Chromosomal alterations among age-related hematopoietic clones in Japan

Abstract

The extent to which the biology of oncogenesis and aging are shaped by factors that distinguish human populations is unknown. Hematopoietic clones with acquired mutations become common with advancing age and can lead to blood cancers^1–10. Here we describe shared and population-specific patterns of genomic mutation and clonal selection revealed by 33,250 autosomal mosaic chromosomal alterations (mCAs) we ascertained from 179,417 Japanese participants in the BioBank Japan cohort and compared to analogous analyses of UK Biobank. In this long-lived Japanese population, mCAs were detected in more than 35.0% (s.e. 1.4%) of individuals older than 90 years, suggesting that such clones trend toward inevitability with advancing age. Japanese and Europeans exhibited key differences in the genomic locations of mutations in their respective hematopoietic clones; these differences anticipated the populations’ relative rates of CLL (which is more common among Europeans) and T-cell leukemia (more common among Japanese). Three different mutational precursors of CLL (including trisomy 12, 13q loss, and 13q CN-LOH) were 2–6x less common among the Japanese, suggesting that the Japanese and European populations differ in selective pressures on clones long before the development of clinically apparent CLL. Japanese and UK populations also exhibited very different rates of clones arising in B- and T-cell lineages, anticipating the relative rates of B- and T-cell cancers in these populations. We identified six new loci at which inherited variants predispose to mCAs that duplicate or remove inherited risk alleles, including large-effect rare variants at NBN, MRE11, and CTU2 (OR=28–91). Our results suggest that selective pressures on clones are modulated by factors specific to human populations, and motivate further genomic characterization of clonal selection and cancer in populations around the world.

Clonal hematopoiesis involving cells with genomic alterations commonly occurs in older individuals and confers risk of hematological malignancies and overall mortality^1–10. Clones harbor diverse mutations– ranging from point mutations to whole-chromosome gains and losses – on every chromosome.

Though populations can differ greatly in their rates of various cancers, the genomic landscape of mosaicism in the absence of known cancer is currently unexplored outside of European-ancestry cohorts^10–13.

Mosaic chromosomal alterations in Japan

We searched for mosaic chromosomal alterations (mCAs) in blood DNA microarray data from 179,417 participants in the BioBank Japan (BBJ) cohort, which was ascertained for 47 diseases¹⁴ (including 13 cancers found in 16.7% of participants; Methods). We ascertained mCAs by analyzing allele-specific hybridization intensities for 515,355 genotyped autosomal SNPs (Supplementary Table 1), analyzing these data with a recently developed approach that detects imbalanced abundance of an individual’s two inherited haplotypes by utilizing the long-range haplotype phase information that can be inferred from large population samples¹⁰ (Methods and Supplementary Note 1).

This analysis detected 33,250 autosomal mCAs (at FDR = 0.05) in 27,910 unique individuals (Fig. 1 and Supplementary Note 2). (The high rate of events, relative to a contemporaneous analysis of 482,789 UK Biobank (UKB) participants, reflects that BBJ participants were older (mean age at enrollment 62.8 years in BBJ vs. 57 years in UKB; s.d. 14.5 years, range 0–113 years) and a larger fraction were male (54.1% in BBJ vs. 45.8% in UKB)¹⁴.) Of these mutations, 5,233 were confidently classified as mosaic deletions, 10,431 as copy-neutral loss of heterozygosity (CN-LOH) events, and 4,044 as duplications (Fig. 2a and Supplementary Table 2); the remaining 13,452 events were present at cell fractions too low or spanned too few genotyping probes to confidently determine copy number (Supplementary Note 1). Some 4,156 individuals harbored two or more non-overlapping mCAs (Supplementary Table 3); analysis of mosaic cell fractions suggested that these events were usually present in distinct clones (Supplementary Note 3). The mCA detection rate was broadly consistent across genotyping arrays (Supplementary Table 4) and across cases of the 47 diseases systematically surveyed by BBJ (Supplementary Table 5); mCAs were strongly enriched (OR=1.93 (1.66–2.25), p=2.1×10⁻¹⁷) among individuals with hematological cancers at registry (i.e., at the time of DNA sampling) as expected from earlier work (Supplementary Note 4), but not among individuals with other illnesses.

Figure 1. Genomic locations of 33,250 autosomal mCAs detected in 27,910 unique BioBank Japan participants.

Loss, CN-LOH, and gain events are plotted as blue, orange, and red horizontal lines, respectively. Events with undetermined copy number are plotted in grey. Commonly deleted regions are labeled in blue; loci associated with CN-LOH mutations in cis are labeled in orange.

Figure 2. Classification of mCAs, frequency as a function of age, and comparison of genomic distributions between BBJ and UKB.

a. Classification of mCAs as loss, CN-LOH, or gain events using log R ratio (LRR, measuring total DNA abundance) and B allele frequency deviation from 0.5 (|ΔBAF|, measuring allelic imbalance; Methods). Unclassified events are indicated in grey.

b. Frequency of detectable mosaicism stratified by age and sex. Frequencies (means) and error bars for 95% CIs are indicated among 179,417 subjects analyzed.

c,d. Distribution of mCAs by chromosome and copy number in BioBank Japan and UK Biobank.

e. Chromosomal coverage of loss and CN-LOH events in BioBank Japan and UK Biobank. Curves indicate frequencies at which each chromosomal position is contained in loss (resp. CN-LOH) events, normalized to 1 on each chromosome.

Numeric data are provided in Supplementary Tables 6 and 13.

