Five decades ago, the analysis of metaphase chromosomes in the hematologic malignancies provided our first broad glimpse into the genetic anatomy of a malignant cell. Today, the advent of high-throughput methods such as next-generation sequencing, capable of surveying the entire genome, provides an unprecedented opportunity to develop an integrated molecular profile of each patient’s tumor, based on the chromosomal pattern, gene/miRNA/lncRNA expression, DNA methylation/epigenomic pattern, and gene mutation status, promising a revolutionary change in the way we diagnose, characterize, and treat cancer.
Although work in the hematologic malignancies has been at the vanguard of efforts to elucidate the molecular pathogenesis of cancer, there are a number of unanswered questions. For example, we do not yet know the full spectrum of genetic changes, the order in which these mutations occur (ie, which mutations are initiating events and which are typically cooperating events), or the prognostic significance associated with many cooperating mutations. Recent studies, including those by Genovese et al1 and Jaiswal et al2—reviewed in this issue of The ASCO Post—have provided new insights into these questions, tying together seemingly disparate observations, such as the phenomenon of clonal hematopoiesis and the increased incidence of hematologic malignancies in the elderly.
Mature blood cells are generated from hematopoietic stem/progenitor cells, which sustain hematopoiesis for decades by virtue of their self-renewal properties. The fact that hematopoietic stem/progenitor cells could undergo clonal expansion—a term called “clonal hematopoiesis”—was first suggested by the observation of clonal remissions in patients with acute myeloid leukemia (AML), based initially on the detection of a skewed pattern of the X-linked glucose-6-phosphate dehydrogenase enzyme variants in female patients and later with X-linked DNA polymorphisms.3
Several decades later, the detection of the translocation-specific fusion transcripts, such as the RUNX1/RUNX1T1 fusion from the t(8;21), in bone marrow cells from AML patients with durable remissions suggested that this clonal hematopoiesis resulted from the expansion of cells that harbored an initiating, driver mutation. These observations dovetail nicely with later findings using genomic studies, which revealed that, in some patients, stem cells carrying a subset of the mutations present in the cancer cells are able to survive chemotherapy, acquire novel mutations, and, ultimately, trigger a relapse.4 Parallel observations of the detection of leukemia/lymphoma-associated chromosomal translocations, such as the t(14;18)/IGH-BCL2 fusion, in blood samples of healthy individuals suggested that cancer-associated early mutations (1) may occur in hematopoietic stem/progenitor cells in the absence of overt disease; (2) may be present many years before disease develops; and (3) are required, but insufficient, to result in disease.
In 2012, several reports suggested that clonal hematopoiesis resulting from an expansion of cells that have an initiating mutation might be a feature of the aging hematopoietic system.4-6 Busque et al identified TET2 mutations in neutrophils of healthy elderly females with clonal hematopoiesis as defined by altered ratios of X inactivation in peripheral blood.4
By using high-throughput sequencing of the genomes of AMLs, as well as the exomes of hematopoietic stem/progenitor cells from healthy individuals, Welch et al concluded that most of the mutations found in AML genomes actually occurred in hematopoietic stem/progenitor cells before they acquired the leukemia-initiating mutation.5 The investigators hypothesized that some genetic mutations confer advantages to affected hematopoietic stem/progenitor cells, resulting in enhanced self-renewal and clonal expansion.
To address the issue of whether such mutations affect additional genes involved in leukemia and lymphoma, they analyzed 2,728 blood samples, sequenced to a relatively high depth of coverage (average of 107.5× coverage) as germline controls for a variety of solid tumors analyzed by The Cancer Genome Atlas (TCGA).6 Surprisingly, this analysis confirmed that the overall frequency of blood-specific mutations increased with age (ranging from 1.2% to 6.8% for subjects in their 40s vs 80s) and revealed that blood cells of more than 2% of all individuals and 5% to 6% of people older than 70 years contain mutations that may represent premalignant events (eg, recurrent mutations in leukemia/lymphoma-associated genes that cause clonal hematopoiesis, most commonly DNMT3A, TET2, JAK2, ASXL1, and TP53).
The studies by Genovese et al and Jaiswal et al approached the issue of clonal hematopoiesis in an unbiased fashion by using whole-exome sequencing data from DNA of peripheral blood cells of large cohorts of individuals who were unselected for cancer or hematologic phenotypes.1,2 Genovese et al1 used data from 12,380 individuals from Swedish national patient registries, whereas Jaiswal et al2 analyzed 17,182 individuals from cohorts (cases and controls) in type 2 diabetes association studies or the Jackson Heart Study.
Key findings of these studies follow. Clonal hematopoiesis, as evidenced by detectable somatic mutations, is rare in individuals younger than age 40 but increases appreciably with each decade of life, reaching approximately 10% in persons older than 65 years and approximately 18% in persons 90 years of age or older. Moreover, mutations persist over time. Intriguingly, the incidence of hematologic malignancies mirrors this pattern. The majority of clonal mutations occurred in three genes, DNMT3A, ASXL1, and TET2, recapitulating the types of mutations in these genes, which have been detected as somatic mutations in myeloid neoplasms.
