Categorizing biomedical research: the basics of translation

Abstract

As biomedical research has evolved over the past century, the terminology employed to categorize it has failed to evolve in parallel to accommodate the implications of these changes. In particular, the terms basic research and translational research as used today in biomedicine seem especially problematic. Here we review the origins of these terms, analyze some of the conceptual confusions attendant to their current use, and assess some of the deleterious consequences of these confusions. We summarize that the distinction between basic and translational biomedical research is an anachronism. Elimination of this often contentious distinction would improve both the culture and the effectiveness of the scientific process, and its potential benefits to society.—Flier, J. S., Loscalzo, J. Categorizing biomedical research: the basics of translation.

There is power in language that often transcends the simplest of intentions in its construction. Such is the case for the term “translational research,” which is defined by the European Society of Translational Medicine as an interdisciplinary branch of biomedical science supported by 3 main pillars: benchside, bedside, and community (1). Defined in this way, translational research involves the application of scientific observations to the human condition, a process that involves many steps from conception of the problem to its ultimate application (2). “Basic research,” by contrast, refers to scientific research conducted without any particular practical purpose in mind a priori. There are, however, many nuances and confusions attendant to the use of these terms. To explore these distinctions and their implications for biomedical research, we should turn first to fundamental definitions

Research is based on intellectual investigation focusing on discovering, interpreting, and revising human knowledge of the world and as such, is a reflective endeavor. “Biomedical research,” as a subset of research is broad in scope, referring to activities spanning many disciplines of biology and medicine. Within these broad disciplines are experiments designed to understand reality by examining events at many different levels of organization, from the atomic level (e.g., structure of key biologic molecules), to the molecular and cellular levels (e.g., biochemistry, cell biology), to the organismal level (e.g., physiology and pathophysiology), and to the population level as well (e.g., population genetics, epidemiology, and public health). These domains are not tightly bounded: many fields of biomedical research, as self-defined or demarcated by professional organizations or academic departments, span many or even all of these levels of experimental inquiry.

Consider the discipline of neurobiology, with research addressing topics as diverse as the atomic structure of ion channels; signal transduction; development of the nervous system; systems properties of neural networks; the basis for the emergent properties of consciousness, cognition, and emotion; the molecular basis for diseases of the nervous system; and many others. Many such studies can be carried out in simple or complex models and increasingly in humans. Investigators can focus selectively on individual elements (e.g., ion channel structure and function), or integrate observations at multiple levels to answer a specific question. Consider a genetic disease of the nervous system in which a defined mutation causes a molecular alteration in a specific protein, the understanding of which requires studying the effects of the molecular defect on neuronal function (e.g., a channelopathy) and on complex neural circuitry (i.e., interneuronal communication) and behavior. Is there a clear line separating which component of such neuroscience research is basic and which is translational? The clarification of the system-wide (cellular or organismal) consequences of the mutation not only informs our understanding of disease pathogenesis but also informs the fundamental biology of the protein that could not be appreciated from studies of the protein in isolation.

Next, consider genetics, a field encompassing diverse, investigative efforts, spanning atomic resolution of DNA structure and DNA–protein interactions, the genetic basis for development, how changes in the genome cause altered function and disease, and the way in which genetic variation affects the fitness of populations. Each of these distinct aspects (and others) may be studied in different model systems, including organisms as diverse as yeast, worms, flies, mice, and most relevant to medicine, humans. Investigators interested in a specific biomedical problem (e.g., aging, metabolism) may carry out research spanning many of these levels of inquiry in more than one of these models. How can we distinguish basic from translational research in this context? Is research on the molecular details of DNA–protein interactions more basic than research on the role of DNA sequence variation in human health? Is research focusing on a specific protein in a simple organism more basic than research on the homologous protein in a human cell? Is a study at the atomic level more basic than a study of molecules, the latter more basic than a study of organelles and cells, and that, in turn, more basic than a study of complex organisms, just as some consider mathematics more basic than physics, physics more basic than chemistry, and chemistry more basic than biology? We think the answer to these questions is no.