Inevitability of mCAs in the elderly

The long-lived Japanese population revealed that clonal hematopoiesis with mCAs becomes extremely common in the very old: detectable mosaicism reached 40.7% (standard error 2.3%) in males and 31.5% (standard error 1.7%) in females over the age of 90 (Fig. 2b and Supplementary Table 6), suggesting inevitability of mCAs in the elderly (Supplementary Note 4). Mosaic chromosomal alterations on different chromosomes and with different copy number changes exhibited various degrees of enrichment in males and in the elderly (Extended Data Fig. 1, Supplementary Tables 7–8 and Supplementary Note 4) and in individuals with anomalous blood counts (Supplementary Table 9), suggesting a spectrum of biological processes involved in the development of different clones.

Population differences in mCA distributions

To compare the genomic distributions of mCAs in the Japanese and UK populations, we co-analyzed BBJ mCAs together with 19,632 autosomal mCAs detected in parallel work¹⁵ on the UK Biobank (UKB) cohort^16,17 (Fig. 2c–e and Supplementary Note 3 and 5, Supplementary Table 10).

The Japanese have a 10-fold higher incidence of adult T-cell leukemias¹⁸ and 5-fold lower incidence of chronic lymphocytic leukemia (CLL, a B-cell malignancy) compared to Europeans^19,20. Our analysis indicated that, even among people without cancer, Japanese and UK populations have dramatically different rates of hematopoietic clones arising in the B- and T-cell lineages, as evidenced by deletions that report on V(D)J recombination in developing T and B lymphocytes and thus identify clonal expansions in the T- and B-cell lineages. Mosaic deletions at the TCR alpha locus on 14q (indicating clonal expansion in the T-cell lineage; Supplementary Note 5) were common in BBJ but rare in UKB (82% vs 11% of loss events on chr14 in BBJ and UKB, respectively); in contrast, deletions at the IGH and IGL immunoglobulin loci (indicating clonal expansion in the B-cell lineage) were common in UKB but rare in BBJ (5% vs 39% of loss events on chr14 and 2% vs 58% of loss events on chr22 in BBJ and UKB, respectively) (Fig. 2e and Supplementary Note 5). (We verified that these differences did not arise from differences in genomic coverage by the genotyping arrays used by BBJ and UKB (Extended Data Fig. 2).) Clones arising from the T-cell lineage (as evidenced by deletions at TRA) were also associated with elevated lymphocyte counts (Supplementary Tables 11–12). Thus, the differences in rates of B- and T-cell malignancies between Japanese and UK populations appear to be preceded by distinct relative rates of subclinical clonal expansions in these lineages.

mCAs affect the various human chromosomes at different frequencies. The frequency of CN-LOH varied across chromosome arms in a way that strongly correlated between BBJ and UKB (R=0.73, p=0.00013), with the exception of 14q (more common in BBJ) and 13q (more common in UKB) (Fig. 2c,d, Extended Data Fig. 3 and Supplementary Table 13). In contrast, the most common loss and gain events in each population (including 20q-, 13q-, 10q-, +21, +15, and +12 events) tended to be much more common in one population than the other (Fig. 2c,d and Supplementary Table 13).

A clear pattern among the most strongly population-differentiated mutations involved the 2- to 6-fold lower frequency in BBJ of +12, 13q-, and 13q CN-LOH events (Fig. 2d): all three mutations are commonly observed in CLL^21,22 and in individuals who later develop CLL¹⁰. Considering the 4–5 times lower incidence of CLL in East Asians, the observation that all three of these precursor mutations are also less common in Japanese hematopoietic clones that have expanded to detectable cell fractions suggests that this population difference in CLL risk originates in reduced selective advantage for clones with (diverse) CLL precursor mutations. Consistent with this hypothesis, we observed that clonal sizes for these events tended to be lower in BBJ than UKB (Supplementary Note 3).

The sub-chromosomal distributions of mosaic deletion events were broadly similar between BBJ and UKB but exhibited a few notable differences (Fig. 2e, Supplementary Table 14 and Supplementary Note 5). Focal deletions frequently targeted DNMT3A, TET2, ETV6, NF1, and CHEK2, as in UK Biobank and previous studies^1,2,5,6,10 (Fig. 1 and Fig. 2e). Interestingly, the CLL-related deletion region at 13q14 was less focal in BBJ than in UKB (Fig. 2e), involving longer deletions in a pattern more similar to the 20q, 5q, and 11q deletion regions. We also observed new focal deletion regions in BBJ: at FHIT on 3p, TNFAIP3 on 6q, ABCA1 on 9q, and PTEN on 10q (Fig. 2e, Supplementary Tables 15 and 16); FHIT, TNFAIP3, and PTEN are known tumor suppressor genes associated with blood cancers^23–25.

Inherited risk variants for mCAs in cis

Recent work has established an inherited component of clonal hematopoiesis involving both common variants that slightly increase risk (of clones with any mutation)^11–13,26 and rare variants that strongly predispose to developing clones with specific mCAs in cis¹⁰. The large number of mCAs detected among the Japanese, together with the presence of distinct low-frequency alleles in Japan, could enable detection of additional risk loci. To identify inherited variants associated with mCAs, we first performed association tests aimed at detecting CN-LOH events in cis that promoted clonal expansion by making risk alleles homozygous or removing them from the genome¹⁰ (the two-hit model²⁷). We tested variants imputed into BBJ using the 1000 Genomes Phase 3 reference panel²⁸ together with 1,037 sequenced Japanese samples²⁹, setting a significance threshold of p<5×10⁻⁹ (Methods). We further performed binomial tests to determine whether each risk allele was consistently duplicated or removed by CN-LOH events (in individuals heterozygous for the risk allele, Methods).

