To the Editor: SETBP1 mutations are linked to poor prognosis in adult acute myeloid leukemia (AML) and have also been found in other myeloid malignancies 1-3. However, the clinical features of pediatric AML harboring SETBP1 mutations remain poorly understood 4. We report a pediatric AML patient with concurrent somatic mutations in SETBP1 and KRAS, who received azacitidine (AZA) treatment both before and after hematopoietic stem cell transplantation (HSCT). A 3-year-old boy with a fever visited a local physician. Blood tests revealed pancytopenia, leading to a referral to our department. Physical examination showed petechiae but no extramedullary lesions or hepatosplenomegaly. Peripheral blood examination confirmed pancytopenia, with no detectable leukemic blasts. Bone marrow examination revealed 20% blasts with dysplasia in the myeloid lineage cells, resulting in a diagnosis of AML. Flow cytometry was consistent with AML, and both G-banding and a commercial fusion panel showed no abnormalities (Supplementary Table S1). The patient was enrolled in the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) AML-12 protocol (Supplementary Table S2) 5 and achieved complete remission after induction therapy. However, three months after completing chemotherapy, the patient experienced an initial bone marrow relapse. Salvage therapy with idarubicin, fludarabine, cytarabine, and granulocyte-colony stimulating factor was administered 6, but it was ineffective, with 35.2% of blasts remaining in the bone marrow. AZA 75 mg/m2 was administered for seven days to reduce the blasts 7. The leukemic blasts decreased from 35.2% to 10%, fulfilling the European LeukemiaNet criteria for partial remission 8. After a myeloablative conditioning regimen (Supplementary Table S3), peripheral blood stem cell transplantation (PBSCT) was performed using cells from the patient's HLA-haploidentical father. Beginning four months after PBSCT, monthly AZA was given for six cycles without recurrence 9. Nine months after discontinuation of AZA treatment and 19 months after PBSCT, the patient experienced a second bone marrow relapse. The patient was started on AZA; however, after five courses with little effect on disease control, a myeloablative conditioning regimen (Supplementary Table S3) was administered, followed by unrelated cord blood cell transplantation (UCBT). Two months after UCBT, monthly AZA was administered for 22 cycles, and the patient remains in remission more than three years later, with persistent neurocognitive sequelae following HHV-6 encephalitis (Figure 1). To identify somatic mutations, genomic DNA was extracted from serial bone marrow samples. Whole-exome sequencing (WES) identified SETBP1 p.D868N and KRAS p.G12D as dominant clonal mutations with variant allele frequencies (VAFs) of 0.38 and 0.44, respectively, and revealed no pathogenic copy-number alterations (Supplementary Table S4 and Supplementary Figure S1). Droplet digital PCR (ddPCR) showed closely matched VAFs for both mutations throughout the disease course, supporting a common ancestral origin (Figure 1). In a comprehensive genetic study of pediatric AML, only one case with SETBP1 mutation and RAS pathway abnormalities has been reported 4. That patient responded well to induction chemotherapy, although the subsequent clinical course remained unclear. Our patient similarly achieved remission following induction chemotherapy but experienced two relapses, ultimately achieving long-term survival after two HSCT and maintenance therapy with AZA. Genetic analysis identified SETBP1 and KRAS mutations in the bone marrow samples at both onset and relapse. The SETBP1 D868N variant is a known hotspot within the SKI-homologous region, and its leukemogenic gain-of-function has been demonstrated by functional analyses 1. Retrospective ddPCR analysis showed that the VAFs of both mutations closely paralleled each other and mirrored WT1 mRNA levels, a well-established molecular marker in AML 10. These parallel trends suggest that the two mutations arose concurrently within a single leukemic clone and may have contributed to chemotherapy resistance. The morphologic blast percentage underestimated the mutation burden, suggesting that the mutations extended to dysplastic myeloid precursors that were not morphologically classified as blasts. In adults, SETBP1-mutated AML is associated with poor prognosis and often accompanied by monosomy 7/deletion 7q or complex karyotypes 1, 2. However, its clinical features in children remain unclear, and our case showed no chromosomal abnormalities. In contrast to adult secondary AML, where SETBP1 mutations often occur first and lead to chemo-resistant myelodysplastic syndrome-related clones 11, our case showed similar VAFs between SETBP1 and KRAS mutations throughout the disease course. This clonal architecture is consistent with previous experimental studies demonstrating that SETBP1 mutations cooperate with RAS pathway alterations to promote leukemic transformation through enhanced proliferative signaling 12. In our case, AZA showed limited activity, consistent with reports of poor responses in adults with SETBP1-mutated AML 13. Because the patient underwent HSCT in non-remission, the durability of benefit remains uncertain, although AZA may have served as bridging or maintenance therapy 14, 15. This case report has limitations. Because whole-transcriptome RNA sequencing and whole-genome sequencing were not performed, additional fusion genes contributing to the disease biology cannot be excluded. In addition, genome-wide methylation profiling was not conducted, and the epigenetic changes associated with AZA exposure before and after treatment could not be evaluated. In summary, we report a chemo-refractory pediatric AML with co-mutated SETBP1 and KRAS that ultimately achieved long-term remission through HSCT, with AZA potentially contributing as bridging and maintenance therapy. Further case accumulation is needed to validate its therapeutic role in SETBP1-mutated pediatric AML. RO, KK, EI, and KT designed the study. RO, TTa, KK, RK, and TTo performed the experiments. RO, KK, KF, and TTo analyzed and interpreted the data. RO and KK prepared the manuscript. RO, KK, TTa, AK, TS, EI, and KT evaluated the patients and collected and interpreted the clinical data. All authors have read and approved the final version of the manuscript. The authors thank H. Kudo and M. Hashimoto for their technical assistance in this study, which was partially supported by the Japan Society for the Promotion of Science (JSPS) through a Grant-in-Aid for Scientific Research (KAKENHI; grant number 21K07813). Written informed consent for publication, including clinical details and genomic data, was obtained from the patient's guardian. All authors declare that they have no competing conflicts of interest. Sequencing data supporting this report are not publicly available because of patient privacy restrictions but are available from the corresponding author on reasonable request. Supplementary Table S1: Results of FISH, immunophenotyping, cytogenetics, and fusion gene analysis at initial diagnosis and relapse. Supplementary Table S2: Outline of treatment schedule—JPLSG AML-12 protocol. Supplementary Table S3: Outline of conditioning regimen. Supplementary Table S4: Nonsynonymous single-nucleotide variants identified by whole-exome sequencing of the relapse sample. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Okuse et al. (Sun,) studied this question.