Within all scientific endeavors, class distinctions can influence career choices and validate the perceived importance of one’s professional output. In a lecture one of us gives trainees on career development, a slide is presented, indicating one approach to hierarchies in science, in this case set by the importance and rigor of quantitative thinking in each discipline: pure mathematicians view themselves as scientifically superior to applied mathematicians and physicists, who view themselves as scientifically superior to chemists and biologists, who view themselves as scientifically superior to physician–scientists. This type of distinction between pure mathematicians and physicists was well illustrated by Peter Rowlett in a commentary in 2011 (3): In 1998, the engineer, Gordon Lang applied Thomas Hales’s 1970 solution to the Kepler conjecture (dating to 1611 and addressing the best way to pack spheres, which turned out to be the greengrocer strategy—6 in 2 dimensions, 12 in 3 dimensions, 24 in 4 dimensions, and 240 in 8 dimensions) to solve the problem of the optimal way to pack signals in transmission lines (modeled best as an 8-dimensional lattice). This solution opened up the internet for broad public use by maximizing the efficiency of signal transmission. When the mathematician Donald Coxeter, who helped Lang understand Hales’s mathematical solution, learned of Lang’s application, he was appalled that this beautiful theory had been sullied in this way. There are many other examples of this highly opinionated view of scientific hierarchies, not least of which is Ernest Rutherford’s comment that “all science is either physics or stamp collecting” (4).

Insofar as such self-affirming, hierarchical distinctions make us feel better about who we are, especially in a highly competitive environment, it is no wonder that the historical distinctions between basic and applied or translational research continue to exist in the minds of some faculty members, persisting well beyond their usefulness. When Michael Brown and Joseph Goldstein were awarded the Nobel Prize in Physiology or Medicine in 1985 for their work on cholesterol metabolism in which they identified the LDL receptor as defective in patients with familial hypercholesterolemia, many of us thought that the distinction between basic and applied biomedical research had become an anachronism and would (should) dissipate. To be sure, as modern medicine moved from an era of observation to the era of molecular biology, scientific questions, methods, analyses, and interpretations became increasingly conflated across the basic-applied spectrum. Clearly, both ends of the spectrum advance knowledge: basic investigation informs our understanding of pathobiology, and translational studies of disease mechanisms inform our understanding of basic biology. Examples of this latter point abound and have led to the New England Journal of Medicine series, “Basic Implications of Clinical Observations” (5, 6). The Wall Street Journal contributor and author, Matt Ridley, has taken this perspective one step further and argued that basic scientific advances can be the consequence, rather than the cause, of applied technological advances (innovation) (7) (e.g., cryoelectron microscopy was developed to limit the consequences of radiation damage for biologic specimens and of structural collapse by dehydration under a vacuum; with the solution to these practical problems came a dramatic expansion of the field of structural biology, now to include high-resolution images of complex macromolecular structures that defied analysis by conventional X-ray crystallography and diffraction, and time-resolved changes in macromolecular structures or intermolecular interactions). Interpreted most generously, these examples illustrate that basic biomedical research and translational biomedical research have been coevolving successfully into a seamless continuum of investigation.

Given the diversity of questions and model systems being investigated within individual fields, can we identify criteria that might be used to facilitate labeling specific research activities as basic or translational? If so, this might clarify public discourse and enhance communication within the scientific community and between the scientific and lay communities.

POTENTIAL CRITERIA FOR CONSIDERING RESEARCH AS BASIC VS. TRANSLATIONAL

The identity of the institution and department in which the research is performed

At most medical schools, many faculty members are members of what are institutionally denoted basic science departments, such as cell biology, genetics, biochemistry, and neurobiology, among others. Many other faculty members are based in school-affiliated hospitals and within departments in which the names reflect clinical fields, such as medicine, pediatrics, surgery, and neurology, among others. These organizational distinctions might suggest that faculty in basic science departments conduct basic research, whereas those in clinical departments, at least in the main, conduct applied translational or clinical research. But that is not always the case. In biomedical research today, much investigation takes place in academic health centers (or hospitals), and much of this work lies within clinical departments, such as medicine, pediatrics, and neurology. In some such departments, most of the research pursued is clinical research on human subjects, much of it involving the testing of therapies or devices. In other clinical departments, including those at our Harvard-affiliated institutions, research spans a broad array of topics, from general cellular mechanisms to disease mechanisms, and such research may also use organisms from worms and flies to mice and, of course, humans. Many researchers in these departments pursue research as a full-time or nearly full-time endeavor, many are not physicians, and substantial numbers might fit just as well, based on the work they do and where they publish it, in traditional basic science departments. For these reasons, we should not categorize research as being basic or translational based on the identity of the institution or department in which it is performed.