We identified five new loci at which inherited variants associated with mosaic CN-LOH events in cis (we also replicated previously reported associations at JAK2^30–32 and MPL¹⁰) (Table 1, Extended Data Figs. 4–6 and Supplementary Note 6). Three of the new loci—NBN, MRE11, and CTU2—involved rare variants with large effects. At NBN, the rare stop-gained variant rs756831345 on 8q associated strongly (OR=91 (52–159), p=9.8×10⁻²³) with 8q CN-LOH events, which consistently made the NBN risk allele homozygous (p=0.00012; Table 1 and Extended Data Figs. 5 and 6). At MRE11, a very rare intronic variant (probably tagging a different causal variant; Supplementary Note 7 and Supplementary Table 17) on 11q associated strongly (OR=37 (17–84), p=2.6×10⁻⁹) with 11q CN-LOH events, which always made the MRE11 risk allele homozygous (p=0.016; Table 1 and Extended Data Figs. 5 and 6). Consistent with strong proliferative advantage in these clones, we observed that these rare risk alleles further associated with detection of multiple CN-LOH clones on the same chromosome arm (with different proximal breakpoints; Extended Data Fig. 7, Extended Data Table 1, and Supplementary Notes 3 and 6). NBN, MRE11, and RAD50 (which did not exhibit a similar association; Supplementary Table 18) encode the components of the MRN double strand break repair complex, which recruits ATM in response to DNA damage, leading to phosphorylation of p53 and CHK2 and initiation of cell cycle arrest, apoptosis, or DNA repair³³. Together with the observations of focal deletions at ATM, TP53, and CHEK2 (Fig. 1) and rare ATM risk alleles for CN-LOH events in cis¹⁰, these results indicate a key role of DNA damage-response dysfunction in clonal selection.

Table 1.

Genome-wide significant associations between inherited variants and mosaic chromosomal alterations.

Mosaic event	Locus	Chr	Pos	Variant	REF	ALT	AF	P	OR (95%CI)	Allelic imbalance in hets
										NREF:NALT	P
Novel associations with mCAs in cis
8q CN-LOH	NBN	8	90949282	rs756831345	C	A	0.00061	9.8×10−23	91 (52–159)	0:14	0.00012
11q CN-LOH	MRE11	11	94160189	11:94160189	G	A	0.00011	2.6×10−9	37 (17–84)	0:7	0.016
14q CN-LOH	NEDD8/TINF2	14	24711798	rs28372734	C	G	0.073	1.0×10−11	1.62 (1.42–1.85)	58:176	5.2×10−15
14q CN-LOH	TCL1A	14	96180242	rs1122138	C	A	0.05	0.015	0.88 (0.79–0.98)	231:107	1.3×10−11
14q CN-LOH	DLK1	14	101175967	rs10873520	G	A	0.30	7.1×10−39	1.38 (1.31–1.44)	1121:689	2.5×10−24
16q CN-LOH	CTU2	16	88781475	rs200779411	C	T	0.00065	7.3×10−20	28 (17–45)	2:11	0.022
Previously reported loci associated with mCAs in cis
1p CN-LOH	MPL	1	45444734	rs560932816	G	A	0.00016	5.3×10−18	54 (30–100)	14:0	0.00012
		1	44074454	rs190159566	C	T	0.0049	2.5×10−13	4.8 (3.4–6.7)	29:6	0.00012
		1	47704269	rs556241419	G	A	0.000062	2.2×10−11	81 (34–191)	7:0	0.016
		1	44579360	rs184778092	C	T	0.00005	5.0×10−9	53 (22–128)	6:0	0.031
9p CN-LOH	JAK2	9	5026293	rs2183137	A	G	0.24	1.1×10−11	1.68 (1.46–1.95)	54:155	1.7×10−12
Novel associations with mCAs in trans
14q CN-LOH	TERT	5	1287194	rs2853677	A	G	0.31	1.5×10−22	1.27 (1.21–1.33)	-	-
chr15 gain	MAD1L1	7	1975624	rs12699483	C	G	0.42	6.9×10−23	1.61 (1.46–1.77)	-	-

Chr, chromosome. Pos, base pair position (hg19). REF, reference allele. ALT, alternate allele. AF, allele frequency in controls. P, association P-value. OR, odds ratio (using ALT as effect allele). N_REF (resp. N_ALT), number of individuals heterozygous for the variant in which a mosaic CN-LOH event in cis made the REF allele (resp. ALT allele) homozygous and removed the other allele. At the MPL locus, P-values for the second, third, and fourth SNPs are from stepwise conditional analysis (Methods); the four risk haplotypes are present in disjoint sets of 1p CN-LOH carriers. Analytic results using 173,599 subjects are shown. Subjects carrying other types of mCAs in the same chromosome are excluded from controls (Methods). Fisher’s exact test (two-sided) is used for the associations and binomial test (two-sided) is used for allelic imbalance associations in heterozygote subjects.

At CTU2, the rare missense variant rs200779411 associated strongly (p=7.3 ×10⁻²⁰, OR=28 (17–45)) with 16q CN-LOH events, which consistently made the CTU2 risk allele homozygous (p=0.022; Table 1 and Extended Data Figs. 5 and 6). CTU2 encodes a component of the cytosolic thiouridylase complex, which is required for maintenance of genome integrity³⁴. The missense variant rs200779411 was predicted to be probably damaging by PolyPhen-2³⁵ and deleterious by SIFT³⁶, suggesting that impaired CTU2 function may promote clonal expansion by reducing genome stability.

