Abstract
Whether individual results of genetic research studies ought to be disclosed to study participants has been debated in recent decades. Previously, the prevailing expert view discouraged the return of individual research results to participants because of the potential lack of analytic validity, questionable clinical validity and medical actionability, and questions about whether it is the role of research to provide participants with their data. With additional knowledge of participant perspectives and shifting views about the benefits of research and respect for participants, current expert consensus is moving toward support of returning such results. Significant ethical controversies remain, and there are many practical questions left to address, including appropriate procedures for returning results and the potential burden to clinicians when patients seek guidance about the clinical implications of research results. In this review, we describe current views regarding the return of genetic research results, including controversies and practical challenges, and consider the application of these issues to research on apolipoprotein L1 (APOL1), a gene recently associated with health disparities in kidney disease. Although this case is unique, it illustrates the complexities involved in returning results and highlights remaining questions.
- genetic renal disease
- research results
- ethics
- humans
- consensus
- apolipoprotein L1
- genetic research
- disclosure
- knowledge
- kidney
- diseases
- Kidney Genomics Series
Introduction
Advances in genome science offer important opportunities for nephrology. Genes associated with rare single-gene kidney diseases, such as Fabry disease and polycystic kidney disease, have been identified, and genetic analysis is being used to discover common gene variants contributing to CKD and ESKD. This research can elucidate disease pathways and may serve as the basis for therapeutic discovery; it may also produce genetic tests to individualize prevention and treatment (1,2). However, the pathway from genetic discovery to clinical benefit is long, complex, and uncertain (2). Gene-disease associations must be replicated, biologic mechanisms and interventions must be defined, and clinical trials must be undertaken to assess the utility of emerging tests or interventions. As scientific discovery proceeds, researchers need to consider whether genetic results should be shared with participants. The issue of returning individual results to research participants has been debated for several decades, with increasing support for returning at least some results, particularly when medically actionable (3). This shift creates a challenge for clinicians because patients are likely to seek their advice about how these new genetic findings derived from participation in a research study relate to their clinical care.
In this review, we describe current views regarding the return of individual genetic research results (Box 1), including controversies and practical challenges, and we consider the application of these issues to research on apolipoprotein L1 (APOL1), a gene recently associated with health disparities in kidney disease (Box 2).
Summary of key points in this review.
Summary of the APOL1 example.
Development of an Expert Consensus on Return of Individual Research Results
Ethicists offer several arguments in favor of returning research results (3), including that sharing results is a way of showing respect to research participants; that participants should have access to research results because they have a right to information collected about themselves in governmental or health records (4); and that, if research results offer valid, actionable health information, researchers have an obligation to provide the information to participants. Providing results may also incentivize research participation. Although the reasons vary, when participants are asked, most express a preference for receiving their results (5,6).
Arguments against providing individual results focus on the nature of research (3). Such results may be provisional or unreplicated, and offering this information to participants could lead to harm (7,8). In addition, because research is designed to produce generalizable knowledge, not individual benefit, it is argued that there is no obligation to provide results to participants (7,8). In this context, diversion of resources to the return of individual results could be construed as a misuse of research funds. Additionally, the notion that participants have a right to information about themselves is disputed (3,4,7,8).
Several advisory groups have considered the issue, and they have identified medical actionability as a key criterion for return. In 1999, the National Bioethics Advisory Commission proposed that individual research results could be disclosed if the results were scientifically valid and confirmed, had significant implications for the participant’s health, and could be addressed with a medical intervention (9). Subsequently, a working group convened by the National Heart Lung and Blood Institute recommended that a gene variant should be returned to a participant only if the risk associated was significant and analytically valid, and they emphasized that disease risk should be both serious and actionable: that is, that proven interventions are available (10). A follow-up working group added additional considerations: return should comply with relevant laws, participants could choose not to receive results, and the obligation to return results would not ordinarily extend beyond the study period (11). This working group noted that researchers could return additional results (e.g., results that are not medically actionable) as long as they met other considerations for return (11). Finally, the same group recommended that community input should be obtained regarding research with identifiable communities.