The motivation of the investigator

Should research qualify as basic because an investigator pursues a question purely for reasons of curiosity, without any interest in the potential practical applications of the work? Likewise, should research qualify as translational because an investigator is pursuing the solution to a practical biomedical problem, such as the treatment of a disease? Perhaps surprisingly, these differences in applicability or practical purpose are common distinctions used to define the following terms: “basic research” is conducted without any practical end in mind, although it may have unexpected results pointing to practical applications (1), whereas, “translational research” applies scientific observations to practical questions on the human condition.

Although many scientists choose to pursue particular questions (e.g., how a complex organism develops from a single cell, what molecular interactions determine cell division or death, how one nerve cell communicates with another, etc.), because they see them as challenging puzzles without consideration of practical applications, it seems unhelpful to label the research as basic or translational solely on the basis of the motivations of the scientists regarding potential, practical impact. A scientist may be motivated to pursue the same research question because of curiosity as to how the world works or because of an intuition that the understanding of a particular pathway or mechanism might lead to an understanding of and potential treatment for a disease. For example, one might study cell death pathways as a challenging biologic problem with relevance to developmental biology or because of a suspicion that such pathways might be relevant to cancer. Likewise, studies of adipocyte-specific gene expression in cell culture might be pursued as a means to understand regulated gene expression, or be motivated by a desire ultimately to understand obesity. Would such practical motivations on the part of the latter scientist render their inquiries less basic? We think not. Although research called basic today, often lacks a consciously applied, practical motivation, that seems not to be an essential reason to label the inquiry as basic. Likewise, research motivated by a desire to understand and eventually facilitate treatment of a disease should not, for that reason alone, be considered translational or less basic. Rather, one or more characteristics of the research itself should determine how it is categorized. This is distinct from the issue, raised by some observers, that current grant proposals force investigators to claim the potential impact of their work on future clinical application before such links are reasonable to assert (8).

The scientific importance of the work

Some discoveries, by virtue of their powerful capacity to change the way we think about an area, are more important and have greater generalizability and impact than others. Whereas some scientific observations provide new, explanatory frameworks and affect many unanticipated areas of inquiry, others provide more limited understanding, with little or no impact on other fields. The term basic could be used to describe research that is fundamental in this way. If this kind of importance or impact is what we mean when referring to research as being basic, then the term can be applied to many different kinds of research, from the purely theoretical to that dealing with events at the atomic, molecular, cellular, organismic, or population level. Likewise, by the criterion of scientific importance, basic research can involve organisms ranging from bacteria, to worms, to mice, or to humans, and some research on humans can readily be judged as more basic (i.e., more important) than some research on worms. Furthermore, molecular discoveries about a specific pathway that are novel but add incrementally to the field and lack impact outside it may be judged less basic in this sense than human studies, revealing new insights about a previously unexplained physiologic response or disease.

Thus, if the term basic is used to refer to research that is scientifically important, then disease-related research, wherever it is carried out, can be more (or less) basic than research conducted in basic science departments. At the very least, this analysis would suggest that the term basic is being used in different ways in these various situations.