Inherited risk variants for mCAs in trans

To additionally detect inherited variants associated with mCAs in trans, we performed genome-wide association tests on each mCA type (classifying events by chromosome and copy number), setting a genome-wide significance threshold of p<5.7×10⁻¹¹ to account for multiple hypotheses tested (Methods). Two trans associations reached significance: common variants in MAD1L1 associated with gains on chromosome 15, and common variants in TERT (previously associated with mosaic JAK2 V617F mutation¹²) associated with 14q CN-LOH (Table 1, Extended Data Figs. 4 and 5, and Supplementary Note 6). At MAD1L1, a cluster of five SNPs in near-perfect linkage disequilibrium (including the missense variant rs1801368) associated (p=6.9 ×10⁻²³, OR=1.61 (1.46–1.77); Table 1 and Extended Data Fig. 5d) with chromosome 15 gain events (mostly full trisomies; Fig. 1). We replicated this association in the UK population with a slightly reduced effect size (p=5.1×10⁻⁴, OR=1.40 (1.16–1.69) for rs1801368). MAD1L1 encodes a component of the mitotic spindle assembly checkpoint (SAC) that ensures proper chromosome segregation³⁷. The MAD1L1 risk allele was also previously observed to increase risk of mosaic Y chromosome loss¹³, consistent with a mechanism involving mis-segregation of chromosomes during mitosis due to impaired SAC function. Lending further support to this hypothesis, the risk haplotype was estimated to also increase risk for large (arm-level or whole-chromosome) gain events in 9 out of 10 chromosomes with at least 50 such events (binomial p=0.02; Supplementary Table 19).

Population-private mCA risk alleles

A comparison of the mCA risk loci detected in BBJ with previously-reported loci from UK Biobank¹⁰ revealed ways in which genetic background can differentially shape clonal hematopoiesis in different populations (Supplementary Note 6 and Supplementary Tables 20–23). Four of the risk variants we found in BBJ (at NBN, MRE11, NEDD8/TINF2, and CTU2) were present at much lower allele frequencies in Europeans³⁸ (Supplementary Table 20). Conversely, all rare variants previously associated with mCAs in UKB (at MPL, FRA10B, ATM, and TM2D3/TARSL2)¹⁰ were absent from the 1000 Genomes Japanese WGS imputation panels, with the absence of FRA10B fragile alleles explaining the lack of 10q25.2-qter deletions in BBJ (Fig. 1). Notably, MPL variants were associated with 1p CN-LOH events in both BBJ and UKB despite most risk alleles in each cohort being population-private, demonstrating a shared path to mosaicism initiated by different variants in different populations.

mCAs and mortality in Japan

Clonal hematopoiesis has previously been linked to poorer health outcomes, with various types of mosaic events observed to increase risk of future blood cancers, mortality, and cardiovascular disease^1–4,10,39. To dissect the link between mCAs and mortality, we analyzed mortality outcomes (including cause of death) available for ~72% of the cohort⁴⁰ (Methods).

We observed a nearly five-fold increase in the risk of death due to leukemia (HR=4.70 (3.26−6.78); Extended Data Fig. 8, Extended Data Table 2, Supplementary Table 24, and Supplementary Note 8). The increased risk of leukemia mortality did not appear to extend to other hematological malignancies (malignant lymphoma and multiple myeloma; Extended Data Fig. 8 and Supplementary Table 24). We also did not observe significantly increased risk of cardiovascular mortality, suggesting that previous associations of clonal hematopoiesis (primarily involving point mutations in DNMT3A, TET2, JAK2, and ASXL1) with cardiovascular outcomes^4,39 may be limited to specific mosaic events (Extended Data Fig 8 and Supplementary Table 24). To refine the association between mosaic status and leukemia mortality, we partitioned mosaic events by chromosome and copy-number change (Methods), identifying six mCAs with significant (p<0.05/88, Cochran-Mantel-Haenszel test), large effects on leukemia mortality risk (Extended Data Fig. 8b and Supplementary Table 25). Mosaic cell fraction and the number of mosaic events carried by an individual each associated with further increases in leukemia mortality risk (Fig. Extended Data Fig. 8c,d and Supplementary Tables 26–27). Mosaic status increased risk of overall mortality (HR=1.10 (1.05–1.16), p=2.7×10⁻⁵; Supplementary Table 24), an association that was driven by mCAs in chromosomes 9 and 14 (HR>1.4; Supplementary Table 28), underscoring the heterogeneity in clinical impact among different mCAs.

Discussion

Our study of hematopoietic clones with mCAs in Japan enabled the first detailed comparison of the genomic landscape of clonal hematopoiesis between populations, revealing broad overall similarities as well as important population differences. A clear pattern among these results was that population differences in cancer rates are preceded by population differences in sub-clinical clonal expansions, at multiple levels: (i) in specific cell lineages (including B- and T-cell lineages), and (ii) with specific cancer-associated mutations (the +12, 13q-, and 13q CN-LOH mutations that are hallmarks of CLL). These results point toward population differences in the clonal advantages that are gained by the same chromosomal mutations in different genetic and environmental contexts.

The interplay between acquired and inherited genetic variation in Japan enabled further insights into the influences of inherited variation on clonal hematopoiesis: population-specific variants at several risk loci pointed to a key role of maintenance of genomic integrity, with corroborating evidence from loci implicated by focal deletions. These results point toward the need for larger and more diverse cohorts in genomic studies of cancer and subclinical clonal expansions as well as inherited variation.

