-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

385 Combined Landscape of Gene Mutations and Copy Number Alterations in Clonal Hematopoiesis: Analysis in 10,612 Japanese Individuals

Program: Oral and Poster Abstracts
Type: Oral
Session: 503. Clonal Hematopoiesis: Aging and Inflammation
Hematology Disease Topics & Pathways:
Diseases, MDS, Biological Processes, genomics, Myeloid Malignancies, hematopoiesis
Sunday, December 6, 2020: 12:45 PM

Ryunosuke Saiki, MD1*, Tetsuichi Yoshizato, MD, PhD1, Yasuhito Nannya, MD, PhD1, Yukihide Momozawa, DVM, Ph.D2*, Chikashi Terao, MD, PhD3*, Yuichi Shiraishi, PhD4*, Kenichi Chiba5*, Hiroko Tanaka6*, Atsushi Niida, PhD7*, Seiya Imoto, PhD8*, Koichi Matsuda, MD, PhD9*, Takayuki Morisaki, MD, PhD10*, Yoshinori Murakami, MD, PhD10*, Yoichiro Kamatani, MD, Ph.D3,9*, Michiaki Kubo, MD, Ph.D11*, Satoru Miyano, PhD6*, Hideki Makishima, MD, PhD1 and Seishi Ogawa, MD, PhD1,12,13

1Department of Pathology and Tumor Biology, Graduate School of Medicine, Kyoto University, Kyoto, Japan
2Laboratory for Genotyping Development, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan
3Laboratory for Statistical and Translational Genetics, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan
4Center for Cancer Genomics and Advanced Therapeutics, National Cancer Center Research Institute, Tokyo, Japan
5Division of Cellular Signaling, National Cancer Center Research Institute, Tokyo, Japan
6Department of Integrated Data Science, M&D Data Science Center, Tokyo Medical and Dental University, Tokyo, Japan
7Laboratory of Molecular Medicine, Human Genome Center, The Institute of Medical Science, The University of Tokyo, Tokyo, Japan
8Health Intelligence Center, Institute of Medical Science, University of Tokyo, Tokyo, Japan
9Department of Computational Biology and Medical Sciences, Graduate school of Frontier Sciences, The University of Tokyo, Tokyo, Japan
10Division of Molecular Pathology, Institute of Medical Science, The University of Tokyo, Tokyo, Japan
11RIKEN Center for Integrative Medical Sciences, Yokohama, Japan
12Department of Medicine, Center for Haematology and Regenerative Medicine, Karolinska Institutet, Stockholm, Sweden
13Institute for the Advanced Study of Human Biology (WPI-ASHBi), Kyoto University, Kyoto, Japan


Clonal hematopoiesis (CH) in apparently healthy individuals has been investigated on the basis of either recurrent driver mutations or copy number alterations (CNAs). However, previous studies have not evaluated both alterations together at a comparable sensitivity. Thus, the combined landscape of CH with regard to both lesions, their co-occurring patterns, and their effects on clinical outcomes have not been fully elucidated.

Materials & Methods

To address this issue, we enrolled a total of 10,612 subjects from the BioBank Japan (BBJ), for whom blood-derived DNA and SNP-array data as well as a standard health record were available. All the subjects were over 60 years old and had no history of malignancy at the time of DNA sampling. Their samples were analyzed using deep sequencing and SNPs-array to detect CH-related mutations in 23 driver genes and CNAs, respectively.


In total, CH was detected in approximately 40% of the entire cohort of apparently healthy individuals over 60 years of age, by the presence of mutations and/or CNAs. We called 3,660 mutations in 2,820 (26.6%) subjects. In accordance with previous reports, DNMT3A (13.5%), TET2 (9.3%), ASXL1 (2.1%), and PPM1D (1.4%) were most frequently mutated. SNP-array based copy-number analysis detected 2,459 regions/segments showing CNAs in 2,049 (19.3%) of the cases. 14qUPD (1.9%), +21q (0.9%), del(20q) (0.8%), and +15q (0.8%) were among the most frequent CNAs. Among CH-positive subjects, ~15% harbored both mutations and CNAs, which was significantly more frequent than expected only by chance (odds ratio [OR]=1.3; P<0.001), suggesting their functional interactions for positive selection in CH.

Focusing on specific alterations, we found a number of significantly co-occurring combinations of mutations and CNAs, including DNMT3A mutation and 2pUPD/del(2p) (n=14, OR=4.99, q=0.02), TET2 mutation and 4qUPD/del(4q) (n=16, OR=9.10, q<0.001), JAK2 mutation and 9pUPD (n=13, OR: 231, q<0.001), and TP53 mutation and 17pUPD/del(17p) (n=8, OR=88.0, q<0.001) (Figure). Also commonly identified in myeloid neoplasms (MNs), bi-allelic involvement of driver genes in these combinations in CH might represent positive selection in early myeloid leukemogenesis. By contrast, other combinations commonly observed in myeloid neoplasms such as CBL/11qUPD, NRAS/1pUPD, or EZH2/7qLOH were rarely seen in CH, probably because these combinations are more related to progression to MNs rather than selected for CH.

Next, we evaluated correlations between CH and clinical parameters. We evaluated risk factors of CH in multivariate analysis; besides the strongest effects of age, we found that male sex independently predicted +15, del(20q), +21, and ASXL1 and PPM1D mutations, and tobacco smoking predicted ASXL1 mutations. We also observed significant associations between CH-related lesions and blood counts: JAK2 mutations with higher platelet counts (P < 0.001), PPM1D mutations and del(20q) with lower hemoglobin levels (P < 0.01), and CBL mutations with higher hemoglobin levels (P < 0.01), suggesting a unique role of each lesion in hematopoiesis.

Finally, we evaluated the effect of CH on prognosis. The presence of high-VAF mutations (VAF≥10%) (HR=2.73, 95%CI:1.14-6.55) and CNAs (HR=2.02, 95%CI:1.11-3.69) was significantly associated with higher mortality from hematological malignancies. Given the established association between CH and cardiovascular diseases, we also evaluated the effects on all-cause mortality and cardiovascular mortality. CH-related mutations (VAF ≥ 5%) were significantly associated with higher all-cause (HR=1.15, 95%CI:1.06-1.26) and cardiovascular mortality (HR=1.33, 95%CI:1.05-1.69), although CNAs conferred almost no effects.


Combined analysis of gene mutations and CNAs in CH disclosed associations between mutations and CNAs commonly seen in MNs such as those involving DNMT3A, TET2, JAK2, and TP53, suggesting the role of biallelic alteration in these genes even in early myeloid leukemogenesis. There were combined effects of CH-related mutations and CNAs on both hematological malignancies and cardiovascular diseases.

Disclosures: Ogawa: KAN Research Institute, Inc.: Membership on an entity's Board of Directors or advisory committees, Research Funding; Chordia Therapeutics, Inc.: Membership on an entity's Board of Directors or advisory committees, Research Funding; Asahi Genomics Co., Ltd.: Current equity holder in private company; Sumitomo Dainippon Pharma Co., Ltd.: Research Funding; Otsuka Pharmaceutical Co., Ltd.: Research Funding; Eisai Co., Ltd.: Research Funding.

*signifies non-member of ASH