10 Addressing Ethical, Legal, and Social Implications (ELSI)

1. Birth of the ELSI Program

During an October 1988 press conference, James Watson, Nobel Laureate and first director of the HGP, announced unexpectedly that a portion of the budget for the soon to be launched HGP would be devoted to studies of the ELSI of the research (Schmeck 1988). Over the years, there has been considerable speculation about what he meant by this statement. Was he genuinely concerned about the potential harms, or was he merely trying to neutralize anything that might hamper the Project’s funding? In effect, was he performing a canny and prophylactic public relations coup? Both the Office of Technology Assessment and National Research Council planning studies had already identified potential concerns, and it was evident that those issues would need to be addressed in some way to secure from the United States Congress the billions of dollars of funding that would be needed to complete the Project (National Research Council [US] Committee on Mapping and Sequencing the Human Genome 1988; United States Congress, Office of Technology Assessment 1988).

Whatever Watson’s intention, the notion that a portion of HGP funding should be allocated to supporting what has come to be called “ELSI research” quickly became codified in the funding legislation for the Project; at first 3 percent of the budget, and later 5 percent, would be set aside to “anticipate the impact of the HGP and address what protections need to be in place, so that the information generated can be of maximum benefit to individuals and society” (NIH Revitalization Act 1993). Congress was so committed to the idea of an ELSI “set aside” that it also provided one for the Department of Energy (DOE)—the National Institutes of Health’s (NIH’s) US partner in the HGP. The DOE ELSI Program—considerably more modest than the NIH program—received 3 percent of the DOE’s HGP budget and focused largely on education efforts and policy research and analysis. The DOE program (the discussion of which is beyond the purview of this paper) was discontinued with the completion of the first genome sequence.

This decision was not without controversy. Many in the genomics research community saw it as, at best, an unwarranted expenditure of limited funds that could better be spent on the basic science (Annas 1991). Some saw it as a “welfare program” for bioethicists, “subsidizing the vacuous pronciamentos of self-styled ethicists” (Marshall 1996). The special set-aside of funding also carried with it the risk that genomics would be made to seem riskier than other types of research—leading, perhaps unjustifiably, to a sense that genetics was “exceptional” in some way.

Initial skepticism was not limited to members of the genomics research community. Far from seeing the creation of an ELSI Program as an unalloyed good, some bioethicists looked askance at the very notion of creating a research program within the larger scientific initiative it was supposed to be charged with examining with a critical, independent, lens. These critics viewed the Program’s creation as little more than a jaded attempt to control potential HGP critics (Murray 1992). Indeed, some pointed to the fact that when the first list of issues to be addressed by the Program was released, it failed to include the most relevant issue: whether the HGP itself should be given so much funding when there were so many other priorities deserving of federal dollars (Hubbard and Wald 1993).

2. The Policy-Research Debate

As planning for the HGP got underway, a Working Group on Ethics was constituted to develop a plan for what would ultimately become the ELSI Program. When the Program was officially established in 1990 as a part of the HGP at the NIH National Center for Human Genome Research (NCHGR) and the DOE, this committee was reformulated to serve as a more permanent working group of the NIH National Advisory Council for Human Genome Research and the DOE Health and Environmental Research Advisory Committee, so that it could continue to “play an active role in refining the agenda” for the ELSI Programs at both agencies. The working group’s agenda included a broad range of activities—not only research, but also activities in the education, public policy, and outreach arenas. In fact, the “development of policy options” was one of the ELSI program’s first stated goals, along with efforts to stimulate public discussion and input; research was merely one of the tools available to the Program as it was initially conceived. As already noted, the DOE ELSI program with its more limited budget focused more on education and public policy efforts, while the larger NIH program prioritized multidisciplinary research and education activities focused on the impact of the HGP in health care settings and on society more broadly.