Methods

The BioBank Japan cohort

All the subjects analyzed in this study were participants of the BioBank Japan Project (BBJ). The BBJ is a multi-hospital-based registry that collected clinical information, DNA, and serum samples from approximately 200,000 patients with one or more of 47 target diseases (including 13 cancers) at a total of 66 hospitals between fiscal years 2003 and 2007¹⁴. The case proportions correlated well with prevalence in the Japanese population and all of the study participants were diagnosed by medical doctors as described elsewhere¹⁴. We complied with all relevant ethical regulations. This project was approved by the ethics committees of RIKEN Center for Integrative Medical Sciences and the Institute of Medical Sciences, the University of Tokyo. Written informed consent was obtained from all of the participants.

Genotyping subjects in the BBJ

Subjects were genotyped in three batches using different arrays or set of arrays, namely: (1) a combination of Illumina Infinium Omni Express and Human Exome, (2) Infinium Omni Express Exome v1.0, and (3) Infinium Omni Express Exome v1.2 (Supplementary Table 1). The SNP content of the three methods was quite similar. DNA was obtained from blood samples for all but one individual (for which DNA was obtained from oral mucosa; this sample was negative for mosaic events).

We excluded outliers from East Asian clusters in a plot in which we projected BBJ subjects in combination with 1000 Genomes Project⁴¹ samples in the principal component (PC)1 and PC2 space. We also excluded samples genetically identical to another sample, samples with call rates less than 0.98, and samples whose reported sex information was not supported by genotypes in the X chromosome. We further excluded three samples with evidence of potential contamination (as suggested by low-cell-fraction mosaic events called on many chromosomes^6,10), leaving 179,417 samples for analysis. We used plink1.9 software⁴² to handle the genotyping data.

Genotyping intensity data used for calling mosaic events

To call mosaic events, we analyzed genotyping intensity data for variants in the intersection of the three primary arrays used for BBJ genotyping (namely, Illumina Infinium Omni Express and Infinium Omni Express Exome v1.0 and v1.2) to allow analysis of the same set of variants in all individuals (to avoid the possibility of differing detection sensitivity across batches due to different numbers of genotyping probes analyzed). When calling mosaic events, we did not include variants typed on the Human Exome array in some samples (see above) to minimize the potential for batch effects arising from different arrays. We did use variants from the Human Exome array in genetic association analyses (see below) since association tests are robust to genotyping heterogeneity when potential confounders are appropriately controlled by correcting for batch covariates and principal components.

Calculation of BAF and LRR from genotype intensity

We computed B allele frequency (BAF) and log R ratio (LRR) values with the use of the BBJ genotyping intensity data⁴³. We modified previous methods used by Jacobs et al¹ and Loh et al¹⁰ to fit the current data set. We computed LRR and BAF values in an array-basis in which all of the subjects genotyped in the same arrays were clustered together. Details are provided in the Supplementary Note 1.

Phasing of genotype data for calling mosaic events

We phased the filtered genotypes mentioned above with the use of Eagle2 software⁴⁴ which enabled us to conduct accurate long-range phasing. This phasing information was used for calling mosaic events (Supplementary Note 1).

Filtering possible non-mosaic trisomy or monosomy events

We excluded chromosomes with mean LRR more than 0.2 or mean LRR less than −0.5 (possible trisomy and monosomy, respectively, Supplementary Note 1).

Calling mosaic events with the use of BAF and LRR

We used the same method to call mosaic events as Loh et al¹⁰. This calling method is composed of the following steps: (1) filtering constitutional duplications; (2) evaluating phased BAF for variants on each chromosome using a parameterized hidden Markov model; (3) calling existence of events using a likelihood ratio test; (4) calling event boundaries; (5) calling copy number; (6) filtering remaining possible constitutional duplications; (7) estimating cell fraction of mosaic events. Details of each step are provided in the Supplementary Note 1.

Associations between array batches or disease status at registry and detectable mosaicism

We conducted logistic regression analyses to evaluate associations between detectable mosaicism and 1) array batches or 2) disease status at registry of the BBJ (47 diseases, a binary trait for each of the diseases). For array batches (Supplementary Table 4), we put mosaic detection status as a dependent binary variable and age, sex, smoking, genotyping arrays, and 10 principal components as independent variables. For disease status at registry (Supplementary Table 5), we put disease presence as a dependent variable and presence of mosaic events, age, sex, genotyping arrays, and 10 principal components as independent variables.

Associations between hematological traits and mosaic events

We extracted data from the BBJ of 13 hematological traits, namely, red blood cell count, hemoglobin, hematocrit, mean corpuscular volume, mean corpuscular hemoglobin, mean corpuscular cell hemoglobin, white blood cell count, neutrophil count, lymphocyte count, monocyte count, eosinophil count, basophil count and platelet count. Associations between 13 hematological quantitative traits and presence of 88 types of mosaic events (see below) were analyzed in logistic regression models with event presence as outcomes. Before analysis in logistic models, the 13 traits were regressed out by covariates specified in the previous BBJ study⁴⁵ for males and females. Residuals were normalized and used as independent variables one by one (a total of 13 models for each mCA). In each logistic model, disease status at registry (for each of the 47 diseases in the BBJ study design), age, sex, smoking, genotyping arrays, and 10 principal components were used as covariates. We took this approach to avoid effects of covariates associated with both mCAs and hematologic traits.

