Translational Research — For the Individual and the Community

Abstract

The potential for translational research to improve human health is unprecedented, as the integration of genetic health risks with other data influencing health provides substantial opportunities for improvement. However, how integrating these data sources in a fair, unbiased and appropriate way without reinforcing pre-existing assumptions requires thoughtful implementation. Furthermore, integration of new technologies requires assessment of needs and benefits for the individual balanced with community needs and goals. Thus, examination of values, goals and implicit assumptions through transparent, authentic engagement of individuals and communities is essential.

Advances in translational research from mechanisms of human biology through environmental modifiers of phenotype and then to application to human disease present exciting opportunities for improvement of health. Focusing on the individual within the medical model of disease, innovative treatment strategies for cancer now include specific enzyme inhibitors, immunotherapy with genetically engineered immune system cells obtained from patients, and immunotherapy with targeted monoclonal antibodies for selected patients whose tumors display the appropriate targets reacting with those antibodies. Characterization of cancer tumor genetics of patients has enabled development of individualized treatments that promise greater efficacy with fewer undesirable side effects. Similar insights for rare diseases with genetic mutations will advance treatment options for people afflicted with such conditions.

In addition to the context of this innovative, translational research environment, enabled by an individualized approach engaging biotechnology, there is a transformative change in the traditional model of translational research -- the “digital model” based on data gathered not only from individual health information, often based on in-person encounters, but also on social determinants of health and other factors.^1–4 This rapidly evolving model, which employs computer-based prediction algorithms (developed in the field of artificial intelligence) blurs the boundaries of individual-level information and challenges us to consider the ways in which data from our use of digital tools, including social media and other public sources, should be governed.⁵ These challenges apply to all of us, but may be particular impactful for the medically underserved. ^6,7

The Belmont Report

The Belmont report, created by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research in 1978, emphasized the importance of respectful, authentic, trustworthy, and transparent relationships with individual study participants involved in biomedical and behavioral research.⁸ The report also emphasized the principles of justice and beneficence. Working from the medical model of human investigation, many interpret the report as giving priority to respect for persons’ autonomy over the general good of society. However, definitions of the ‘general good’ and fairness for both the individual and the society may vary according to the social and philosophical frameworks being applied.^{9, 10}

The boundaries between biomedical ethics, health equity, and social policy may blur, and these factors may merge. Individual interests and established community values may not be well aligned. These fundamental tensions are brought into relief with new biomedical innovations and with the digital model of investigation for which permissions and intent may be less clearly understood by participants, especially in the context of the use of social media data.¹¹

Examples for consideration.

(1) Realizing the potential of the Human Genome Project. The ever-increasing information about human genetics as the roadmap for each individual provides new opportunities to improve human health. Translational research projects to define individual genetic information and relate this information to current and future individual health status have great potential for more effective health care.^12,13 However, the information is incomplete, and the genetic roadmap has multiple roadways, the choice of which is influenced by the individual’s personal habits and behaviors as well as by his/her environment.^6,7 Thus, the perspective of genetic determinism in health needs a broader view, and some suggest a framework phrased as genetic “contextualism” to convey the concept that while genetic factors are important, other environmental factors also contribute importantly to health and health outcomes^14,15 As a result, practical and bioethical questions arise when considering translational studies involving genetics: 1) how do we set expectations for participation; 2) how do we inform people about results as they develop in a way that respects privacy and prevents discrimination and is cognizant of the interdependence of genes, the environment, and social factors; 3) how do we explain future use; and 4) how do we account for different levels of understanding so that decisions are informed and participants engaged? No single approach can address these issues for all participant groups. As translational investigators and health care providers grapple with these issues,^16–19 they must be sensitive to and respectful of individual levels of health literacy and potentially different group norms and to be authentic about the implications and applications of the genetic information.

Examples for Consideration.

