Search

Clonal hematopoiesis of indeterminate potential (CHIP) is a common form of age-related somatic mosaicism that is associated with significant morbidity and mortality. CHIP mutations can be identified in peripheral blood samples sequenced using approaches that cover the whole genome, whole exome or targeted genetic regions; however, differentiating true CHIP mutations from sequencing artifacts and germline variants is a considerable bioinformatic challenge. We present a stepwise method that combines filtering based on sequencing metrics, variant annotation, and novel population-based associations to increase the accuracy of CHIP calls. We apply this approach to ascertain CHIP in ∼550,000 individuals in the UK Biobank complete whole exome cohort and the All of Us Research Program initial whole genome release cohort. CHIP ascertainment on this scale unmasks recurrent artifactual variants and highlights the importance of specialized filtering approaches for several genes includingTET2andASXL1. We show how small changes in filtering parameters can considerably increase CHIP misclassification and reduce the effect size of epidemiological associations. Our high-fidelity call set refines prior population-based associations of CHIP with incident outcomes. For example, the annualized incidence of myeloid malignancy in individuals with small CHIP clones is 0.03%/year, which increases to 0.5%/year amongst individuals with very large CHIP clones. We also find a significantly lower prevalence of CHIP in individuals of self-reported Latino or Hispanic ethnicity in All of Us, highlighting the importance of including diverse populations. The standardization of CHIP calling will increase the fidelity of CHIP epidemiological work and is required for clinical CHIP diagnostic assays.

Truncating mutations in the terminal exon of protein phosphatase Mg2+/Mn2+ 1D (PPM1D) have been identified in clonal hematopoiesis and myeloid neoplasms, with a striking enrichment in patients previously exposed to chemotherapy. In this study, we demonstrate that truncating PPM1D mutations confer a chemoresistance phenotype, resulting in the selective expansion of PPM1D-mutant hematopoietic cells in the presence of chemotherapy in vitro and in vivo. Clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated protein-9 nuclease mutational profiling of PPM1D in the presence of chemotherapy selected for the same exon 6 mutations identified in patient samples. These exon 6 mutations encode for a truncated protein that displays elevated expression and activity due to loss of a C-terminal degradation domain. Global phosphoproteomic profiling revealed altered phosphorylation of target proteins in the presence of the mutation, highlighting multiple pathways including the DNA damage response (DDR). In the presence of chemotherapy, PPM1D-mutant cells have an abrogated DDR resulting in altered cell cycle progression, decreased apoptosis, and reduced mitochondrial priming. We demonstrate that treatment with an allosteric, small molecule inhibitor of PPM1D reverts the phosphoproteomic, DDR, apoptotic, and mitochondrial priming changes observed in PPM1D-mutant cells. Finally, we show that the inhibitor preferentially kills PPM1D-mutant cells, sensitizes the cells to chemotherapy, and reverses the chemoresistance phenotype. These results provide an explanation for the enrichment of truncating PPM1D mutations in the blood of patients exposed to chemotherapy and in therapy-related myeloid neoplasms, and demonstrate that PPM1D can be a targeted in the prevention of clonal expansion of PPM1D-mutant cells and the treatment of PPM1D-mutant disease.