The breadth of the initial ELSI mandate, the poorly defined division of responsibility among an extramural research program and a Working Group to the Advisory Council, and the explicit inclusion of policy development in the Program’s mandate proved to be problematic. The positioning of the ELSI Working Group within the NCHGR and DOE advisory process structure made it difficult for that group to have the tools or the reach to exert a meaningful impact on policy development at a national level (Roberts 1993). Nor did the ELSI Program—situated as an extramural grant program operating within the NIH peer review system—have the flexibility or authority to play an active role in policy development. As the policy issues raised by the HGP continued to unfold and as the HGP leadership grew increasingly frustrated with the slow pace at which they were being addressed, it gradually became clear that neither the Working Group nor the research program were appropriately positioned to meet the growing need for effective policy development.

In 1996, as part of a routine five-year planning process for the HGP, a committee was constituted to evaluate the structure and function of the ELSI Working Group in relation to the extramural ELSI Program and other federal advisory bodies. This evaluation group published a report recommending that “a federally-chartered Advisory Committee on Genetics and Public Policy should be established in the Office of the Secretary of HHS to undertake the formulation of public policy resulting from advances in genetics” (Rothstein and Spence 1996). The report also recommended that a smaller, more focused ELSI Research Evaluation Committee be established to provide advice and guidance to the ELSI Program within the NIH. This division of responsibilities relieved some of the initial expectations that the ELSI Program play a direct role in policy development and allowed the Program gradually to refine its agenda to focus on the support of projects to develop data that could inform, but not necessarily drive, the development of research, health, and public policies.

Despite the reorganization of the ELSI Working Group and the creation of a policymaking body within HHS, the expectation that the ELSI Program would play a more direct role in policy development has continued to exert subtle pressure on the ELSI Program through the years. This has occasionally presented significant challenges. An example can be seen in the Program’s years-long attempts to address in a systematic way the complex issues around gene patenting and intellectual property concerns more generally—issues identified almost from the outset (and, in fact, mentioned explicitly in the enabling legislation) as in need of urgent policy development. The ELSI Program did support some early research by Mildred Cho and others in this area and supported work by David Blumenthal and Eric Campbell in the related area of conflict of interest; work on intellectual property by scholars such as Rebecca Eisenberg was also an important early focus of DOE’s then-still extant ELSI Program (Blumenthal et al. 1996; Eisenberg 2002; Cho et al. 2003). The ELSI Program also issued a 2005 Request for Applications (RFA) on intellectual property issues, but as will be discussed further below, it was not until years later, after the 2004 development of the Centers of Excellence in ELSI Research Program and the identification of a cohesive team of scholars with the requisite grasp of the full range of scientific, legal, and economic issues, that the Program was able to address these issues in a way that could exert a major influence on policy.

The tension between the desire for policy development and the need for research to undergird the development of such policy remained an important tension in the relationship between the ELSI Program and the rest of the HGP enterprise throughout the course of the HGP. Indeed, this tension remains to this day an ongoing theme in the relationship between the ELSI Program and the genomics research community more broadly (Burke et al. 2015).

3. Building a Conceptual Foundation

When the ELSI Program was established in 1990, Eric Juengst, a highly regarded young bioethicist, was recruited by NIH to make Watson’s vision of an ELSI research initiative a reality. Dr. Juengst worked closely with the Working Group to develop an initial road map for the program. This plan identified nine categories of ELSI research priorities (in additional to education activities), which were later streamlined into four broader categories: Privacy and Fair Use, Clinical Integration, Research Issues, and Education. Still, in the very earliest years, the research the Program supported was largely anticipatory—dominated by broadly themed conferences and research projects focused not so much on issues of immediate import but rather, largely, on more theoretical issues—issues thought likely eventually to arise, once the human genome sequence was completed and made available to the public. Projects by legal scholars and bioethicists on the use of genetic information to discriminate against individuals in various settings, ranging from healthcare to employment and beyond, dominated in these years, as did more conceptually focused projects on such issues as the challenges associated with ensuring equal access to the benefits of genetic information and the dangers of germline enhancement. Examples of this work can be seen in the early projects by David Wasserman and Bernard Gert (Berger and Gert 1991; Wasserman 1993). The Program also funded a number of studies examining the history of eugenics and its legacy for modern genomics research, exemplified by the work of William Schneider, Howard Markel, and David Micklos (Schneider 1995; Markel 1997; Micklos and Carlson 2000).

