Therapy-related clonal hematopoiesis in pediatric cancer survivors

Background: Clonal hematopoiesis of indeterminate potential (CHIP) refers to clonal expansion of hematopoietic stem cells due to somatic mutations and is associated with heightened risk of myeloid neoplasms and cardiovascular diseases. Genotoxic therapies such as chemotherapy and radiation therapy are well-recognized contributors to CHIP in adult cancer population. However, CHIP prevalence, mutational patterns, and treatment associations in pediatric populations remain poorly characterized. Methods: We conducted a retrospective analysis of pediatric patients (age ≤ 18) diagnosed with cancer with blood samples collected after cancer therapy. CHIP was defined as the presence of a pathogenic somatic mutation in an accepted driver gene with variant allele frequency (VAF) ≥2%. CHIP prevalence was estimated overall and stratified by histology and therapeutic exposure. Logistic regression with Firth correction and exposure adjustment was used for multivariable analyses. TP53 variants with known Li-Fraumeni syndrome association and VAF >40% were considered germline mutations and excluded from analyses. Results: CHIP mutations were identified in 23 of 1,052 (2.2%) patients in this cohort. Mutations occurred most frequently in the TET2 (26%), DNMT3A (22%), and KRAS (13%) genes. CHIP prevalence was numerically higher among patients who received chemotherapy (3.3% vs. 1.4%, p=0.055). Multivariable analysis showed a modest female predominance (OR 0.45, p=0.04) and a trend toward higher CHIP risk with chemotherapy (OR 1.9, p=0.16). Among drug classes, antimetabolite therapy showed the highest CHIP prevalence (OR 3.46, p = 0.048), driven by methotrexate exposure (OR 21.5, p = 0.001). No correlations were found with age, histology, or radiation. Conclusion: We identified a low overall prevalence of CHIP in pediatric cancer patients and an association with antimetabolite therapy. Further studies are needed to define the long-term clinical implications of clonal hematopoiesis in pediatric cancer survivors. Baseline characteristics by CHIP status Characteristic Overall (n = 1052) No CHIP mutation (n = 1029) CHIP mutation (n = 23) p-value Age (mean + SD) 11.5 ± 7.4 11.6 ± 7.4 9.3 ± 5.7 0.083 Gender 0.091 Male 516 (49.0%) 509 (49.5%) 7 (30.4%) Female 536 (51.0%) 520 (50.5%) 16 (69.6%) Cancer type 0.5 Lymphoid leukemias 225 (21.4%) 216 (21.0%) 9 (39.1%) Sarcoma 179 (17.0%) 177 (17.2%) 2 (8.7%) Glioma 136 (12.9%) 135 (13.1%) 1 (4.3%) Neuroblastoma 72 (6.8%) 71 (6.9%) 1 (4.3%) Renal/Hepatic tumors 62 (5.9%) 61 (5.9%) 1 (4.3%) Hodgkin lymphoma 49 (4.7%) 48 (4.7%) 1 (4.3%) Benign tumor 46 (4.4%) 45 (4.4%) 1 (4.3%) Medulloblastoma / Embryonal tumors 44 (4.2%) 43 (4.2%) 1 (4.3%) Non-Hodgkin lymphoma 43 (4.1%) 42 (4.1%) 1 (4.3%) Other CNS 36 (3.4%) 36 (3.5%) 0 (0.0%) Germ cell tumors 35 (3.3%) 33 (3.2%) 2 (8.7%) Carcinoma NOS 27 (2.6%) 26 (2.5%) 1 (4.3%) Endocrine tumors 25 (2.4%) 25 (2.4%) 0 (0.0%) Myeloid leukemias 22 (2.1%) 22 (2.1%) 0 (0.0%) Ependymoma 20 (1.9%) 20 (1.9%) 0 (0.0%) Retinoblastoma 17 (1.6%) 16 (1.6%) 1 (4.3%) Other tumor 14 (1.3%) 13 (1.3%) 1 (4.3%) Chemotherapy 0.055 Yes 430 (40.9%) 416 (40.4%) 14 (60.9%) No 622 (59.1%) 613 (59.6%) 9 (39.1%) Chemotherapy type 0.3 Antimitotic Agent 153 (14.5%) 149 (14.5%) 4 (17.4%) Antimetabolite 110 (10.5%) 103 (10.0%) 7 (30.4%) Methotrexate 13 (1.2%) 10 (1.0%) 3 (13.0%) Alkylating Agent 84 (8.0%) 81 (7.9%) 3 (13.0%) Antitumor Antibiotic 55 (5.2%) 55 (5.3%) 0 (0.0%) Plant Alkaloid 26 (2.5%) 26 (2.5%) 0 (0.0%) Miscellaneous Agent 1 (0.1%) 1 (0.1%) 0 (0.0%) Topoisomerase I Inhibitor 1 (0.1%) 1 (0.1%) 0 (0.0%) Immunotherapy >0.9 Yes 6 (0.6%) 6 (0.6%) 0 (0.0%) No 1046 (99.4%) 1023 (97.2%) 23 (100.0%) External beam RT >0.9 Yes 119 (11.3%) 117 (11.4%) 2 (8.7%) No 933 (88.7%) 912 (88.6%) 21 (91.3%) Radiopharmaceuticals >0.9 Yes 36 (3.4%) 36 (3.5%) 0 (0.0%) No 1016 (96.6%) 993 (96.5%) 23 (100.0%)