Introduction: Clonal hematopoiesis of indeterminate potential (CHIP) may occur when a hematopoietic stem cell (HSC) acquires a fitness-increasing mutation resulting in its clonal expansion. A diverse set of driver genes, such as regulators of DNA methylation, splicing, and chromatin remodeling, have been associated with CHIP, but it remains largely unknown why HSCs bearing these mutations are positively selected. It has been challenging to identify the genetic and environmental factors mediating clonal expansion in humans, partially due to a lack of large cohorts with longitudinal blood sampling of participants. To circumvent this limitation, we developed a method to infer clonal expansion rate from single timepoint data called PACER (passenger-approximated clonal expansion rate). Methods: PACER is based on the principle that genomic passenger mutations can be used to infer the birth date of pre-malignant clones because these mutations accumulate fairly linearly over time. Thus, an individual with CHIP with a greater number of passenger mutations in the mutant clone is expected to have acquired the clone at a later age than someone with fewer passenger mutations. For two individuals of the same age and with clones of the same size, we expect the clone with more passengers to be more fit, as it expanded to the same size in less time. Typically, one would need to isolate single-cell colonies derived from HSCs in order to calculate the total passenger mutation burden. However, we hypothesized that this measure could also be approximated from whole genome sequencing of blood cell DNA, such as that used in large biobank projects. The expansion rate (PACER) is then estimated by adjusting the total passenger count for age and variant allele fraction in each individual. The ability of passengers to predict future clonal expansion was validated using longitudinal blood samples from 51 CHIP carriers in the Women's Health Initiative taken ~10 years apart (Figure 1). It also accurately predicted the known fitness effects due to different driver mutations in 5,551 CHIP carriers from the Trans-Omics for Precision Medicine (TOPMed) program (Figure 2). Results: Having validated the approach, we next hypothesized that we could identify germline variants influencing PACER, thus revealing genes and pathways mediating clonal expansion. The lead hit in a genome-wide association study (GWAS) of PACER was a common single nucleotide polymorphism (SNP) in the TCL1A promoter that was associated with slower clonal expansion in CHIP overall (Figure 3). TCL1A is an oncogene that is activated via translocation in T-cell prolymphocytic leukemia, but has no known role in CHIP or myeloid malignancies. A gene-level analysis indicated that the TCL1A SNP was associated with slower growth of clones bearing TET2 mutations, but had no effect on DNMT3A-mutant clone growth. We further found that those carrying two copies of the protective SNP had 40-80% reduced odds of having clones with driver mutations in TET2, ASXL1, SF3B1, SRSF2, and JAK2, but not DNMT3A. A concomitant decrease in incident myeloid malignancies was also seen in carriers of this protective SNP. Next, we interrogated how the protective SNP influenced TCL1A activity in HSCs. Normal human HSCs lacked open chromatin at the TCL1A promoter and TCL1A expression, but inducing frameshift mutations in TET2 via CRISPR editing led to accessibility of the promoter and gene/protein expression in HSCs (Figure 4). This effect was abrogated in HSCs from donors of the protective TCL1A SNP in a dose-dependent manner. Finally, we found that HSCs from donors homozygous for the protective SNP had markedly less expansion of phenotypic stem and progenitor cells in vitro after the introduction of TET2 mutations than TET2-edited HSCs from donors with two copies of the reference allele. Conclusions: In summary, we developed a novel method to infer the expansion rate of pre-malignant clones and performed the first ever GWAS for this trait. Our results indicate that the fitness advantage of several common driver genes in CHIP and hematological cancers is mediated through TCL1A activation, which may be a therapeutic target to treat these conditions. PACER is an approach that can be widely adopted to uncover genetic and environmental determinants of pre-malignant clonal expansion in blood and other tissues. Figure 1 Figure 1. Disclosures Desai: Bristol Myers Squibb: Consultancy; Kura Oncology: Consultancy; Agios: Consultancy; Astex: Research Funding; Takeda: Consultancy; Janssen R&D: Research Funding. Natarajan: Blackstone Life Sciences: Consultancy; Boston Scientific: Research Funding; Novartis: Consultancy, Research Funding; AstraZeneca: Consultancy, Research Funding; Apple: Consultancy, Research Funding; Amgen: Research Funding; Genentech: Consultancy; Foresite Labs: Consultancy. Jaiswal: Novartis: Consultancy, Honoraria; AVRO Bio: Consultancy, Honoraria; Genentech: Consultancy, Honoraria; Foresite Labs: Consultancy; Caylo: Current holder of stock options in a privately-held company.

