LAUSR

Exposure to ionizing radiation is a well-established risk of cancer development. Atomic bomb (A-bomb) survivors were exposed to A-bomb radiation, then developed acute myeloid leukemia (AML), acute lymphoblastic leukemia, and chronic myeloid leukemia with the highest incidence around five to ten years after the bombing. On the other hand, the risk of myelodysplastic syndromes (MDS) were significantly high among the survivors even after 40 to 60 years from the bombing (Iwanaga et al. J Clin Oncol 2011), indicating a long-lasting influence of A-bomb radiation on human hematopoiesis. In general, genetic mutations are found in almost all MDS, however, the landscape of genetic alterations in MDS among A-bomb survivors remains to be elucidated. To study the role of A-bomb radiation on the pathogenesis of MDS and related conditions, we analyzed 32 patients with MDS and 3 cases diagnosed as idiopathic cytopenia of undetermined significance (ICUS) among A-bomb survivors using next generation sequencing technologies and SNP array karyotyping. Patients were categorized into two groups by the exposure distance; proximally exposed group (PE-group, 18 patients, < 2.7 km from the hypocenter) and distally exposed group (DE-group, 17 patients, directly exposed in 2.7-10 km, or those who entered the bombing area [inside of 2 km from the hypocenter] within two weeks after the bombing [no acute radiation syndromes]). Median exposure distance was 1.1 and 3.4 km in PE- and DE-group, respectively (P < .001). There was no significant difference in sex, subtype of MDS, and ages at bombing and the diagnosis, between the two groups, whereas frequencies of abnormal karyotype (78% vs. 59%, P = .29) and complex karyotype (33% vs. 12%, P = .23) were higher in PE-group without statistical significance. Whole genome sequencing of three patients in PE-group revealed that most frequent nucleotide substitution was cytosine-to-thymine (C to T), which is a hallmark of age-related mutational changes. Using whole exome sequencing and 154 genes-targeted capture sequencing (T-S), we identified significant differences in mutated genes between the two groups. No patients of PE-group had mutations in TET2 although it was most frequently affected in DE-group (29% [5 out of 17 patients], P = .019), and the frequency of gene mutations in DNA methylation pathway was 5.6% (1 out of 18 patients) and 41% (7 out of 17) in PE- and DE-group, respectively (P = .018). TP53 and the genes coding RNA splicing factors (SFs) were mutated equally in both groups (TP53, 11% and 12% [P = 1]; SFs, 33% and 35% [P = 1] in PE- and DE-group, respectively). Copy number alterations were analyzed using SNP-array and T-S data. We found that loss of long arm of chromosome 11 (11q loss) was significantly accumulated in PE-group (33% [6 out of 18 patients] vs. 0%, P = .019), whereas loss of chromosome 5q and chromosome 7 were almost equally identified in both groups (chromosome 5q, 22% and 12% [P = .66]; chromosome 7, 22% and 29% [P = .71] in PE- and DE-group, respectively). On the commonly affected region of 11q in this study, there were three genes recurrently altered in MDS and AML; ATM, KMT2A, and CBL. Copy number loss was found in these three genes, and the alteration of ATM was significantly increased in PE-group (28% [5 out of 18 patients] vs. 0%, P = .046) than DE-group. The remaining allele of ATM also possessed a mutation in two out of five patients in PE-group, indicating a deleterious effect on the function of ATM. In summary, we revealed two unique genetic features in MDS among A-bomb survivors who were proximally exposed to A-bomb radiation; significantly less frequent mutations in DNA methylation pathway and significant accumulation of 11q loss, suggesting a long-lasting effect of ionizing radiation and an important role of 11q loss in the initiation of MDS. No relevant conflicts of interest to declare.