We subdivided mosaic events by copy-number state (loss, CN-LOH, gain) and by p vs. q arm for loss and CN-LOH events. To reduce multiple testing burden, we restricted analyses to mosaic events with more than 20 carriers (Supplementary Table 2). As a result, 88 mosaic events were analyzed in association with 13 hematologic traits. The statistical significance threshold was set at p<0.05/88/13 (4.4×10⁻⁵); results are reported in Supplementary Table 9.

Comparison of mosaic frequency between BBJ and UKB

We co-analyzed BBJ mosaic calls with mosaic calls in UKB from 482,857 individuals¹⁵. We calculated frequencies of mosaic events subdivided by chromosome arm and copy number among all mosaic events in both data sets. We assessed correlation of event frequencies in the two data sets using Spearman’s correlation coefficient.

Relative coverage of the genome by mosaic events in BBJ and UKB

We determined mosaic coverage as follows. We divided chromosomes into 0.1Mb bins and calculated the fraction of loss (respectively, CN-LOH) events covering each bin to compute mosaic coverage. We scaled the coverage in each mosaic type in each chromosome (to set maximum coverage as 1). We compared mosaic coverage in BBJ and UKB using Pearson’s correlation coefficient.

Genomic coverage by genotyping arrays in BBJ and UKB

We computed mean numbers of heterozygous genotyped sites across individuals in each 1Mb region of the genome for the BBJ and UKB genotyping arrays to confirm difference in mosaic frequency between the two populations were not driven by different coverage of the genome by DNA microarrays.

Association between mosaic events indicating T-cell expansions and lymphocyte counts

We used a Wilcoxon rank sum test to compare lymphocyte counts between subjects carrying TRA deletions (indicating clonal expansions of T-cells) and subjects without TRA deletion. We also evaluated Spearman correlation between cell fraction of TRA deletions and lymphocyte counts.

Distribution of breakpoints of CN-LOH in BBJ and UKB

We computed relative frequencies of estimated CN-LOH breakpoint locations in each chromosome in BBJ and UKB. We smoothed breakpoints over +/−2Mb and rescaled to 1.

Genes affected by focal deletion

We evaluated importance of genes by taking numbers of gene involved in loss events into account. We counted the number of genes involved in each loss event and defined a score of each loss event as one divided by the number of genes (i.e., when a loss event contained only one gene, the gene received a score of 1). We summed up scores of all loss events containing each gene. To pick up genes frequently involved with focal deletions only in the Japanese, we identified genes covered by at least 5% of loss events in a chromosome, having a 10-fold larger score in BBJ than in UKB, and scoring more than 0.5.

Genetic association studies

We excluded subjects showing high degree of kinship (1^st-degree or closer as detected by plink⁴²) with other subjects, leaving 173,599 subjects for genetic association studies. Among related pairs, we retained subjects having mosaic events. We also integrated the genotyping data used for calling mosaic events with genotyping data from additional variants typed on the Human Exome Array in some samples when also available on the Omni Express Exome Arrays in other samples (Supplementary Table 1) to maximize the number of variants used for imputation. (We did not integrate these data at the stage of calling mosaic events to minimize the potential for batch effects.) We phased the integrated data using Eagle2 software⁴⁶. The phased genotypes were imputed using a reference panel containing 2,504 1000 Genomes Phase 3 samples and 1,037 Japanese high-depth (30x) whole-genome sequencing samples (Data set 1 of ref.²⁹) using Minimac3 software⁴⁷. Variants imputed with R²>0.3 were used for the association studies. We filtered variants with minor allele count less than 5. Best guess data was used to conduct Fisher’s exact tests using plink software (plink --fisher --ci 0.95). We used Fisher’s exact test to prevent inflated type I error when testing associations between rare variants and rare mosaic events⁴⁸. To confirm that significant associations were not driven by confounding factors, we re-analyzed significant associations (detected by Fisher’s exact test) in logistic regression with and without covariates (10 principal components, disease status at registry, age, sex, smoking and genotype batches) and verified that the associations were robust. We used GWAS array genotyping data if available to rescue rare variants not included in the reference panel, not well-imputed or with low allele frequency. As a result, 26.6 million variants were used for association studies.

We analyzed mosaic events in each chromosome as distinct phenotypes, treating loss, CN-LOH and gain separately. In order to maximize power to identify significant associations with CN-LOH, we included unclassified “likely CN-LOH” events extending to one telomere with |LRR|<0.02 when testing variants for association with CN-LOH events. We subdivided loss and CN-LOH events in each chromosome into p-arm and q-arm events. We set a threshold of at least 20 event carriers to consider an event in genetic association studies. This led to a total of 88 copy number-chromosome pairs analyzed (Supplementary Table 2). We tested each of these phenotypes for association with variants in cis (i.e., on the same chromosome and contained within a mosaic event) or trans (i.e., on any chromosome). For cis associations, we also conducted allelic imbalance analyses to assess whether one of the alleles at each variant was preferentially duplicated by mosaic CN-LOH events. Details of each test and corresponding significance thresholds are described in Supplementary Note 6.

At significantly-associated loci, we additionally performed stepwise conditional analyses (by iteratively removing carriers of high-risk rare alleles) to test for additional independently-associated variants.

Associations between risk variants and presence of multiple CN-LOH clones with different breakpoints

In a small fraction of individuals, we detected evidence of multiple clonal expansions of CN-LOH events affecting the same chromosome arm but with different breakpoints. To detect such events, we applied a modified hidden Markov model as described previously¹⁰ (Supplementary Note 3). Briefly, this analysis searched for evidence of CN-LOH with increasing BAF deviation toward the telomere. We evaluated associations between presence of risk variants found above and presence of single vs. multiple breakpoints among individuals with CN-LOH spanning the variants using Fisher’s exact test. Details are described in Supplementary Note 6.4.6 (“Multiple breakpoints driven by rare penetrant variants”).