Clonal hematopoiesis of indeterminate potential (CHIP) occurs when hematopoietic stem cells (HSCs) acquire a mutation, most commonly a null variant in TET2 or DNMT3A, that confers a selective advantage. Blood cancers may result if additional cooperating mutations are acquired. However, CHIP may also cause atherosclerosis and other inflammatory diseases because these mutations alter the function or development of effector immune cells derived from the HSCs. Genome-wide association studies have implicated microglia, the resident myeloid cells in the brain, as key players in the biology of Alzheimer's disease (AD). Here, we asked whether CHIP associated with AD dementia or neuropathologic change, and whether mutant marrow-derived cells could be found in the brains of CHIP carriers. To test for an association, we used data from the Trans-omics for Precision Medicine project (TOPMed) and the Alzheimer's Disease Sequencing Project (ADSP), where whole genome or exome sequencing data as well as AD phenotype data was available on 5,730 persons. TOPMed contained population-based cohorts unselected for AD, while ADSP was a case-control study for AD. We surprisingly discovered that the presence of CHIP was associated with a reduced risk of AD dementia in both projects (fixed-effects meta-analysis odds ratio 0.64, p = 3.0 x 10-5, adjusted for age, sex and APOE genotype) (Figure 1). The protective effect of CHIP was strongest in those with APOE e3 or e4 alleles, but not seen in those with APOE e2 allele. No substantial differences in AD risk were seen based on mutated driver gene. In addition, the presence of CHIP was associated with a reduced burden of amyloid plaques and neurofibrillary tangles in the brains of those without dementia. In sum, our human genetic analyses indicated that CHIP was robustly associated with protection from AD dementia and AD-related neuropathologic changes. A causal link between CHIP and AD would be strengthened by finding the mutated cells infiltrating the brain. However, it is presumed that bone marrow progenitors have minimal contribution to the adult microglial pool. To determine if the mutations seen in the blood of CHIP carriers could also be found in the brain, we obtained 8 occipital cortex samples from autopsy of donors with CHIP, 6 of whom were cognitively normal at the time of death. The 8 CHIP carriers had mutations in DNMT3A, TET2, ASXL1, SF3B1, and GNB1 with the highest frequency in DNMT3A and TET2, which is representative of the relative proportion of these mutations in the general population. We detected the CHIP somatic variants in the microglia enriched (NeuN- c-Maf+) fraction of brain in 7 out of 8 CHIP carriers, with a VAF ranging from 0.02 to 0.28 (representing 4% to 56% of nuclei) (Figure 2), but at low levels or absent in the other fractions of brain. We then performed single-cell ATAC-sequencing on brain samples from 2 CHIP carriers and 1 control to specify the cellular population harboring CHIP mutations. This revealed that hematopoietic cells in the 3 samples formed a single myeloid cluster that had accessible chromatin at the microglia marker genes TMEM119, P2RY12, and SALL1, but not in genes specific to monocytes or dendritic cells. We further determined that the proportion of cells in this cluster bearing the CHIP mutations ranged from ~40-80% in these two samples, indicating widespread replacement of the endogenous microglial pool by mutant cells. We show here that, unexpectedly, the presence of CHIP is associated with protection from AD dementia. CHIP is also associated with lower levels of neuritic plaques and neurofibrillary tangles in those without dementia, indicating a possible modulating effect of CHIP on the underlying pathophysiology of AD. Consistent with this hypothesis, we also detect substantial infiltration of brain by marrow-derived mutant cells which adopt a microglial-like phenotype. We speculate that the mutations associated with CHIP confer circulating precursor cells with an enhanced ability to engraft in the brain, to differentiate into microglia once engrafted, and/or to clonally expand relative to unmutated cells in the brain microenvironment. These non-mutually exclusive possibilities could provide protection from AD by supplementing the phagocytic capacity of the endogenous microglial system during aging. Figure 1 Figure 1. Disclosures Jaiswal: Novartis: Consultancy, Honoraria; Foresite Labs: Consultancy; Genentech: Consultancy, Honoraria; AVRO Bio: Consultancy, Honoraria; Caylo: Current holder of stock options in a privately-held company.

