What do the Changes in the Aging Genome Mean for Pharmacogenomics?

Keywords::

• aging genome
• microsatellite
• nextgen sequencing
• personalized genomics
• single nucleotide polymorphism

Pharmacogenomics is the study and application of an individual’s genome to make therapeutic selections, to optimize the application of drugs in therapy and to direct the development of new drugs. The decreasing cost of sequencing and the consequent increased availability of personal genomes have been hailed as the ushers of a new world of personalized medicine [1,2].

However, there is early evidence that an individual’s genomes may evolve and change over a lifetime [3]. There is also early evidence that an individual’s genome may vary in different cells or parts of the body, not only between diseased and healthy tissue, but between healthy organs [4–6]. The accuracy and utility of genetic testing presumes a certain stability of the genome. What are the implications for personalized medicine if a whole-genome sequence is merely a snapshot of time and place?

Until recently, the focus of genomics has been on population scale methods at one locus driven by linkage information [5,7]. The analysis of many genomes representing affected and unaffected individuals has made it possible to identify new causative and/or actionable genomic variations [8,9]. To ensure that such variant discoveries are indeed indicative of their contribution to a given trait, disease or response, it is still important to best match the affected and unaffected group members, thus removing known and hidden biases – ethnicity, sex and environmental localization. One possible hidden variable, or at least one not yet recognized as important, is the age of a genome.

A recent report highlights the potential impact of aging on an individual exome [3]. This study analyzed the DNA of three individuals drawn over time spans of 9–16 years. The analysis demonstrated that the genomes in living and aging tissues do acquire variants with the passage of time at different rates. In these samples, the variation rate spans at least one order of magnitude, from 9.6 × 10^-7 to 8.4 × 10^-6 bp^-1 year^-1 for nonsynonymous single nucleotides variants and 2.0 × 10^-4 to 1.0 × 10^-3 locus^-1 year^-1 for microsatellite loci. To put these numbers in context, the accumulated variants are about one thousandth of a percent per year. Therefore, while the variants may have potentially deleterious effects, the baseline genome we were born with essentially captures the genome. These observations suggest certain interpretations: genetic information may only represent an individual at a specific time of his/her lifespan; an individual’s genome may be dynamic with respect to age, and may have a unique and specific variation rate, which itself may be diagnostic; microsatellite loci, relative to SNPs, are about ten-times more susceptible to age-induced variation, which may also be an opportunity; genomic information identified in a single measurement cannot be assumed to be unvarying; and the variation accumulation rate is sufficiently small to suggest that risk-based diagnostic test results are informative throughout a lifetime, but to be absolutely precise, the genome may have to be occasionally remeasured. These early results suggest that the existing paradigm of a static genome in healthy individuals may not be entirely accurate and that age may be an important additional variable in the analysis of the human genome. Our capabilities to interpret singular genomes will determine the effectiveness of personalized genomic medicine and this approach establishes a novel paradigm, ‘multiple genomes per individual’. The existence of multiple dynamic genomes per person – mosaicism – creates complications, as well as opportunities.

Scientific & clinical implications

Since the successful sequencing of the human genome a decade ago, slow progress has been made in identifying the genetic causes of known heritable disease. Individual gene mutations and SNPs have provided only a fraction of the expected answers. As scientists move toward evaluating whole genomes across healthy and diseased populations and identifying informative patterns, which distinguish the populations, more answers will emerge. An awareness of the potentially dynamic nature of the genome is another factor to input in determining the utility and reliability of a genetic signature for risk of disease.

Clinicians are learning of the enormous potential of personal genomics for improving patient care and outcomes. Identifying individuals at increased risk for disease provides an enormous benefit in focusing resources on monitoring most closely those most likely to suffer from heritable diseases.