The most common change was a base-pair change, namely a cytosine-to-thymine transition, resulting from the deamination of a cytosine base, considered to be a somatic mutational signature of aging.5,7 The presence of somatic mutations was associated with an increased risk of hematologic malignancies (hazard ratio [HR] = 11.1, 95% confidence interval [CI] = 3.9–32.62; HR = 12.9, 95% CI = 5.8–28.71). Notably, a higher proportion of mutant leukocytes (a variant allele frequency ≥ 0.10, which is equivalent to ≥ 20% of cells) conferred a higher risk of developing a hematologic malignancy. In two cases, DNA-sequencing analysis of bone marrow cells at the time of diagnosis of AML confirmed that the leukemia was derived from the earlier mutant clones.2 Somatic mutations were also associated with an increase in all-cause mortality1,2 as well as an increased risk of incident coronary heart disease and ischemic stroke.2
The estimates of mutation frequency derived from the three studies1,2,6 summarized here are very conservative, since the sequencing depth of coverage limited mutation detection to the level of a 2% to 3% variant allele frequency (≥ 4%–6% of cells in a population); ultimately, greater than 1,000× sequencing depth or single-cell sequencing may be required to determine the actual incidence. Moreover, other types of alterations—eg, gene fusions, chromosomal abnormalities, and copy-number changes—were not included in the analyses.
Ramifications and Next Steps
These pivotal studies provide new insights into the multistep process of leukemogenesis, as well as risk factors for increased mortality and cardiometabolic disease in the elderly.
The findings support emerging data that mutations in genes such as DNMT3A, JAK2, ASXL1, TET2, TP53, and others are initiating events for myelodysplastic syndrome, AML, myeloproliferative neoplasms, and chronic myelogenous leukemia, whereas other changes characteristic of these diseases are likely to occur later as cooperating mutations (eg, IDH1, RUNX1, NRAS, NPM1, or FLT3 mutation), which are important for disease progression, a distinction that is relevant to the current dogma of therapeutic targeting of the primary changes.
In this current era of next-generation sequencing, caution is warranted when using blood as a surrogate reference for the germline genome, particularly in elderly individuals. Care must also be taken when interpreting leukemia-associated mutations detected in peripheral blood as an incidental finding: Although they signify an elevated risk, they do not support a diagnosis of hematologic malignancy without other corroborating evidence.
The incidence of myeloid neoplasms in the elderly is less than 0.1%, yet approximately 6% have leukemia-associated mutations in blood cells. The fact that most individuals with evidence of clonal hematopoiesis do not progress to overt disease makes it premature to sequence DNA from blood samples to enable early detection of hematologic malignancies.
In the future, expanded studies are needed to identify those individuals who are at the highest risk—including ascertaining the risk associated with specific mutated genes or combinations of genes, particularly in cell type-specific contexts—as well as to identify the effect of coexisting germline cancer predisposition mutations and the influence of lifestyle factors. Ultimately, single-cell sequencing may be required to identify high-risk combinations of genes co-occurring in the same cells. Although we currently lack interventions that are suitable for large groups of patients who may have only a low or modest risk of cancer, in the future, the use of sequencing of DNA from peripheral blood to identify high-risk persons or to monitor clonal expansions could facilitate clinical trials to reduce the risk of progression to hematologic malignancies or other cancers. ■
Disclosure: Dr. Le Beau reported no potential conflicts of interest.
1. Genovese G, Kähler AK, Handsaker RE, et al: Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med 371:2477-2487, 2014.
2. Jaiswal S, Fontanillas P, Flannick J, et al: Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med 371:2488-2498, 2014.
3. Jacobson RJ, Temple MJ, Singer JW, et al: A clonal complete remission in a patient with acute nonlymphocytic leukemia originating in a multipotent stem cell. N Engl J Med 310:1513-1517, 1984.
4. Busque L, Patel JP, Figueroa ME, et al: Recurrent somatic TET2 mutations in normal elderly individuals with clonal hematopoiesis. Nat Genet 44:1179-1181, 2012.
5. Welch JS, Ley TJ, Link DC, et al: The origin and evolution of mutations in acute myeloid leukemia. Cell 150:264-278, 2012.
6. Xie M, Lu C, Wang J, et al: Age-related mutations associated with clonal hematopoietic expansion and malignancies. Nat Med 20:1472-1478, 2014.
7. Alexandrov LB, Nik-Zainal S, Wedge DC, et al: Signatures of mutational processes in human cancer. Nature 500:415-421, 2013.
Dr. Le Beau is Arthur and Marian Edelstein Professor of Medicine in the Section of Hematology/Oncology and Director of the Comprehensive Cancer Center at the University of Chicago.
In two studies recently reported in The New England Journal of Medicine,1,2 whole-exome sequencing of DNA from peripheral blood cells of individuals unselected for hematologic phenotype showed that clonal hematopoiesis with somatic mutations is increasingly common with increasing age and is...