an inherited germline mutation. Persons with heterozygous CBL mutations who appear healthy may be present.