Other statements suggest an expert consensus favoring the offer of individual genetic results to participants when the results can be used to inform clinical care for serious health conditions, with the acknowledgments that participants may choose not to receive results and that researchers may choose to return additional results beyond those that are medically actionable (12–16). However, several judgments are involved in complying with these recommendations. Which conditions are deemed medically serious? How is “medical actionability” defined? What constitutes sufficient evidence of analytic validity? Furthermore, although there seems to be consensus favoring return of research results that are valued by participants, the return of results seems to be uncommon.
Although there are no formal guidelines for the return of genetic research results, the 2018 report from the National Academy of Sciences (NAS) on results generated in research laboratories provides the most definitive resource (3). This report noted that “strong justifications can be made for returning results in many circumstances beyond traditional and current practices” (ref. 3, p. 11). However, it also acknowledged that there may be reasons to limit return. The committee’s recommendations (Table 1) include that researchers routinely consider this issue as part of planning for research and ensure that any results returned to participants are analytically valid. The report also emphasized the importance of assessing and incorporating participant preferences and values in the decision-making process and returning results in a manner that ensures participant understanding. In determining which results should be returned from a given study, the report noted two key elements: (1) the value of the result to the participant and (2) the feasibility of return, acknowledging that genetic studies often involve large numbers of participants and that feasibility at that scale could be limited. Within this framework, they argued that researchers ought to return some results, such as urgent clinically actionable results, but should be discouraged from returning results with certain characteristics, such as results that cannot be interpreted at an individual level or are of low value to the participant while requiring significant resources to return (3). There is no legal obligation in the United States to return genetic research results to participants or any relevant case law (17). However, as expert groups make recommendations about the return of certain results (e.g., those that are medically actionable), they may contribute to a perceived standard of practice that could create liability for researchers who fail to return such results (17).
Summary of recommendations in National Academy of Sciences report (3) with application to APOL1
Although the NAS framework acknowledges the significance of medical actionability, it defines “value to participants” as the most relevant consideration, with the concomitant need to determine how participants value different kinds of research results through participant engagement and rigorous research. For example, participants may place a different value on results relevant to an inherited childhood kidney disorder, like congenital nephrotic syndrome, than on those for a common condition, like CKD, in adulthood.
Significant controversy remains regarding how the standard of analytic validity should be met. The Center for Medicare and Medicaid, the federal agency responsible for laboratory standards, has stated that research results can be returned only if they were performed in a Clinical Laboratory Improvement Amendments (CLIA)-compliant laboratory (https://www.cdc.gov/clia/law-regulations.html). Thus, if a research laboratory is not CLIA compliant, CLIA guidance would require the researcher to obtain a new sample from the participant without disclosing results and confirm the result in a CLIA-compliant laboratory before returning it. This requirement represents a substantial barrier to returning results in many cases (12,18,19). Others argue that this requirement creates an ethical dilemma when significant results are identified that have not been performed in a CLIA-compliant laboratory (19) and that it unfairly prohibits participants’ access to results that they may value (4). The NAS report noted the importance of ensuring that results are analytically valid before returning them; however, the report noted that this did not necessarily require a CLIA-compliant laboratory. Instead, they recommended the development of less burdensome quality standards for research laboratories, and in the interim, they proposed that institutional review boards use peer review to determine whether adequate quality standards are in place before results are returned (3). There remains substantial disagreement about the optimal approach to this issue (3,4,11,12,19).
The NAS report also noted a conflict between the CLIA requirement and mandatory disclosure requirements of the Health Insurance Portability and Accountability Act (HIPAA; https://www.hhs.gov/hipaa/for-professionals/index.html). HIPAA requires that all information in the designated record set, which is defined as information about a patient maintained by or for a HIPAA-covered entity, be disclosed on patient request. This requirement would apply to research laboratories that are not CLIA compliant if they are part of a HIPAA-covered entity (such as a university medical center). To resolve this potential conflict, the NAS report recommended that the designated record set be defined to include only individual results that meet quality standards for analytic validity and that steps to harmonize CLIA and HIPAA regulations be undertaken by the federal agencies involved (3).