Clonal hematopoiesis of indeterminate potential (CHIP) was recently identified as a major risk factor for development of both hematologic malignancies and atherosclerotic cardiovascular disease in humans. The most commonly mutated gene in CHIP, DNMT3A, is a de novo DNA methyltransferase. The second most commonly mutated gene is TET2, an enzyme which can lead to loss of DNA methylation, and thus is thought to have an opposing biochemical function to DNMT3A. Surprisingly, mutations in both genes lead to convergent phenotypes, such as clonal expansion of mutated stem cells, increased risk of malignant transformation, and increased risk of coronary heart disease. A molecular mechanism linking CHIP and cardiovascular disease has been explored only for loss of function mutations in the Tet2 gene (Jaiswal et al., NEJM 2017; Fuster et al., Science 2017). Here we tested the ability of null mutations in Dnmt3a to contribute to atherosclerosis in hypercholesteremic mice. We further explored the biological basis for this association through gene expression analyses and single-cell RNA sequencing. To model cardiovascular disease associated with DNMT3A-mutated CHIP, atherosclerosis-prone Ldlr-/- mice received bone marrow from Dnmt3a+/+ mice (WT), or from Dnmt3a-/- mice (KO) and WT mice in a 1:9 ratio to mimic a typical variant allele fraction observed in human CHIP. Mice then consumed a high-fat, high-cholesterol diet (HFD), and underwent assessment of atherosclerosis. At 9 weeks, mice that had received 10% Dnmt3a-/- bone marrow displayed an average lesion size that was 40% larger compared to mice receiving control marrow only (p=0.04). The increase in lesion size resembles that we previously observed in mice receiving Tet2-/- marrow (Jaiswal et al., NEJM 2017). De novo DNA methylation by Dnmt3a can alter gene expression. To elucidate how such changes may accelerate atherosclerosis, we first performed transcriptome analysis using bulk RNA sequencing of cholesterol-stimulated bone marrow derived macrophages (BMDM) from either WT or KO mice. BMDMs lacking Dnmt3a showed significantly augmented expression of genes belonging to the CXC chemokine cluster, Cxcl1, Cxcl2 and Cxcl3, as well as increases in mRNAs encoding canonical pro-inflammatory cytokines Il1b and Il6. These changes mirrored those we saw in macrophages lacking Tet2 (Jaiswal et al., NEJM 2017). We next asked how transcriptomic changes observed using the ex vivo BMDM system translated into the in vivo lesional environment. Single-cell RNA sequencing (10X Genomics) was performed on atherosclerotic aortae from mice that had been competitively transplanted with WT, Dnmt3a-/-, or Tet2-/- marrow at a 1:9 ratio. Clustering demonstrated broad changes in lesional immune cell composition in mice harboring CHIP. Lack of either Tet2 or Dnmt3a substantially expanded the myeloid compartment, containing cells that drive atherogenesis. A reciprocal reduction mainly affecting T lymphocyte populations accompanied this expansion. Within the myeloid cell compartment, Dnmt3a-/- or Tet2-/- donor cells, but not WT donor cells, gave rise to a distinct lesional macrophage population. These cells expressed markers associated with tissue-resident macrophages (Mrc1, Lyve1, F13a1), but also highly expressed several inflammatory mediators (Cxcl1, Pf4, Ccl2, Ccl7, Ccl8), and a characteristic set of transcription factors (Jun, Fos, Egr1). Overall, the present study reveals broad changes to the lesional cellular composition and transcriptome induced by the most common CHIP mutations, and provides novel insight into the mechanisms by which CHIP accelerates atherosclerosis. Despite exerting opposite catalytic functions, lack of Dnmt3a or of Tet2 function lead to a myriad of similar downstream transcriptomic and cellular changes. These results indicate that mutations in Dnmt3a and Tet2 accelerate atherosclerosis through convergent mechanisms. Disclosures No relevant conflicts of interest to declare.

Recent genetic analyses of large populations have revealed that somatic mutations in hematopoietic cells leading to clonal expansion are commonly acquired during human aging. Clonally restricted hematopoiesis is associated with an increased risk of subsequent diagnosis of myeloid or lymphoid neoplasia and increased all-cause mortality. Although myelodysplastic syndromes (MDS) are defined by cytopenias, dysplastic morphology of blood and marrow cells, and clonal hematopoiesis, most individuals who acquire clonal hematopoiesis during aging will never develop MDS. Therefore, acquisition of somatic mutations that drive clonal expansion in the absence of cytopenias and dysplastic hematopoiesis can be considered clonal hematopoiesis of indeterminate potential (CHIP), analogous to monoclonal gammopathy of undetermined significance and monoclonal B-cell lymphocytosis, which are precursor states for hematologic neoplasms but are usually benign and do not progress. Because mutations are frequently observed in healthy older persons, detection of an MDS-associated somatic mutation in a cytopenic patient without other evidence of MDS may cause diagnostic uncertainty. Here we discuss the nature and prevalence of CHIP, distinction of this state from MDS, and current areas of uncertainty regarding diagnostic criteria for myeloid malignancies.