The best diagnosis and treatment guidelines today are based on the positive statistical results of clinical trials [1]. However, 10–40% of patients in these studies do not respond appropriately to the therapy. Personalized genomics presents us with new opportunities to recognize these individuals and treat them differently from those who respond as desired [1,2,10,11]. If the genome is dynamic, then attention must be paid to confirm that the informative genetic signature remains accurate at the relevant diagnostic moment. It is highly likely, however, that clinically relevant genomic signatures will be made up of very minute portions of the genome. With the decreasing cost of sequencing and the increasing sophistication of target enrichment kits, such confirmatory diagnostics will likely be able to be conducted quickly and inexpensively.

Big data, cost & economic challenges

There are technological considerations as well, for the growth of genomic information is accelerating, growing at a pace greater than computing capabilities, especially data storage. This further exacerbates the ‘big data’ problem in both population-scale genomics and personal genomics, not only in research, but also in the amount of data that we, as patients, will have to carry around with us [12]. Only time will tell how many actionable genomic differences our genomes contain and how frequently we will have to refresh our knowledge of them, as well how practical and useful it turns out to be in the clinic. This additional genomic data dimension may add to the overall cost of personal genomics, not only in the amount of data, but also in the significant computational resources needed to consider the clinical implications of such data, and the time clinicians must devote to decision-making for each patient.

Ethical, legal & social implications

There exists a considerable reluctance among some government regulators and clinicians to provide patients with detailed information about their genome for fear that the information will be misused, misunderstood and lead to misguided decisions. As a consequence, there is a concern that further discoveries regarding the complexity and nuances of the genome – such as changes with aging or nonconcordance across organs – will only bolster this new paternalism and further delay the availability of patients’ access to genomic diagnostics. However, available data support the conclusion that most patients wish to be informed about their health risks and will obtain appropriate medical advice before making clinical decisions [13].

Genomic risk diagnostics potentially provide valuable information to patients that empower them to make medical and lifestyle decisions that address their potential risk for future illness. The fact that a genetic profile may represent a snapshot in time and may need to be updated in the future puts it on the same footing as nongenetic risk diagnostics. The fact that the genome may be changing in small ways over the course of an individual life from the impact of environmental stressors and aging enhances, rather than reduces, the value of such testing.

The lesson to be extracted from these early findings regarding the potential impact of aging is a call for further research and further refinement of available diagnostics to enhance patient understanding and increase the power and sensitivity of diagnostic information.

The future

To understand and exploit our individual internal genomic diversity, it must first be characterized. Akin to the 1000 Genomes Project, which measured the genome from the blood of many individuals, a program to measure our genome in two additional dimensions should be conducted: the longitudinal measurement of many individual’s genomes throughout their lives, from cells collected at birth and at frequent times throughout one’s life; and the spatial measurement of individuals’ genomes throughout their bodies, collected from the many organ sites from healthy individuals. Rather than waiting a lifetime to gather longitudinal samples, a sufficient number of tissue/blood samples may currently exist that would permit the study. Also, a cross-organ study could be conducted using a plurality of samples from all tissue types from an an individual that perished from traumatic injury or disease. These datasets would establish the basis for exploration of the amount of genomic diversity within an individual, whether there are temporal or spatial hotspots for enriched variation, and how this internal variation differs across many individuals. This baseline knowledge would provide opportunities for the pharmacogenomics researcher to ensure they are using the proper and well-understood reference, and for clinicians who will increasingly depend on personal genomics for treatment selection and optimization to order the correct laboratory tests for a given individual and their disease site.

Having now recognized that genomes are not always temporally or spatially consistent, the measurement and exploitation of our various personal genomes may help us accelerate our understanding, explain previous inconsistencies and defeat disease, while resolving emerging scientific conflicts.

Financial & competing interests disclosure

MB Waitzkin is chief executive officer and HR Garner is co-owner and founder of Genomeon, LLC, which has licensed these aging genome findings. Genomeon did not fund this study or influence or participate in any way. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.

Additional information

Funding

MB Waitzkin is chief executive officer and HR Garner is co-owner and founder of Genomeon, LLC, which has licensed these aging genome findings. Genomeon did not fund this study or influence or participate in any way. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed. No writing assistance was utilized in the production of this manuscript.