This early conceptual and analytical work provided a foundation for many subsequent analyses, and in the case of the work on genetic discrimination, for the eventual (2008) passage of the Genetic Information Nondiscrimination Act (2008). Much of this early work also continues to inform current debates within the research and policy communities about how to regulate and govern the use of emerging genomic technologies. For example, early projects on genetic enhancement and germline editing, which at the time seemed to some as fanciful, are informing current debates about the regulation of CRISPR-CAS gene-editing technology.

Still, for many in both the research and policy communities, the work of these early investigators was viewed as too speculative, not closely enough tied to the realities of existing genomic science, and unlikely to exert much of an impact on how genomic information and technologies would ultimately be developed and used in real-world settings (Hanna 1995; Lehrman 2000). Thus, even though, over the years, it has often been the very research criticized at the time as most theoretical and speculative that has ultimately turned out to be most influential, continuing doubts about the value of research that is primarily conceptual in nature have remained a persistent theme.

In 1993, Dr. Juengst left the program to take a position at Case Western Reserve University. Before his departure, Elizabeth Thomson, a pioneering nurse genetic counselor at the University of Iowa, joined the program to help develop and manage the Programs’ clinical and human subjects research efforts—efforts which were rapidly becoming a major focus as the sequencing of the first human genome began in earnest.

4. One Person’s Genome?

From the beginning of the HGP, it was widely assumed that the actual first sequence would be based on an amalgam of genetic material derived from many different individuals. However, in 1995, as the HGP began to turn its attention to the creation of the first or “reference” human genome sequence, it became apparent that most of the material to be used for the sequence was based on the bacterial artificial chromosome library of a single person. The initial reaction from the scientists involved in the project was that while this was not necessarily the original vision for the sequence, it was not a significant problem since, after all, the sample had been de-identified, making it unlikely that the individual donor would suffer any negative psychological or social harms. Some in the ELSI community, however, voiced concerns about the issues that could arise and cause problems for the project in the future—particularly if the identity of the individual were ever to be revealed. Not only would that person run some risk of discrimination or stigmatization if their sequence were to suggest predisposition to particular diseases or disorders, but their family members might also be at risk. In addition, the notion of basing the reference sequence on the DNA of single person seemed inconsistent with the message about the reference sequence as a reflection of all of humanity—not just one human being.

As the discussion around this issue continued, it thus became apparent that steps would need to be taken to address the problem. Ultimately, HGP staff from NCHGR and the DOE worked together with the genomics research community to develop a protocol to recruit and consent participants to donate DNA samples that could be used to expand the libraries and ensure that the sequenced material would be more diverse and adequately consented. This process resulted in the development of the first guidance on Human Subjects in Large Scale Sequencing (NCHGR-DOE 1996). It also marked the first time that the bioethics expertise of the ELSI Program made a direct contribution to the day-to-day work of HGP researchers.

Although it could be argued that this contribution slowed down the pace of the research and was thus initially viewed with some ambivalence by the genomic research community, it was eventually recognized as a necessary and prudent step. This became especially apparent when, years later, the commercial sector announced the creation of a sequence based on the DNA of a single well-known genomic scientist and faced criticism because of that fact.

5. Anticipating the Future: Initial ELSI Forays into the Clinic

Research on how best to ensure the safe and effective integration of genetic and genomic information into clinical practice has always been a high priority for the ELSI Program, and over the years, it has consumed a substantial portion of the Program’s budget. This area of research has addressed a wide range of issues, such as what information should be provided to patients and their families, how this information should be provided to ensure equal access to health services, when this information should (and should not) be used clinically, and how to ensure that health-care providers have the knowledge and confidence to order tests and advise patients on results and that patients have the information they need to make informed decisions.