Background Clonal hematopoiesis of indeterminate potential (CHIP) is an age-related phenomenon characterized by the presence of somatic mutations in peripheral blood (PMIDs: 25426837, 25426838). Although CHIP was originally defined in healthy older adults without cytopenias, it can be found in other contexts as well. For example, one recent report described four patients with therapy-related myeloid neoplasm (TMN) arising after treatment for other cancers, all of which were driven by TP53 mutations that could be found at very low levels in samples drawn years before the development of TMN (PMID: 25487151). However, there has not yet been a more systematic study of CHIP in this type of context. In this study, we sought to understand how CHIP behaves and influences outcomes in the context of autologous stem cell transplantation (ASCT), arguably the most extreme selective pressure that can be studied in the context of native hematopoiesis. We hypothesized that in patients with Non-Hodgkin Lymphoma (NHL) undergoing ASCT, the presence of CHIP at the time of transplantation would be associated with an increased risk of TMN and other adverse outcomes. Methods We analyzed exome sequencing data from 10 patients with TMN after ASCT (City of Hope Cancer Center, Duarte, CA), and performed targeted sequencing of 116 genes on banked, mobilized peripheral blood from an additional 401 patients with NHL who underwent ASCT (Dana Farber Cancer Institute, Boston, MA), to determine whether there is a clonal connection between CHIP at the time of ASCT and subsequent TMN, and to determine whether the presence of CHIP at the time of ASCT influences subsequent outcomes. Results In 7 of 10 TMN patients for whom we analyzed exome sequencing data, mutations present at the time of TMN were also detectable in the pre-ASCT sample. PPM1D, a key mediator of the DNA damage pathway, was mutated in 2 patients, as was TP53 (2 patients), TET2 (2 patients) and PRPF8 (1 patient). In our larger cohort of 401 unselected ASCT patients, CHIP was common (121 patients, 30.2%) and was associated with older age but not with other demographic or treatment-related factors. PPM1D was the most commonly mutated gene (54 mutations in 48 patients). In the ASCT cohort of 401 patients, 18 patients developed TMN. The presence of CHIP at the time of ASCT significantly increased this risk: the 10-year cumulative incidence of TMN, with death and allogeneic transplant as competing risks, was 12.4% for patients with CHIP, compared to 3.5% for patients without CHIP (P=0.002, Figure 1A). Moreover, the presence of CHIP at the time of ASCT conferred significant risks beyond TMN alone, as patients with CHIP had significantly inferior overall survival compared to patients without CHIP (10-year OS 30.6% versus 60.9%, P=0.0003, Figure 1B). This difference was driven primarily by late mortality and not by an increased risk of relapse or by the difference in rate of TMN. Although other variables were associated with OS in univariate analysis, multivariate analysis in a Cox proportional hazards model showed that only older age (60 or above), aggressive lymphoma, and presence of CHIP were significantly associated with survival. Conclusion We show that CHIP at the time of ASCT for NHL is common and is associated with an increased risk of TMN and decreased overall survival independent of the TMN risk. These results have substantial clinical and translational implications. They suggest the need to specifically study the connection between CHIP and lymphoma more deeply, which could be accomplished by assessing CHIP in patients with newly diagnosed lymphoma prior to the administration of any chemotherapy or mobilizing agents. They also suggest the need to consider alternative therapeutic approaches for patients with lymphoma and a high risk of TMN who are being considered for ASCT. Finally, they underscore the need to study clonal hematopoiesis in the context of treatment for other cancers to determine whether these results may be relevant to an even larger number of patients. Disclosures Lindsley: MedImmune: Research Funding; Takeda Pharmaceuticals: Consultancy. Mar:H3 Biomedicine: Other: Spouse's employment. LaCasce:Forty Seven: Consultancy; Seattle Genetics: Consultancy; Seattle Genetics: Consultancy. Koreth:LLS: Research Funding; amgen inc: Consultancy; takeda pharmaceuticals: Membership on an entity's Board of Directors or advisory committees; kadmon corp: Membership on an entity's Board of Directors or advisory committees; prometheus labs inc: Research Funding; millennium pharmaceuticals: Research Funding. Ritz:Kiadis: Membership on an entity's Board of Directors or advisory committees. Soiffer:Kiadis: Membership on an entity's Board of Directors or advisory committees; Juno: Membership on an entity's Board of Directors or advisory committees; Jazz Pharmaceuticals: Membership on an entity's Board of Directors or advisory committees.

