The past decade has witnessed the cataloging of genetic mutations in cancer genomes, providing new insights into how and in what ways cancer can develop and spread (1, 2). The focus has been on defining specific “driver” mutations, genetic errors in cancer cells that reveal basic biological processes gone awry that are required for cancer initiation and progression. These drivers are the target of new therapies—this concept is central to precision oncology efforts to treat patients according to the genetic changes that are present in their tumors (3). Along the way, it has also become apparent that cancer genomes harbor many additional “passenger” mutations (4). Patterns of driver and passenger DNA mutations derived from cancer genomes have provided clues about the different ways that cancer can manifest as a disease of genetic mutations (5, 6). In some circumstances, they can be linked to strong environmental carcinogens (for example, mutation patterns caused by tobacco smoke, ultraviolet radiation, or the fungal toxin aflatoxin) (7). Moreover, these forensic mutational patterns can be used to estimate how long it has taken for a tumor to develop (5). On page 911 of this issue, Martincorena et al. (8) turned their attention toward the detection of mutations in normal tissue, addressing a long-standing paradox that mutations arise in normal tissues but do not necessarily lead to cancer.
Martincorena et al. found that a larger-than-anticipated fraction of cells of the esophagus harbor mutations in cancer driver genes, which increases with age. Although only nine individuals without evidence of cancer were studied, the results raise two important questions. What is the role of mutations in tumorigenesis, and what factors are required to progress to cancer, including additional mutations or local events in the microenvironment or immune system? The findings also reveal that genome integrity deteriorates with age and perhaps in tissue-specific ways, which could reflect distinct external exposures and tissue-specific properties.
The authors compared the genome sequences of nondiseased esophagus and skin (9) to mutations in 74 known cancer driver genes. The total number of mutations was higher in the skin, which is not unexpected because the outer layers of skin are continually exposed to a strong mutagen, ultraviolet light. However, it came as a surprise that esophageal epithelial cells have a higher number of driver mutations, particularly mutations in the NOTCH1 and TP53 genes, which often occur in esophageal squamous carcinoma (10).
It is intriguing to speculate how such driver cancer mutations might be induced in the esophagus. The authors propose that the accumulation of mutations increases with age, mainly owing to errors in intrinsic processes such as aberrant DNA replication or repair. It is also plausible that the complex mix of what we swallow daily could contribute to the high mutation burden. The cumulative lifetime probability of developing esophageal squamous carcinoma in the United States is about 1 in 125, which is substantially smaller than that which would be predicted by the occurrence of the observed NOTCH1 and TP53 mutations. This raises the question of what other events are needed to induce cancer. Alternatively, perhaps there are mechanisms that prevent cancer from developing.
These findings suggest that the transition from healthy tissue to cancer may be more complicated than we have modeled (see the figure). A next step will be to survey the entire genome (rather than only selected driver genes), where we may find mutations in new genes as well as the elusive “dark matter” between genes. It is important to focus on this intergenic space to understand how regulation of the genome is disrupted through mutations or chemical modification, a hallmark of epigenetics. Larger comparative studies are also needed to more accurately estimate the actual distribution of mutational events in other tissues, particularly those with high cancer rates, such as the lung, and tissues with very low cancer rates, such as the heart. A long-term goal should be to generate a genetic atlas of mutations in tissues of healthy individuals, sampled across the age spectrum and from distinct ancestries. Armed with a more comprehensive catalog of aging-associated mutations, we should be in a better position to understand the trigger events for cancer.
Genetic mutations accumulate in normal esophageal epithelium over time—often this does not initiate cancer. The initial mutations can be generated by intrinsic processes as well as induction by environmental factors (e.g., alcohol, tobacco). The progression to esophageal squamous carcinoma requires additional factors, which could include environmental triggers, additional mutations, ineffective immune surveillance and local changes to the microenvironment.
GRAPHIC: VERONICA FALCONIERI/SCIENCE
Although more comprehensive cataloging of mutations will inform our understanding of the genetic events that underlie cancer development, it is also necessary to investigate the suppression of tumor development through epigenetic changes as well as the influences of the host immune system and the local tissue microenvironment. Investment in prospective studies, which follow individuals sampled serially over time, will enable researchers to understand the timing and dynamics of transformation to cancer. Additionally, it could be interesting to prospectively study individuals in various locations because there are substantial differences in the rates and regional risk factors for esophageal squamous carcinoma and other cancer types. The potential implications of such future work are enormous because this study suggests that it might be more daunting than anticipated to develop genetic tests as a diagnostic tool for cancer.
For some time, there have been hints that the genome deteriorates with age, including telomere shortening at the ends of chromosomes (11), and the detection of subpopulations of blood cells that harbor cancer-associated mutations, known as clonal mosaicism. These observations have arisen from the analyses of circulating blood cells drawn from tens of thousands of apparently healthy individuals (12). The most common large event involving an entire chromosome, mosaic loss of the Y chromosome in blood cells, has been shown to increase with age and tobacco use, prompting some to speculate its use as a biomarker for chronic diseases associated with age (13). Similarly, point mutations in driver genes that are particularly important in hematologic cancers occur more frequently with age (14). This is known as clonal hematopoiesis and can be associated with increased risk for cardiovascular disease or a hematological cancer (14, 15).
We should not restrict the application of the study of Martincorena et al. to cancer. Imagine that the generation of mutations in ostensibly healthy cells could be important in the pathogenesis of other chronic diseases of aging, such as diabetes, heart disease, and neurodegenerative disorders. If we can comprehend the basic mechanisms that underlie differences in mutational rates by tissue sites, together with how mutated cells are held in check, we could be a step closer to slowing down the destabilization of the genome with age.
원문: 여기를 클릭하세요~