100k Genome Project: sequencing and much more

Advances in genome-based technologies have led to the rapid generation of new data regarding the genetic etiology of many traits and conditions. In particular, new and cheaper sequencing technologies have enabled the use of whole-genome sequencing (WGS) and whole-exome sequencing (WES) as oft-used research tools, and increasingly as clinical tests for hard-to-diagnose cases. Thus, it is anticipated that WGS/WES may eclipse single- and multi-gene tests used today for disease diagnosis and treatment, and potentially be used for health screening. In December 2012, British Prime Minister David Cameron announced an ambitious initiative to sequence the genomes of 100,000 Britons (100,000 Genome Project or 100kGP) affected with cancer or rare diseases, budgeting UK£100 million for this extraordinary effort [101]. In July 2013, the Department of Health announced that this initiative will be coordinated by the newly-established, government-owned company Genomics England (London, UK). This project will undoubtedly result in new discoveries and demonstrate whether use of WGS can result in improved health outcomes.

Clinical implementation of WGS/WES in a clinical setting poses a formidable barrier. Although it may be prudent to wait until evidence of clinical utility has been gathered, given that clinical implementation studies are relatively less costly and can be done concurrently with clinical effectiveness studies, the 100kGP provides a golden opportunity to inform the delivery of WGS if and when the technology is demonstrated beneficial. In particular, there are two major areas of clinical implementation research that could be conducted concurrently with these 100,000 samples of sequenced Britons and vastly contribute to developing strategies for the effective and safe use of WGS: communication and behavioral impact. While decades of clinical genetics research and clinical experience will help inform the clinical delivery of WGS, extrapolating these findings and practices may not be as straightforward, given the scope of information that can be generated from WGS and, thus, warrant new studies specific to WGS [1].

Communication

Effective patient communication remains a challenging but critical part of medical practice. Successful clinical implementation of WGS must address the challenges of patient communication given the breadth and value of information it can generate, notably how to appropriately obtain consent from patients and communicate results. Regarding informed consent, the Public Health Genomics Foundation (Cambridge, UK) has called for development of a strong informed consent policy to ensure that individuals understand the purpose, risks and benefits of the project [102]. The Genomics England Ethics Working Group has suggested that consent be obtained in the context of clinical care, but have also recommended discussion of several issues related to the research context (e.g., data sharing) [103].

Developing appropriate and effective consent and communication approaches for WGS will depend in a large part on what type of information is being offered or disclosed, the intended purpose of offering or disclosing the information, and the patient. A major issue currently preoccupying many researchers, health practitioners and organizations is what, if any, other information (often referred to as incidental information) derived from WGS should be either offered or communicated to patients [2,3]. For the 100kGP project, the Public Health Genomics Foundation has called for the return of incidental findings about diseases with known preventive or therapeutic interventions [103]. The desire to protect participants from harm such as unnecessary follow-up care or anxiety caused by returning results with little clinical utility have been raised as factors in decisions not to return results, whether or not they are related to the patients’ current health conditions. However, studies assessing psychological risk associated with learning of a diagnosis or risk for an inherited condition, both in the clinical and research settings, have reported few adverse responses. For example, even after learning one’s risk for Alzheimer’s disease, an untreatable condition, remarkably few adverse effects were reported [4]. However, it is possible that learning of multiple disease risks uncovered in WGS may lead to harm if individuals are not able to emotionally manage the scope of information, struggle to prioritize how to respond to the information, and/or feel a sense of futility or hopelessness.

Ironically, the debate about what results to return has occurred primarily in the absence of the public and patients [5]. The Wellcome Trust Sanger Institute has commendably taken such steps to gather public opinion through their survey GenomEthics [104]. In conflict with some groups’ positions of disclosing only information about diseases with proven interventions, several studies have reported that research participants and patients wish to be involved in decisions about what, if any, results they would like to learn (or not learn) of [6,7]. In particular, the threshold of clinical utility may not be the only litmus test that patients consider in deciding whether they wish to learn of their genomic risk for a given condition(s) [8].

The 100kGP population will likely provide a rich and diverse patient population to not only better understand the type of incidental findings of interest to patients, but also what types of process and decision-making support one needs to promote informed consent and comprehension of results. Different approaches could be evaluated, such as whether to seek consent for every disease risk that may be revealed or use a blanket or tiered consent. Furthermore, the 100kGP can enable study of how best to communicate genomic information so as not to overwhelm patients, potentially develop a type of ‘genomic triage’ system, provide support services for patients (and their family members) to optimize comprehension and risk management, and assess short- and long-term psychological impact. Since communication is not limited to verbal communication, different combinations/approaches for decision-making support, delivery of risk information and educational tools (e.g., print, web-based tools and video/audio) may be needed for different testing purposes, and can be evaluated and compared among participants to identify the most effective strategies. For example, the delivery of risk information may benefit from a multimode approach to maximize comprehension of not only the test result, but also of the recommended intervention, if any. This multimode approach may include inperson consultation with a health provider, a printed summary of results and access to a secure database that enables review of results and links to resources about the disease and interventions. If patients do not clearly understand their genomic risks or the options available to them to ameliorate their risk, the potential benefit of WGS may not be realized.