Associations between mosaic events and mortality

The BBJ project has follow-up data to survey mortality and cause of death⁴⁰. A total of 141,612 BBJ subjects who have one of 32 out of 47 diseases were prospectively followed up (~10 years) after DNA collection. For subjects who died, further detailed surveillance was made to identify causes of death coded by ICD10 by accessing national vital registration system used for input survey of medical and social welfare at Ministry of Health, Labor, and Welfare Japanese Government.

We restricted subjects to those who were followed at least one year after registry and free from malignancy at blood collection. We found 86,546 subjects in the current study were included in the follow-up data for mortality. Among them, 16,812 deaths were recorded during the follow-up period. The average follow-up period was 7.6 years (median 8.3 years, s.d. 2.8 years).

Associations between mortality (overall or specific causes) and presence of mosaic events (regardless of mosaic types) were analyzed as an initial evaluation. We analyzed overall mortality, hematopoietic malignancy mortality and non-hematopoietic malignancy mortality. We compared subjects with mosaic events (loss, CN-LOH, or gain) at cell fraction ≥1% to subjects without mosaic events on any chromosomes. Cox regression analysis was used for the analyses conditioning for age, agê2, sex, disease status, genotyping array and smoking. We used follow-up period as censoring factor. When we analyzed specific causes of death (e.g., non-hematopoietic malignancy), we only used subjects whose deaths were not reported during follow-up as controls to use consistent control samples across analyses. We used significance thresholds based on Bonferroni’s correction (based on the number of tested mortality phenotypes).

After evaluating associations between mortality phenotypes and presence of any mosaic event in any chromosome, we searched for associations between specific mosaic event types and mortality. We analyzed the same set of 88 mosaic event types (defined by copy-number state and chromosomal location) we used when testing associations with hematological traits and inherited variants. In these analysis of associations between mortality phenotypes and specific mosaic types, we divided subjects based on age, sex and smoking status and computed associations using Cochran-Mantel-Haenszel tests to avoid inflation of statistics arising from small number of subjects carrying mosaic events. We set a significance level based on Bonferroni’s correction (p<0.00057, 0.05/88).

We also analyzed cardiovascular mortality (defined as ischemic heart diseases and ischemic stroke) as previous studies have reported associations between mosaic point mutations and cardiovascular outcomes.

Definition of cancers based on ICD10 codes for causes of death

We categorized causes of death to decrease multiple testing burden. Hematopoietic malignancy was defined by ICD10 codes of C81-C96 and D45, 46 and D47. Leukemic diseases were defined by ICD10 codes of C91-C96, D45 and D46. Malignant lymphoma was defined by ICD10 codes of C81-C88. Multiple myeloma was defined by C90. Cancers were defined as ICD10 codes starting with “C” together with hematopoietic malignancies defined not starting with “C”. We did not regard other ICD10 codes starting with “D” as cancer since most of them are benign tumors.

Associations between multiple mosaic events and mortality

We extended the mortality analyses to investigate the effect of multiple mosaic events within a single individual. We limited analyses to subjects with at most three mosaic events. We divided subjects into three groups: (1) subjects without mosaic events, (2) subjects with a single mosaic event, (3) subjects with multiple mosaic events (2 or 3 mosaic events in different chromosomes). We analyzed an association of presence of multiple mosaic events with leukemia mortality in comparison with presence of a single mosaic event. The analyses conditioned on age, sex, disease status, genotyping array and smoking status.

Associations between cell fraction of mosaic events and mortality

We also extended the mortality analyses to investigate the effect of mosaic cell fraction. For subjects with multiple mosaic events, we took the highest cell fraction. We divided subjects into categories according to cell fraction of mosaic events and analyzed associations between cell fractions and outcomes with which presence of mosaic events were significantly associated.

Extended Data

Extended Data Fig.1|.

Age and sex of carriers of mosaic event types.

Mean age and sex of carriers of specific mCA types (defined by chromosome and copy number) with at least 100 carriers in the 179,417 subjects. Marker sizes are proportional to mCA frequencies. Error bars, s.e.m. Numeric data are provided in Supplementary Table 7.

Extended Data Fig.2|.

Comparable chromosomal coverage by heterozygous genotypes in BBJ and UKB.

Average numbers of heterozygous genotyped sites (averaged across individuals) in each 1Mb region of the genome for the BioBank Japan and UK Biobank genotyping arrays.

Extended Data Fig.3|.

Similar breakpoint distributions of CN-LOH events in the BBJ and UKB.

Relative frequencies of estimated CN-LOH breakpoint locations in BioBank Japan and UK Biobank. Breakpoints were smoothed over +/−2Mb to allow plotting of frequency curves, which were rescaled to 1.

Extended Data Fig.4|.

Quantile-quantile plots of mosaic events with significant associations demonstrating no inflation of association statistics.

Quantile-quantile plots of results for mosaic events with significant associations are indicated. Analytic results of Fisher’s exact test (two-sided, nominal p-values) using 173,599 subjects are shown. We defined as hit loci 42–49M at chr1 (1p CN-LOH), 88–94M at chr8 (8q CN-LOH), 92–96M at chr11 (11q CN-LOH), 88–90M at chr16 (16q CN-LOH), 23–26M and 100–103M at chr14 (cis association of 14q CN-LOH), 4–6M at chr9 (9p CN-LOH), 0–2M at chr5 (trans association of 14q CN-LOH) and 1–3M at chr7 (trans association of chr15 gain).