Whether the research is disease related and/or conducted on humans or human tissues

Historically, research in basic science departments has not prominently involved human disease models or subjects (there are exceptions), and this fact (which could have evolved differently and might still in the future), combined with the fact that much human and/or disease-related research is conducted in hospitals, has created a cultural divide. In this context, it might not be surprising that research on humans would be viewed by some as less basic than research typically not involving humans or their tissues carried out in basic science departments. But does this idea, or related constructs, have merit? Research on human subjects or tissues may be basic, based on criteria of scientific importance, having great impact on understanding human biology and disease and leading to a fundamental understanding of general biologic principles. Of course, most researchers who conduct human studies are motivated by or are aware of potential applications to disease. Is this a sound basis for applying the term translational to the research activity rather than viewing it as basic? Perhaps so, provided that the use lacks hidden implications of relative scientific value. As departments redefine their scope of interests (e.g., departments of physiology focusing on whole-organ genomics in humans), these distinctions of purpose, application, and relative value truly become moot.

Beginning about a decade ago, the field of translational research was redefined and re-emphasized as the ultimate formalism for all biomedical research, regardless of the nature (purely basic or translational) of the initial scientific question posed. The intention of redefining the field of translation was both a noble one, emphasizing, as it does, our academic community’s efforts to apply fundamental observations to the human condition, as well as a practical one, serving to emphasize to our legislators, in an era of fiscal constraint, that even basic research funded by the U.S. National Institutes of Health (NIH) has (or may have) practical applications that may benefit human health. Yet, it appears that this type of nomenclature has had an unintended, divisive effect within the academic biomedical community. Why? The newfound emphasis on the translation of basic observations to the clinical arena evolved in the setting of a severely cost-constrained NIH budget and applied yet another criterion for the use of limited government funds (viz., the potential for clinical application of even the most fundamental finding). This mandate was established in the setting of an expanded pool of biomedical, newly minted Ph.D. researchers through the doubling of the NIH budget in the early part of this century. As an increased number of Ph.D.s began to seek support from a significantly constrained pool of NIH funds, they found that they also now needed to justify their proposals based on potential translation to clinical application, no matter the topic. As a result, growing concern arose in the basic biomedical community about society’s commitment to support research without express practical purpose. For those members of the community who have always believed, if not insisted, on the age-old hierarchical distinctions between basic and applied biomedical research, the ascendancy of translational medicine gave them pause. Furthermore, this distinction has now led some members of the basic community to argue for the need to enhance support for basic studies so that fundamental scientific investigation not be lost in translation (9). The push for (rapid) clinical application is viewed by some as overvaluing translation (10) and, furthermore, has led to conflation of the more clinical end of the translational research spectrum with the more fundamental end of that spectrum (11). Thus, this transition has had the unintended consequence of promoting the age-old, hierarchical distinctions between intellectually and methodologically equivalent research strategies in domains called basic and translational research, inapplicable, though they now are, for the reasons delineated above.

Thus, in a most unfortunate way, the excessive emphasis on translational research has led to a reversion to outdated approaches as to how biomedical investigators relate to one another. We feel strongly that there is no place for these artificial distinctions in today’s biomedical research enterprise. They are intellectually limiting and in the worst case, create artificial barriers to sharing ideas and resources.

Summary interpretation of bases for categorizations

Based on the foregoing analysis, we think it is fair to conclude that use of the terms basic or translational biomedical research should not be dependent on the following criteria: 1) whether the scientist pursuing it is motivated by a desire for the work to have practical impact; 2) the scientific importance of the work; 3) the level of inquiry (e.g., atomic, molecular, cellular, physiologic, community); or 4) the name of the department in which the work is pursued. Yet, the question remains: is there an unambiguous use of the terms that can be widely accepted and, most important, advance the cause of the scientific enterprise?

The most meaningful distinctions that arise when describing and categorizing biomedical research relate to two specific issues: the precise question being asked and the approach used to address the question. The scientific questions can be broad in scope, ranging from studies of single molecules to populations of human subjects and from normal mechanisms to mechanisms of disease pathogenesis. Likewise, the methodological approaches applied can be equally broad, ranging from structural studies of single molecules to statistical genetic analysis of variant allele frequencies in populations of human subjects, with or without a disease phenotype. Each line of inquiry and experimental strategy can produce scientific results that are more or less important, with their importance only reliably ascertainable with the passage of time; can be motivated by a desire to produce practical/applied results or not; and can be performed by individuals conducting research in what we now call basic or clinical departments. Thus, the formation of distinctions between basic and translational research is not especially useful in the current era. This distinction is an anachronism that can best be appreciated by understanding its historical origins.