There are a number of implications for clinicians in these debates about returning research results. To the extent that medical actionability is a criterion for returning results, clinical input is needed in setting policies. When research results are returned, clinicians need information resources to enable them to provide appropriate assistance when their patients seek advice about these research results. If a research laboratory returns results generated under an acceptable but non-CLIA quality standard (as endorsed by the NAS [3]), the clinician needs to ensure that the results are confirmed in a CLIA-compliant laboratory before they are used in clinical decision making.
Practical Challenges to Returning Results
Box 3 outlines considerations about return of results across the research timeline. These point to a number of practical challenges. Optimal approaches for returning individual genetic research results to participants have not been defined, including the resources needed, the feasibility of returning results from studies such as large epidemiologic studies (e.g., the Chronic Renal Insufficiency Cohort and Chronic Kidney Disease in Children), and perhaps most significantly, how to return results respectfully, responsibly, and effectively (Table 2). Many studies do not have personnel or other resources available to return results to participants. The initial consent process may not have included discussion of the issue, requiring new procedures to inform participants about this option and determine participant preferences. The NAS report recommends that the costs of returning results be included in research budgets (3), but this is a prospective solution; the lack of resources may be a significant barrier for existing studies.
Considerations for return of individual research results during the course of the research process. CLIA, Clinical Laboratory Improvement Amendments; IRB, institutional review board.
Practical challenges in returning individual genetic research results
Thus, although growing consensus suggests that individual genetic research results valued by participants ought to be returned when feasible, empirical research is needed about how best to accomplish this goal. Participants may seek advice from their clinicians when they receive research results, and therefore, the process must ensure that both participants and clinicians have adequate information to understand and make appropriate use of the results and that returning research results does not overly burden clinical practice (3).
APOL1 as an Instructive Example
To illustrate the issues involved in returning research results, including areas of ongoing controversy and uncertainty, we offer the example of the APOL1 gene and its variants associated with ESKD (Box 2) (20–29). This example entails significant complexity. There is an active research agenda regarding the role of this gene in kidney disease etiology. Its variants are causally associated with a serious health outcome, yet no specific measures are known to address this risk, and the associated health outcome is a well recognized health disparity disproportionately affecting blacks, a group that has reason to distrust research and medicine.
Having two copies of variants in the APOL1 gene (G1 and G2) is associated with a seven- to tenfold increased risk of developing ESKD versus having one or no copies (20). These variants are thought to have originated relatively recently, increasing in frequency under positive selection because they seem to be protective against trypanosomiasis (African sleeping sickness). They are found almost exclusively in people of West African descent. Among blacks, 13% carry the high-risk genotype (21). Only a minority of individuals with APOL1-associated risk will develop ESKD, however, and there are not yet clear data to characterize which individuals with the genotype will progress to ESKD. Furthermore, although APOL1 is associated with FSGS, HIV nephropathy, and hypertension-attributed ESKD (20,24), it is not associated with diabetic nephropathy, the etiology reported as the primary cause of CKD and ESKD (26).
Many questions about APOL1-associated risk remain unresolved (28,29). Interactions between social determinants of kidney disease risk (e.g., low socioeconomic status) and APOL1-associated risk are largely unstudied (22,23,28). Questions about BP management and cardiovascular disease risk have also been raised. One study found that strict BP control did not change APOL1 kidney outcomes, and there was no mortality difference by genotype (26); another found an association between strict BP control and reduced mortality in patients with APOL1-associated risk (25). Similarly, results vary on the association of APOL1 genotypes with cardiovascular risk, and mechanisms of APOL1-associated kidney injury remain undefined (27). These questions point to the importance of ongoing research about APOL1-associated risk (28,29).