For providers, effective laboratory reports and clinical decision support for WGS will be needed to assist them in interpreting and applying results appropriately. Perhaps the primary results (for which the test was ordered for) can be communicated to the provider via a standard laboratory test report, but incidental findings may be made available through a secure database, enabling the provider to review the results meeting certain criteria set by the provider and/or patient. Since one of the goals for the project is to support training and development of health-system infrastructures for data storage, the 100kGP provides an ideal opportunity to work with clinical testing laboratories to explore the development and use of a WGS patient results database and other tools to help health providers integrate genome information into clinical care.

Behavior

Much can be done by patients and providers to reduce one’s overall disease risk through lifestyle changes or preventive intervention and screening. Thus far, the data have been equivocal regarding the impact of genetic information on behavior and of somewhat poor quality [9,10]. Indeed, the communication of genomic risks without information about recommended clinical follow-up or lifestyle interventions may partly explain the absence of observed behavior change, as evidenced by those individuals who learned of their genomic risk through direct-to-consumer testing companies [9,11]. The delivery of personalized risks combined with generic health recommendations may not be as impactful as a personalized follow-up plan to reduce disease risks [12]. The 100kGP can provide a robust population to identify the barriers to behavior change, whether they are specific to genetics and genomics compared with a control population not enrolled in the study, and how genomic risk influences risk perceptions and self-efficacy to adopt healthy behaviors to reduce risk. The 100kGP will further enable development and assessment of personalized plans compared with generic recommendations for affecting behavior change. However, given that participants in the 100kGP will already be affected with a condition, the types of behaviors to improve health outcomes may be limited without involvement of a health provider. Thus, some consideration should be given to enrolling healthy individuals, as well assessing the impact of genomic risk on behavior changes to prevent or reduce risk of specific diseases.

Behavior changes are not limited to patients, but extend to providers in how they choose to manage patients’ genomic risk (be it increased or decreased) and recommend steps to reduce overall disease risk. Studies of provider behavior regarding family history-taking and follow-up have shown differences in knowledge and practice [13,14]. Providers may be uncertain as to how to use genomic information, such as that related to drug response (pharmacogenetics) with respect to dose adjustment or drug selection. Some have speculated that knowledge of genomic risks will unnecessarily increase demand for follow-up care to confirm test results or reduce anxiety of those found to have an increased genomic risk [15,16], although no such impact has been documented [17]. Individuals who choose to share their genomic risk results ordered through direct-to-consumer testing companies with providers appear more likely to engage in healthy behaviors, potentially due to learning about interventions to reduce disease risk [17]. As WGS becomes more widely used, it will not be feasible for genetics professionals to be involved in delivery of every WGS test and so it is important that nongenetic medical specialties are prepared to integrate this new source of risk information into patient care.

Conclusion

It is anticipated that the 100kGP will position the UK as a leader in genomic medicine by demonstrating the clinical utility of genomic information. WGS presents an exciting new clinical tool, but also a somewhat challenging application to integrate into clinical care with its excessive generation of disease risk information. In the era of evidence-based medicine, it is equally important to develop a clinical delivery model informed by empirical research, and the 100kGP appears to miss an opportunity to generate some of these data that are so critical to the clinical implementation of WGS. Thus, as patient recruitment and sequencing commence for the 100kGP, studies addressing clinical implementation issues should be codeveloped alongside clinical effectiveness studies. Clinical implementation of WGS will also have significant implications for testing laboratories and health informatics, and, thus, collaborating with these and other groups will be essential to providing a comprehensive and effective delivery system to promote patient and provider comprehension, and optimize likelihood for improved health outcomes. While delivery of WGS will evolve as both patients and health professionals become familiar with the benefits and risks of this innovative tool, in the interim, it is likely that use of WGS will require extensive communication and education for both patients and providers. Given recent calls for sequencing of all Britons (healthy and ill) [18], time is of the essence to inform the clinical delivery of WGS to minimize potential harms and maximize benefits. The 100kGP provides a golden opportunity to address many of these issues in a practical clinical setting.