One of the adverse consequences of chemotherapy exposure is the development of therapy-related myeloid neoplasms (t-MNs). However, the cause and origin of most t-MNs are unknown, and the prognosis remains dismal. Novel work has shown that in addition to TP53, PPM1D is selectively mutated in 15% of therapy related MDS (Lindsley et al., ASH Abstract). Truncating mutations of PPM1D are also found to commonly occur in clonal hematopoiesis (Jaiswal et al., NEJM 2014; Genovese et al., NEJM 2014; Xie et al., Nat Med 2014), as well as in the blood of cancer patients, particularly after exposure to chemotherapy (Ruark et al., Nature 2013; Swisher et al., JAMA Oncol 2016). We hypothesized that PPM1D truncating mutations confer chemotherapy resistance, causing selective outgrowth of PPM1D mutant hematopoietic stem cells in states of genotoxic stress. In addition, since PPM1D mutations lead to a gain of function, we examined the potential of targeting PPM1D-mutant cells pharmacologically. The protein phosphatase PPM1Dis a direct regulator of TP53 activity and the DNA damage response pathway (Fiscella et al., PNAS 1997). Consequently, gain-of-function PPM1D mutations lead to decreased TP53 activity. To examine whether PPM1D mutations drive chemotherapy resistance through an abrogation of the TP53 dependent DNA damage response, we engineered PPM1D-mutant subclones using the CRISPR-Cas9 system in the TP53 wild-type AML cell line MOLM-13. PPM1D exon 6 truncation led to increased expression of PPM1D and resistance to DNA damaging agents, including cytarabine, cyclophosphamide and cisplatin. In addition, PPM1D-mutant cells exhibited a selective advantage over wild-type cells in the presence of chemotherapy, expanding 100-fold over a 24-day period (Fig. 1). While treatment with chemotherapy induced phosphorylation of Chk1 and p53, cell cycle arrest and apoptosis in wild-type cells, this response was abrogated in PPM1D-mutant cells due to gain of function of PPM1D. Using phosphoproteomic analysis, we further demonstrate decreased phosphorylation of known and novel targets in PPM1D-mutant compared to wild-type cells. We next investigated the effect of PPM1D mutation on normal marrow progenitors in response to chemotherapy treatment in vivo. We performed a competition experiment in which mouse bone marrow c-Kit+ cells expressing Cas9 were transduced with guide RNAs targeting exon 6 of PPM1D or a control guide, and then transplanted into mice in a 1:5 ratio. We observed a selective outgrowth of PPM1D-mutant myeloid cells in the peripheral blood of mice after exposure to chemotherapy. In addition, we found expansion of PPM1D-mutant cells in the lineage negative, c-Kit+ Sca-1+ population, which is enriched for hematopoietic stem cells and multipotent progenitors, indicating that PPM1D-mutant stem and progenitor cells have a competitive advantage over wild-type cells after exposure to genotoxic stress. The generation of a selective, allosteric inhibitor of PPM1D (Gilmartin et al., Nat Chem Biol 2014) allowed us to examine whether pharmacologic inhibition of PPM1D decreases the chemotherapy resistance or survival of PPM1D-mutant cells. We found that PPM1D-mutant cells have a significantly increased sensitivity to PPM1D inhibition when compared to wild-type controls. In addition, PPM1D inhibitor treatment was able to re-sensitize mutant cells to chemotherapy and abrogate the selective outgrowth of PPM1D-mutant cells during chemotherapy exposure. Lastly, we demonstrate that the proteome-wide phosphorylation profile characteristic of PPM1D-mutant cells can be reversed through treatment with the PPM1D inhibitor. In sum, these results demonstrate that PPM1D mutations confer a competitive advantage to hematopoietic stem cells undergoing genotoxic stress by abrogating the DNA damage response, and are likely to be the initiating mutation in a large proportion of t-MNs. Due to the gain-of-function nature of this mutation, PPM1D-mutant cells are differentially sensitive to treatment with a PPM1D inhibitor. PPM1D inhibition may therefore provide an opportunity for the prevention and targeted treatment of hematologic malignancies that harbor PPM1D mutations. Figure 1 PPM1D mutant cells have a competitive advantage under the selective pressure of chemotherapy (cytarabine) treatment. Figure 1. PPM1D mutant cells have a competitive advantage under the selective pressure of chemotherapy (cytarabine) treatment. Disclosures No relevant conflicts of interest to declare.