(2) Advances in organ transplantation. For many patients, the shortage of organs available for donation limits access to life-saving procedures. For people with kidney failure, renal dialysis provides a temporizing and life-sustaining option while awaiting kidneys for transplantation. However, patients on dialysis have accelerated rates of mortality, and the option of receiving a humanized porcine kidney (xeno-transplantation) is nearly at hand.²⁰ This remarkable biotechnical advance, resulting from translational research, creates a series of practical and bioethical questions: 1) when is it appropriate to move forward to xeno-transplantation in humans; 2) among all who might benefit, how are the first people to receive this procedure selected; 3) are there contextual factors -- social, economic, spiritual, in addition to medical and physiological factors -- to be considered before proceeding with xenotransplantation in patients? Individual medical need, constraints on availability of allogeneic matched donors, and individual commitment to contribute to the advance of medical care require balance with the norms in more conventional transplantation^20–22.

Examples for consideration.

(3) Predictive analytics and artificial intelligence. Large amounts of data are now available on each of us as individuals -- not only biometric data that may come from wearable devices as well as from the more conventional digital medical records (with permission) but also data reflecting a person’s environment, including social determinants of health, such as location of residence, socioeconomic status, and educational background.^4,5 These data, coupled with information available through social media, may enable development of computer algorithms that not only reflect but also predict health status and outcomes. Predictive analytics, including polygenic risk scores, can be a powerful tool for guiding decision-making and allocation of healthcare resources.^10–12 This capacity points to a series of practical and bioethical questions: 1) does machine-learning with resultant predictive algorithms inadvertently create or perpetuate inequitable treatment; 2) how do we establish trust in the ‘who’ and the ‘how’ guidelines are developed; 3) how do we deal with information in the domain of social media within which the boundaries of privacy and permission are different from those in the current system of healthcare records; 4) how do we balance the tools of data science for the public good? Much has been written about approaches to achieve fairness; about concern over implicit bias, transparency, and governance; and about the potential to recapture the essential in-person connection in medical care.^{7, 8, 23–26} Within these discussions of machine-based algorithms to define insightful patterns of disease risk and outcomes, a recurrent theme is that fairness requires human intelligence and oversight.⁸

Examples for Consideration.

(4) Access to biomedical innovation. Among the highly effective biomedical advances reflecting innovative technologies are specific humanized or fully human, therapeutic monoclonal antibodies for a range of cancers and autoimmune diseases and anti-microbial agents enabling cure of conditions that previously lacked treatments.^27–29 These advanced treatments are generally costly. Although they represent remarkable advances made through translational science and do not press the bioethics of performing the translational research underpinning their development, the access to these treatments does challenge the boundaries of ethics, equity, affordability and social policy. With treatment costs that range from $100,000 to more than $1,000,000 per person, health care plans are challenged to meet costs. To realize the full potential of translational research, new models of access may be necessary, such as the Louisiana “Netflix payment model” for hepatitis C medications.³⁰ Such partnerships point to innovative ways by which communities can come together to address these boundaries.

Opportunity for the present.

As the pace of innovation continues to quicken, the potential for translational science to yield even greater advances in health care and health status has never been brighter. The introduction of the digital model, bringing both individual and community data from multiple sources together on an unprecedented scale has created a new set of challenges in trust, transparency, agency, and equity.³¹ Opportunities and challenges are presently at hand, and as communities come together to address these challenges, the public good will be well served.

Some Bioethical Issues ^9,10,11.

• What is “fairness”, and how is it defined? How is individual benefit balanced with community benefit?

• How does one define the boundaries of community information and technology use when individual health is involved?

• What types of decisions should be delegated to computer -based systems (artificial intelligence)? Who decides the underlying algorithms and decision patterns?

• What is the appropriate philosophical framework(s) for discussion (frames of faith, of pragmatism, of utilitarianism)?

Summary Points.

Exciting new opportunities in translation research (e.g., human genetics and genomics, predictive modeling with artificial intelligence, xenotransplantation) require examination of tacit and explicit values, goals, and assumptions.

Authenticity, transparency, respectful consideration of different perspectives and desires for privacy are essential.

Trustworthy governance of translational science initiatives is necessary.

Communities must come together to create the paths forward in the implementation of translational knowledge for better health care.