Engagement with patients and community members indicates strong interest in receiving APOL1 results either from research or in the context of clinical care (5,30,31). Furthermore, providing the option to learn research results may strengthen relationships between researchers and minority populations (32). Our data suggest that researchers and clinicians support the return of APOL1 information to research participants but with substantial cautions about ensuring that there are resources to reduce confusion and promote comprehension (33). Practical considerations include ensuring that adequate counseling is readily available and accessible and if needed, the cost of confirming research results in a CLIA-compliant laboratory before they are returned. For APOL1 testing, the cost is $375 per sample (https://www.apol1genetest.com/place-an-order/).
Additional Considerations
The APOL1 example illustrates additional issues that arise when considering the return of genetic research results.
Genotype as a Factor in Study Recruitment
Using participants’ genotype in studies to characterize disease biology and develop interventions can be a powerful research approach, particularly when the genetic variant of interest is relatively rare (34). In genotype-driven recruitment, participant genotypes are used to determine eligibility for a study (e.g., a retrospective cohort study to assess ESKD risk relative to APOL1 genotype) or to assign participants to study arms (e.g., a clinical trial to assess the effectiveness of specific hypertension interventions on ESKD outcomes stratified by APOL1 genotype). This information may be derived by querying previously genotyped data in biobanks and existing databases or collected for the researchers’ previous study (35).
Key ethical challenges in genotype-driven recruitment include harms from learning unwanted or unanticipated information prior to recruitment and consent for the new investigation (i.e., a person’s right “not to know” [36]) and conversely, the potential to deceive and undermine informed consent if the genotype-based reason for the individual’s eligibility is not disclosed (35). If the purpose of the study is to characterize the phenotypic consequences of a specific genotype, communication must emphasize current uncertainties so that the participant does not mistake the genotype disclosure with indicating known clinical meaning or actionability. Even with such care, it is possible that a given participant will be concerned by receiving unexpected genetic information or learning this information from someone with whom he or she may not have a prior research relationship.
These issues have not yet been resolved (37), and appropriate procedures are likely to depend on the context of the study: whether the issue of recontact was addressed in a prior consent form; if participants were identified from a registry; how and when results are offered; how participants have interacted with the research (e.g., long term or as patients of the researcher/clinician); whether the research findings to be disclosed involve conditions that participants are already aware of or rather, provide new information about their risk or prognosis; whether study participants are vulnerable in various ways (i.e., minors or those who have cognitive, social, or economic vulnerability); the study design, such as involvement of family recruitment; and finally, the participants’ information needs and expectations.
Studies Using Genome/Exome Sequencing: Secondary Findings
Because sequencing of the whole genome and sequencing of the whole exome (the protein-coding regions of the genome) are increasingly used in genomic research, additional questions must be addressed about returning research results not specified as part of the primary inquiry. After a genome sequence is ascertained, the analyst selects which locations in the sequence to query. If the whole sequence is already in hand (contrasting with targeted genotyping methods that only ascertain a specified set of genes or genetic variants), it is possible to query genes beyond those required for the research study. Given that the data exist, which genes, if any, should be queried using whole-genome sequencing or whole-exome sequencing?
The American College of Medical Genetics and Genomics (ACMG) recommends a specified list of results to return in clinical sequencing, including pathogenic variants from 59 genes for 20 conditions, on the basis of a set of criteria, including severity and medical actionability. Fabry disease is one such condition affecting the kidney (https://www.ncbi.nlm.nih.gov/clinvar/docs/acmg/). Use of the ACMG list is voluntary, and the list only includes a small subset of all genes of potential interest to an individual patient (38,39).