Hematological malignancies are associated with recurrent somatic mutations in specific genes, and may be preceded by a pre-malignant state in which only the initial driver mutation(s) are present. For example, monoclonal gammopathy of unknown significance often precedes multiple myeloma and monoclonal B-lymphocytosis can precede chronic lymphocytic leukemia. Recent sequencing studies have identified genes that are recurrently mutated in acute myeloid leukemia, myelodysplastic syndromes, myeloproliferative neoplasms, acute lymphoblastic leukemia, and other hematological neoplasms. We hypothesized that a pre-malignant state comprised of a clonal expansion of cells harboring some of these recurrent mutations would be detectable in the blood of elderly individuals not known to have hematological disorders. To address this question, we analyzed whole exome sequencing data from peripheral blood cell DNA of 17,182 individuals. Most of these were sequenced for type 2 diabetes genetic association studies and were therefore unselected for hematological phenotypes. We looked for candidate somatic variants by identifying previously characterized single nucleotide variants (SNVs) and small insertions/deletions (indels) in 160 genes recurrently mutated in hematological malignancies. The presence of these variants was analyzed for association to hematological phenotypes, survival, and cardiovascular events. We identified a total of 805 candidate somatic variants (hereafter referred to as mutations) from 746 individuals in 73 genes. Somatic mutations were rarely detected in individuals younger than 40, but rose appreciably with age (Figure 1). At ages 70-79, 80-89, and 90-108 these clonal mutations were observed in 9.6% (220 out of 2299), 11.7% (37 out of 317), and 18.4% (19 out of 103) of individuals, respectively. The majority of the variants occurred in 3 genes: DNMT3A (403 variants), TET2 (72 variants), and ASXL1 (62 variants). The median variant allele fraction for the detected somatic mutations was 0.09, from which we infer that the pathologic clone represents on average 18% of circulating white blood cells. Clinical outcome data was available on a subset of subjects, with a median follow-up period of 8 years. Carrying a somatic mutation was associated with increased risk of developing a hematological malignancy (hazard ratio [HR] 11, 95% confidence interval [95% CI] 3.9-33 by competing risks regression). Harboring a mutation was also associated with an increase in all-cause mortality that could not be explained by death due to hematological malignancies alone (HR 1.4, 95% CI 1.1-1.8 by Cox proportional hazards model). We further found that these mutations were associated with type 2 diabetes (odds ratio 1.3, 95% CI 1.1-1.5) and increased risk of incident coronary heart disease (HR 2.0, 95% CI 1.2-3.4) and ischemic stroke (HR 2.6, 95% CI 1.4-4.8) in multivariable regression models. We conclude that clonal hematopoiesis associated with a somatic mutation in a known cancer-causing gene is a common pre-malignant condition in the elderly. This entity is associated with increased risk of transformation to hematological malignancy, as well as increased all-cause mortality, possibly due to increased cardio-metabolic disease. While the link between somatic mutations and cancer is well established, the relationship between clonal hematopoiesis and cardio-metabolic disease requires further study. Figure 1 Figure Prevalence of somatic mutation by age. Colored bands represent 50, 75, and 95 percent confidence intervals. Figure. Prevalence of somatic mutation by age. Colored bands represent 50, 75, and 95 percent confidence intervals. Disclosures Getz: The Broad Institute, Inc.: PCT/US2013/057128​ (Detecting Variants in Sequencing Data and Benchmarking Methods and Apparatus for Analyzing and Quantifying DNA Alterations in Cancer)​ Patent pending Patents & Royalties; Appistry: Certain NGS analysis tools of Broad Institute are made available for commercial use via Appistry, Certain NGS analysis tools of Broad Institute are made available for commercial use via Appistry Other. Ebert:Genoptix: Consultancy, Patents & Royalties; Celgene: Consultancy, Research Funding.