The earliest ELSI research in this area focused on the handful of conditions for which predictive genetic risk information was already, at that time, available. For example, in 1991, the ELSI Program released an RFA focused on the issues surrounding the possible implementation of carrier screening for cystic fibrosis (CF) and funded seven studies at institutions across the country. To facilitate discussion among the investigators and to avoid duplication of effort, the investigators were brought together to form a consortium that met periodically over the course of the studies. The results of these studies supplied the basis for a 1997 NIH Consensus Development Conference that produced recommendations on the use of CF genetic screening; those recommendations, in turn, were used as the basis for practice guidelines issued by the American College of Obstetrics and Gynecology and the American College of Medical Genetics, marking the first time ELSI research was used directly to inform the development of policy in the clinical arena (National Institutes of Health 1997).

The CF consortium was followed by the issuance in 1994 of an RFA focused on ethical issues in predispositional testing for breast, ovarian, and colon cancers and the formation of a similar consortium. This highly productive consortium issued a series of recommendations, published in JAMA, which were highly influential in the development of clinical guidelines for the use of predispositional genetic testing more generally—and which remain important to this day (Burke et al. 1997a, 1997b; Geller et al. 1997).

This practice of issuing an RFA followed by the formation of a related research consortium, pioneered by Elizabeth Thomson, proved to be so effective that it has become a model for many subsequent RFAs issued by the Program. Members of these consortia have become remarkably conversant in genetics and genomics, and over the years, the consortia have been instrumental in producing quality research. The use of this model has also helped to build a strong multidisciplinary research community skilled in interdisciplinary collaborations—a critical feature of successful ELSI research.

Despite the success of the early ELSI forays into the clinical research arena, concerns began to be expressed—especially by clinical geneticists—about a perceived overemphasis on the risks associated with genetic information relative to its potential benefits. Critics complained that some ELSI research in this area tended to promote an exaggerated view of genetic information as uniquely scary and dangerous, thus giving rise to the very kind of “genetic exceptionalism” (and in turn, the risk of stigmatization and discrimination based on genetic status) that the ELSI community itself had identified early on as a major risk related to genetic information (Grody 1995). Some ELSI scholars observed that there was some irony to this criticism; after all, it was often the most enthusiastic proponents of genetics and genomics who, at least arguably, may have contributed the most to a “genetic exceptionalist” stance, by “hyping” the role of genetic factors as a major, almost deterministic, explanatory factor in all health and disease (Caulfield 2016). Nevertheless, and as will be discussed further below, debates over where concerns about “genetic exceptionalism” might originate, and about the extent to which such exceptionalism is a useful or unhelpful construct, continue to this day—the source of another ongoing yet productive tension.

6. Entering the Arena of Genomic Research Ethics

In addition to the concerns about the promotion of genetic exceptionalism, a second criticism of the early clinically focused ELSI studies was that they were still too anticipatory—too focused on issues likely to arise mainly in the future, when genetic medicine would be more widely implemented into clinical practice, and not on the “here-and-now” issues genomic researchers were facing in their work. The Program gradually did turn more of its attention toward these kinds of issues, but when it did, another tension—presaged to some extent by the tension surrounding the earlier debate over how many people’s genomes to sequence—soon emerged. This tension can perhaps best be illustrated by the history of the controversy around the use of stored tissue samples that emerged in the 1990s.

Historically, the research use of stored tissues to which individual identifiers were not attached was not considered to involve human subjects, and thus, informed consent was not thought to be required for the samples’ future use. In the early 1990s, however, ELSI researchers began to question this assumption. The ELSI Program funded a number of studies that examined the adequacy of informed consent and privacy protections for stored samples, and out of these studies came a proposal for heightened protections to guide their use (Weir and Horton 1995a, 1995b). In 1994, amid growing concerns, the ELSI Program, along with the Centers for Disease Control, convened a meeting of ELSI investigators, genomic and clinical genetics researchers, and members of the pathology community to further explore these issues. Following intensive and sometimes explosive deliberations, recommendations were developed and published in a 1995 issue of JAMA (Clayton et al. 1995). These recommendations, which stated that in order to reuse stored samples, researchers should be required to go back to the original donors and obtain new informed consent (or alternatively a waiver of consent), generated considerable controversy. Many researchers charged the ELSI scholars who had participated in their development with having once again overemphasized the “exceptional” character of genetic and genomic information. They complained that implementing the recommendations would require spending more time and money and slow the progress of research, to the ultimate detriment of public health (Grody 1995; Stephenson 1996).