Extended Data Fig.5|.

Local plots for cis and trans associations.

Associations of inherited variants with a. 8q CN-LOH, b. 11q CN-LOH, c. 16q CN-LOH, d. chr15 gain, e.1p CN-LOH, f.9p CN-LOH, and g-i. 14q CN-LOH are shown for regions containing the NBN, MRE11, CTU2, MAD1L1, MPL, JAK2, NEDD8/TINF2, DLK1 and TERT loci, respectively. a-c and e indicate rare cis associations, f-h indicate common cis associations and d and i indicate trans associations. a-d and g-i are unreported regions. Purple points indicate lead variants. Other variants are color-coded according to LD with lead variants. The TCL1A variant that significantly associated with 14q CN-LOH allelic imbalance is not shown here because it did not significantly associate with 14q CN-LOH risk. Analytic results of Fisher’s exact test (two-sided, nominal p-values) using 173,599 subjects are shown.

Extended Data Fig.6|.

Action of CN-LOH events on rare and common inherited variants.

Schematic images showing the patterns of selection or elimination of inherited variants by CN-LOH events. Asterisks indicate risk alleles. (For the TCL1A locus, which did not significantly associate with presence of 14q CN-LOH, we depict TCL1A as a gene for which mCAs select an allele.)

Extended Data Fig.7|.

Examples of multiple overlapping CN-LOH clones in a single chromosome.

We identified 185 individuals carrying multiple CN-LOH clones on a single chromosome.

a.Multiple clones were observed in at least one individual for all chromosomes except chromosomes 18, 20 and 22. The plots show phased BAF deviations (y-axis) as a function of chromosome position (x-axis) for the individual with the largest clone per chromosome (among all individuals with multiple CN-LOH clones on that chromosome). Colored horizontal lines of different colors indicate distinct BAF deviations corresponding to overlapping CN-LOH events.

b.the number of subjects carrying multiple CN-LOH clones on a single chromosome is shown in each chromosomal arm.

Extended Data Fig.8|.

Mortality risk conferred by mosaic chromosomal alterations.

a. Risk of mortality from various causes conferred by presence of an mCA at >1% cell fraction. Leukemia, malignant lymphoma, and multiple myeloma are subdivisions of blood cancer. Cardiovascular mortality includes deaths from coronary artery disease (CAD) and ischemic stroke (IS).

b. Risk of leukemia mortality conferred by specific mCAs (grouped by chromosomal location and copy-number change) reaching Bonferroni significance.

c. Risk of leukemia mortality conferred by mosaic status stratified by mosaic cell fraction.

d. Risk of leukemia mortality conferred by mosaic status stratified by mosaic cell fraction and number of mosaic events detected (one vs. two or more).

All analyses were restricted to individuals with no previous cancer diagnosis and were corrected for age, sex, smoking status, and genotyping array (Methods). Error bars, 95% CIs. Numeric data are provided in Supplementary Tables 24–27. Results using 86,546 subjects are indicated. Cox Proportional Hazard model (two-sided) is applied for a, b, and d. Cochran-Mantel-Haenszel test is applied for c.

Extended Data Table 1.|.

Rare variants associated with CN-LOH further increase risk of multiple overlapping CN-LOH clones.

Variant	Gene	mCA	Ncarrier / Nmulti-CN-LOH	p	OR (95%CI)
Rare variants in novel genes
rs756831345	NBN	8q CN-LOH	1 / 1	0.11	Inf(0.22-Inf)
11:94160189	MRE11	11q CN-LOH	3 / 12	8.0×10−5	64.8 (8.3–446.9)
rs200779411	CTU2	16q CN-LOH	0 / 4	1	-
Previously-reported MPL locus
rs560932816	MPL	1p CN-LOH	1 / 20	0.29	3.2(0.071–23.8)
rs190159566			0 / 20	1	-
rs556241419			1 / 20	0.17	6.4 (0.13–57.1)
rs184778092			1 / 20	0.15	7.7 (0.16–73.9)
Aggregated very rare variants*			3 / 20	0.027	5.5(1.0–21.0)
Common variants
rs2183137	JAK2	9p CN-LOH	20 / 25	0.086	2.50 (0.88–8.69)
rs28372734	NEDD8/TINF2	14q CN-LOH	3 / 17	1	0.76 (0.14–2.75)
rs10873520	DLK1	14q CN-LOH	42 / 57	0.074	1.77 (0.96–3.44)

N_carrier / N_multi-CN-LOH indicates the fraction of subjects carrying multiple clones spanning the variant who also carried the indicated risk variant(s).

OR: odds ratio for carrying multiple clones spanning the variant vs. carrying a single clone CI:confidence interval

P:p-value in Fisher’s exact test (two-sided)

^*aggregating rs560932816, rs556241419 and rs184778092 with variant frequency less than 0.001

Extended Data Table 2.|.

Breakdown of associations between mCAs and death of leukemia.

Phenotype	OR (95%CI)	p value
All leukemia	4.70 (3.26–6.78)	1.0×10−16
1) myeloid leukemia	5.18 (3.35–8.01)	1.4×10−13
2) lymphoid leukemia	3.77 (1.72–8.26)	0.00093
2-1) B cell lymphoid leukemia	4.11 (1.42–11.85)	0.0090
2-2) T cell lymphoid leukemia	3.41 (1.07–10.92)	0.039

CI: confidence interval

Cochran-Mantel-Haenszel test is applied for 86,546 subjects.