Whether these or any other variants of interest should also be returned to individuals in the course of sequencing for research purposes is an open question. For example, in a whole-exome sequencing study of adults with nephropathy of unknown or familial etiology, beyond answering several questions regarding kidney disease etiology, researchers identified a pathogenic BRCA2 variant from the ACMG list that led to a new breast cancer diagnosis (40). This case demonstrates the potential benefit of returning secondary findings in a research context. However, a consensus statement from two research consortia funded by the National Human Genome Research Institute affirms that there is no obligation for researchers to look for secondary findings in research data (12).
As the APOL1 example illustrates, participants might value secondary results for reasons other than medical actionability (41) because APOL1 results nevertheless provide health-related information that may be of interest to patients and that may be relevant for individuals considering kidney donation (42). Participants may see APOL1 status as offering a fuller picture of their health, potentially motivating behavior change, or the offer of such information as showing respect for the participant’s contribution to research (5). Given these values, questions remain regarding how context dependent the reporting of secondary results should be; what role participant preferences should play, particularly when there is no clear change in care, and how these preferences would be ascertained in absence of standard findings to return (41).
Legacy of Harm and Distrust among Communities
Not disclosing research results may heighten distrust of research. In the APOL1 example, the specter of distrust animates many conversations (5). Notably, some may perceive not sharing information about APOL1 with blacks as discriminatory and a breach of trustworthy research. Although many examples exist, two highly influential cases may contribute to black community views: The Public Health Service Study of Syphilis at Tuskegee and the Henrietta Lacks case (43,44). The syphilis study involved withholding effective treatment to study the natural history of disease among disadvantaged black men in Alabama. The Henrietta Lacks case involved obtaining research samples from a disadvantaged black woman without knowledge or participant consent. In both cases, procedures for informed consent were lacking, and in the syphilis study, information germane to the harms of participation (that is, the development of effective treatment during the course of the study) was not disclosed. The distrust resulting from these and other examples of violations of current research ethics standards continues to influence willingness to participate in research among blacks and other under-represented communities (45,46).
Proactively offering information as a show of respect for participant autonomy and contribution to the research enterprise may help to rebuild trust in the research and medical systems among black communities (5). Furthermore, results of genetic research can have implications for families and communities through the shared nature of genetic risk and potential for group stigma and discrimination (47). Altogether, these concerns call for community input on ethical and procedural considerations for returning results, clear in the APOL1 example (5) but relevant beyond black communities as evidenced by research missteps and resulting harm affecting other communities (48). The return of individual genetic results in the context of research has received a great deal of attention in recent years. Because participant interest and ethical consensus supporting this disclosure are on the rise, however, more research is needed to understand how to return research results effectively and responsibly. An important component of this challenge is how to define and develop resources for clinicians in anticipation of patients seeking guidance on results of research participation and the feasibility and costs of doing so (49). Decision making about returning genetic research results depends on the institutional context, participant information needs, and what is known about the implications of the specific result. Research funding agencies have opportunities to support development of best practices, incorporating community viewpoints about return of genetic results. The development of best practices should incorporate the needs of clinicians whose patients may seek help interpreting results that they have received as a result of research participation.
Disclosures
Dr. Burke reports receiving a grant from the National Institutes of Health outside of the submitted work. All remaining authors have nothing to disclose.
Funding
This work was supported by National Institutes of Health, National Human Genome Research Institute grant R01HG007879 (to Dr. Young and Dr. Burke) and National Center for Advancing Translational Sciences Clinical and Translational Science Award UL1TR002243. Dr. Cavanaugh is supported by a grant from the U.S. Department of Defense award number W81XWH-18-1-0412, outside of the submitted work. Dr. Umeukeje is supported by National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases grant 1K23DK114566-01A1. Dr. Young is supported by grant U2CDK114886-01 from the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases and in part by funding from the Veterans Affairs Puget Sound Health Care System.
Acknowledgments
Opinions expressed in this manuscript are those of the authors and do not necessarily reflect the official views of the National Institutes of Health or the authors’ institutions.
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
- Copyright © 2020 by the American Society of Nephrology
References
In this issue
Jump to section
More in this TOC Section
Cited By...
- No citing articles found.