The recommendations were eventually adopted, however, and became the foundation for many of the standard procedures for obtaining consent for stored tissue research that are in use today. The rationale behind the recommendations also led to the subsequent development of more stringent access policies for stored genomic data, as reflected in the NIH Genomic Data Sharing Policy and similar data access policies developed at institutions around the country (Health 2014; NIH 2014). Genomic researchers gradually began to accept the rationale for more protective measures, and today, nearly twenty-five years after the publication of the initial recommendations, most people generally recognize their appropriateness. Today, the issue is most often addressed prospectively, and as genomics researchers now generally seek informed consent for the future use of samples, they plan to store at the time they first obtain them.

The evolution in attitudes and practice that has occurred in this area since the 1990s has no doubt been influenced, in large part, by growing awareness of how easy it has become to identify individuals (and their close biological relatives) by combining de-identified DNA data with other data—including data readily available from an ever-expanding array of public sources. For example, the DNA ancestry testing industry, scarcely envisioned in the 1990s, is today being used not only to help long-lost relatives find each other, but also, increasingly, by law enforcement, who in a growing number of well-publicized cases have been able to combine DNA ancestry test data, leads from relatives, and information readily available from other public sources to identify suspects in unsolved criminal cases.

Another explanatory factor behind the gradual change in attitudes around the treatment of stored samples and data may well have been a study funded by the ELSI Program in the early 2000s showing that most people who donate samples for research are, in fact, willing to let them used in future studies; they merely, as a matter of respect, wish to be asked (Ludman et al. 2010). This study (and others funded by the Program documenting research participants’ views on a range of research ethics issues) provide an example of how ELSI research has been and can be used to inform not only changes in formal policy, but in the broader culture of genomics research—albeit with some bumps along the way.

7. Integration and Embedding: Initial Experiments

Following the completion of the reference sequence, the genomics research community began to turn its attention to the subject of human genetic variation: examination of the ways in which individuals differ (see chap. 5). Most genomics researchers were aware, however, that the research on which they were about to embark would have to be conducted with special attention to its ethical, legal, and societal implications. This was because the most efficient route, at least initially, to understanding how individuals vary was to look at how patterns of variation differed among people from different ancestral groups. Examining such group differences, researchers understood, would raise questions extending well beyond issues like individual informed consent, privacy, and the handling of research results—the issues that, up until then, had been the major focus of ELSI research. This is because genetic variation research has heightened potential to be misunderstood and misused by the public in ways that can lead to group stigmatization, discrimination, and broader social harms.

By the time genomics researchers began to turn their attention to the study of genetic variation, the ELSI Program had already funded a large body of work on the risks inherent in assessing genetic group differences. This work, which included projects by Troy Duster and William Schneider, described the negative societal consequences that can ensure when erroneous, oversimplified, or deterministic beliefs about genetics intersect with folk beliefs about race, ethnicity, and social identity—especially in societies where pervasive biases, both explicit and implicit, are already deeply engrained (Duster 1995; Schneider 1996). The Program had also funded some anthropological work by Morris Foster highlighting the importance of engaging in an affirmative consultation process with representatives of communities whose members are being approached to participate in genetic variation research as “representatives” of broader populations (Foster et al. 1997; Foster et al. 1998; Foster and Freeman 1998). However, the Program recognized this as an area in need of additional research, so once again, in order to anticipate rather than merely react to the issues likely to be raised by this emerging line of research, the Program issued a 2000 RFA to solicit studies addressing these issues.

The ELSI investigators funded through this RFA represented an exceptionally diverse array of disciplinary perspectives, ranging from anthropology and sociology to the history and philosophy of science. As had become the standard practice, a consortium of these researchers was formed and began meeting regularly to share methods and findings. Thus, by 2001, when the International HapMap Project was about to be launched, these researchers were well positioned to provide advice about the design and conduct of the Project. Most were eager to put the insights derived from their research to use in a concrete, real-world setting—and this time, most of the genomics researchers, cognizant of the need to conduct their research with particular attention to the associated ethical issues, readily welcomed their input.

A more complete discussion of the way ELSI input was incorporated into the basic science component of the HapMap Project can be found (see chap. 5). Although many HapMap Project decisions were of necessity made based on a mix of scientific and practical concerns, Genetic Variation Consortium members—some of whom became part of a more formal international ELSI working group for the Project—helped to inform many important decisions. For example, they provided crucial input on the ethical considerations that ultimately informed decisions about which populations to include, as well as on how to incorporate local concerns into study protocols while simultaneously adhering to internationally recognized ethical norms. They also helped genomics researchers understand the need for precision in how each HapMap population was to be labeled, as well as the need for participant communities to have some input (within scientific constraints) into how the broader populations their samples were purporting to represent would be named. This led to the publication of formal guidelines for population labeling, which over the years have helped to minimize (even if not completely alleviate) the risk that who study the samples or the data will apply sloppy, imprecise labels that simplistically conflate ancestral geography with folk conceptions of race.

Perhaps the major contribution of ELSI experts to the HapMap Project was their insistence that some form of community consultation or community engagement, in addition to the obtaining of individual informed consent, be conducted with representatives of each local community approached to provide samples for the Project. These processes, described elsewhere in greater detail (see chap. 5) (Rotimi et al. 2007), were extensive and time-consuming, and slowed the pace of the project (Rotimi et al. 2007). While this created frustration for some of the Project’s genomic researchers, it helped to ensure that group concerns could be identified and, to the extent possible, proactively addressed. The extensive community consultation and community engagement processes employed in the Project also helped inoculate the Project from accusations of biopiracy and “helicopter science” that had plagued some earlier genetic variation research efforts, such as the Human Genome Diversity Project (Cavalli-Sforza et al. 1991).

The HapMap Project marked the first time in the ELSI Program’s history that a conscious effort was made to integrate ELSI research fully into an ongoing genomics research initiative. This made it possible to address ethical, legal, and social concerns at the same time that the underlying research was being conducted. In several cases, ELSI research was actually embedded directly into the community engagement processes. Although this raised some occasional questions about investigator independence, for the most part, they were navigated successfully.

The community engagement processes produced a substantial body of qualitative data that has helped to inform the way subsequent genetic variation studies, such as the 1000 Genomes Project (launched a few years after the HapMap), were approached. In the 1000 Genomes Project, a more streamlined, location-specific approach to community consultation and engagement was employed. Nevertheless, ELSI experts again interacted extensively with genomics researchers through the Project’s course.

The integration and embedding of ELSI investigators and basic genomic scientists in the HapMap and 1000 Genomes Projects—although initially approached as a bit of an experiment and encountering some challenges the way—was generally regarded as a success. For many of the ELSI and genomic researchers involved in these Projects, the collaborations they established helped them learn more of each other’s professional languages and cultures, which in turn seemed to generate an atmosphere or greater trust than had existed before.

8. Building on the Trend Toward Enhanced Integration

Building on the success of the partnerships established between ELSI and genomics investigators through these genetic variation initiatives, a growing recognition emerged of the need for a formal research mechanism to facilitate more such interactions. This led in 2004 to the creation of the ELSI Centers of Excellence (CEER) Program, specifically designed to support the development of transdisciplinary teams to better integrate investigators from the full range of ELSI research disciplines with genomic and clinical genetic investigators. Interestingly, the initial idea for these Centers grew out of a suggestion by a prominent HGP researcher during one of the 1998 NHGRI planning meetings—a signal, perhaps, that the value of incorporating ELSI input into genomic research was beginning to be more fully recognized within the broader genomics community.

The CEER Program was designed not only to enhance such integration but also to play a prominent role in the training of the next generation of ELSI researchers. In addition, in a nod to the continuing frustration of HGP leadership with the lack of a more direct ELSI impact on policy development, the Centers were tasked with translating ELSI research to facilitate the development of policy. These three programmatic goals, codified as the “three Ts” (Transdisciplinarity, Translation, and Training), became the CEER Program’s primary mission.

Four full centers and three exploratory centers were funded in 2004, and by 2019, more than eleven such Centers had been funded. The Centers have been organized around a variety of topic areas, ranging from the delivery of genomic health care to the medically underserved to psychiatric and behavioral uses of genomics. Collectively, the Centers have produced hundreds of peer-reviewed publications and books and have played an essential role in training a new and more diverse community of ELSI researchers.

Despite the many concerns expressed over the years about the appropriateness of federally funded researchers being directly involved in policy deliberations, a number of the Centers have also been able to provide essential input into the development of state and federal legislation and other public policies. For example, one previously funded Center, at Duke University, provided important leadership on some of the intellectual property issues which, as previously noted, were recognized as an important component of the Program’s mission from the Program’s very earliest days but had theretofore been inadequately addressed.

9. Revisiting Clinical ELSI Issues in the New Age of Genome Sequencing

In addition to launching the CEER Program, the ELSI Program in recent years has taken other steps to more proactively integrate its funded researchers into major NHGRI-led genomics initiatives. A major development occurred when sequencing costs began a rapid decline, so that whole exome and whole genome sequencing became increasingly viable alternatives to genotyping. Unlike the more targeted genetic testing of the previous generation, the use of sequencing technologies had the potential to generate an enormous number of secondary or incidental findings—often of uncertain significance. Genomic researchers, and increasingly, clinicians who were incorporating these technologies into their practices, were at a loss as to how to handle them, and ELSI Program staff quickly recognized that there was a pressing need to address the questions related to whether—and of so, how—to offer such findings to research participants and patients. The Program thus issued an RFA to address these questions, and as had become the routine practice, organized the funded investigators into a consortium—the Return of Results Consortium.

Shortly after the formation of this consortium, NHGRI’s Division of Genomic Medicine issued a separate RFA to support exploratory research addressing a broad range of issues relating to the generation, analysis, interpretation, and translation of sequencing data being generated specifically in clinical settings. That initiative also incorporated a distinct ELSI research component at each funded site, providing an opportunity for ELSI investigators to work hand in hand with researchers directly involved in cutting-edge clinical sequencing work. A consortium of the investigators and projects funded through this initiative, called the Clinical Sequencing Exploratory Research (CSER) Consortium, was formed in 2011, and shortly after its launch, that Consortium was expanded to include the ELSI investigators and projects previously constituted as the Return of Results Consortium. Folding the investigators in the Return of Results Consortium into the much larger CSER Consortium led to the integration of an exceptionally broad and diverse collection of clinicians, genomic researchers, social scientists, informaticians, bioethicists, and legal scholars—all working closely together and learning from each other’s work (Green et al. 2016).

The research of some CSER ELSI investigators was directly embedded into the sequencing projects underway at particular sites, which provided a natural laboratory for the generation of useful empirical data. The quantitative and qualitative data emerging from these studies not only provided insight into patients’ attitudes, preferences, and beliefs about secondary and incidental findings, but was also able to document real-time behavioral and psychosocial responses and reactions to the receipt of such information. ELSI researchers in the CSER Consortium published influential papers on a wide range of other topics as well, such as on differing approaches to the return of results in the research and clinical context (and the murky boundary between the two contexts), the pros and cons of various consent models, and the legal implications of incorporating genomic information in electronic health records (Hazin et al. 2013; Jarvik et al. 2014; Wolf et al. 2018; Yu et al. 2019).

The inclusion in the CSER Consortium of bioethicists and legal scholars—some embedded directly within the clinical sites and others working more independently—proved to be very effective. The normative and legal analyses these researchers conducted, especially when integrated with the empirical data, helped to ensure that the best practice recommendations that eventually emerged from the Consortium were not only evidence based but legally and ethically sound.

The high level of integrated (and in many cases directly embedded) research facilitated by the CSER Consortium’s structure led to many new and productive (and continuing) collaborations among ELSI researchers, clinicians, and others involved in various aspects of genomics. The Clinical Sequencing Evidence-Generating Research Consortium (established in 2017 as a successor to CSER and bearing the same acronym) has extended many of these collaborations; although the RFA through which the investigators in that Consortium were funded, unlike the first CSER Consortium, did not incorporate a mandatory component specifically focused on ELSI, it did incorporate a related component on stakeholder engagement.

Despite the close and generally productive collaborations the CSER Consortium successfully fostered, tensions were occasionally in evidence, stemming from the divergent professional orientations among ELSI investigators, clinicians, bioinformaticians, and other Consortium researchers. For example, clinicians and other Consortium participants occasionally voiced frustration about the approach to the return of secondary and incidental findings some ELSI participants were advocating, which they viewed as overly conservative. These participants saw ELSI investigators’ recommendation that clinicians should limit the extent to which they provide such findings as invoking the same old “genetic exceptionalist” stance they thought had long been discredited. Indeed, some clinicians in the Consortium took the position that regardless of whether or not a patient wished to receive such findings, so long as the information had even potential medical relevance, they not only should be allowed to provide it, but indeed arguably had an ethical—or perhaps even legal—duty to do so.

In the end, out of this productive tension (and further informed by the perspectives of patient participants in the underlying research), the Consortium was able to reach a “truce” of sorts on the issue. More importantly, though, the empirical data and the normative and legal analyses relevant to this issue that ELSI researchers within the Consortium produced eventually led to several important revisions in a set of recommendations that had been issued by the American College of Medical Genetics a year or so earlier—before they had the benefit of this scholarship (Kalia et al. 2017). This was an example of how the productive tensions that occasionally arise between ELSI researchers and others involved in the genomic enterprise can often be resolved based on the ELSI research itself—and ultimately exert a concrete impact on broader policy development.

10. Moving into the Future

This paper has described a number of productive tensions that have emerged—and have occasionally resurfaced—over the life of the ELSI Program. Other productive tensions have also been present, at least subliminally, since the Program’s inception, and will present ongoing challenges for the future. For example, what should be the balance between normative-conceptual research and empirical research in the Program’s portfolio? What should be the balance between research with direct policy relevance and research (such as that occasionally conducted by humanities scholars) focused on issues that seem futuristic or largely hypothetical? When is it appropriate for ELSI researchers to play a direct role in the development of policy (Walker and Morrissey 2012)?

Should all research supported by the ELSI Program have at least some direct biomedical relevance? Or would it be an abdication of responsibility for the Program not to allocate at least some of its resources to the support of research on the ELSI of genomics in society more broadly? For example, should the ELSI Program support research on the implications of DNA testing in such areas as criminal forensics, immigration law, family law, mass disasters, or similar contexts? Or are such issues beyond the ELSI Program’s appropriate purview (Research Group 2002)?

Finally, what should be the balance between fully integrated (or perhaps even embedded) research and research conducted by ELSI scholars independently from genomicists or clinicians? Does the support of integrated or embedded research run the risk of simply turning ELSI researchers into “handmaidens of the science,” as some critics have alleged? Can there even really be such a thing as truly “independent” ELSI research, given that the ELSI Program is housed within the very institution that funds the underlying genomic science? Might there, in fact, be some truth to some skeptics’ claim that all ELSI researchers are necessarily, at least to some extent, beholden to their funders, making true independence impossible (Annas 1991)?

The ELSI Program has navigated these tensions since its inception and will continue to do so into the future. In fact, it has often been these very tensions, and the efforts to resolve them, that have allowed the Program to evolve productively. Equipped with the tools necessary to anticipate and address the ever-changing implications of genomics for individuals, families, and society more broadly, the Program seems well poised for the future.

References