HISTORICAL AND CULTURAL PERSPECTIVES UNDERLYING THE CATEGORIZATION OF BIOMEDICAL RESEARCH

For much of history, and, importantly, in 19th century Europe, a bright line was drawn between fundamental scientific inquiry and applied research. Whereas the former was carried out by professors in universities, the latter was often conducted in industry. Some of the greatest scientists and discoveries arose at the interface between fundamental and applied research—witness the discoveries of Pasteur, Langmuir, and those related to atomic energy. In prior years, when medicine had an extremely circumscribed, scientific basis, the limited research that existed was conducted by university professors, whereas clinicians pursued their work in a relative scientific vacuum. As biomedical science developed in the 20th century and especially following World War II, with the growth of research within academic health centers and hospitals, the line between basic and applied biomedical research, as defined in the preceding century, became much less relevant.

Over the past 50 years, the growth of funding for biomedical research and the opportunities created by research advances in cell and molecular biology stimulated clinical departments to expand their research operations. This growth was especially robust during periods with clinical financial surpluses and increasing levels of NIH funding. Both M.D. and Ph.D. researchers were recruited into clinical departments, and research techniques used there progressively included approaches similar or identical to those used in basic science departments, such as gene cloning, transgenic animal studies, etc. Whereas, much of this work might also have been situated within traditional basic science departments of medical schools and universities, many hospitals and clinical departments aggressively pursued it as well, although the faculty typically remained associated with departments bearing clinical names. Depending on the school or university, such faculty may or may not also have appointments in basic science departments.

The role of the biotechnology and modern pharmaceutical industries in the historical evolution of biomedical research categorization also warrants comment. With the passage of the Bayh-Dole Act and the explosion of biotechnology and pharmaceutical industry interactions with academic institutions, another aspect of basic research must be reconsidered. Scientists who might previously have viewed their work as unrelated to practical outcomes (e.g., fundamentals of cloning and eukaryotic gene expression) realized the potential for their discoveries to be applied to human therapeutics, and many of these basic scientists turned their attention aggressively to pursuit of these goals, including moving into the commercial sphere. In this way, another barrier between basic and translational research collapsed. Among faculty in basic science departments who pursue research in cell biology, genetics, or neuroscience, some continue to pay little attention to the practical implications of their work. Others, doing identical or substantially similar work, are constantly on the hunt for the practical implications of their studies that might be patented and exploited, within or outside of the academy, for eventual diagnostic and therapeutic uses. Whereas the former scientists might prefer that their work be described as basic (and devoid of practical use), the latter might proudly describe their (substantially similar) work as having translational or even clinical implications.

CONCLUSIONS

In the end, does it matter whether we call some biomedical research basic or translational? If we understand what the particular research is about and grasp its content, context, and implications, probably not. However, current use promotes confusion and hampers efforts to promote scientific understanding and collaborations across our diverse, creative academic biomedical research enterprise. We should clarify or modify the use of these terms and consider a new, cohesive classification nomenclature that realistically reflects the contemporary biomedical research environment. The descriptive analysis of biomedical research should focus on perceived quality as objectively assessed as possible (12), rather than the promotion of a self-affirming, anachronistic distinction. Perhaps, instead of use of the term basic to describe particular entities of biomedical research, our descriptive terminology should reflect one of the alternate meanings of the term, namely, essential or vital. These terms could apply to all biomedical research that is effective in creating new knowledge, because such research is essential for the growth of the biomedical enterprise for the benefit of society, either directly or remotely. The elimination of the unnecessary distinction between basic and translational biomedical research (and researchers) would improve the effectiveness of the scientific process and its potential benefits for the society to which its practitioners are ultimately obligated.

Discovery is optimall promoted by reliance on talented investigators, whatever their philosophical scientific persuasion (13). Particularly in an era of big data and interdisciplinary research, a nonhierarchical atmosphere of collegiality and mutual support is required for optimal success. The mandate that all research be framed as offering a potential for translation to clinical benefit is misguided, misleading, and disingenuous. Legislators should be educated about the nature, natural history, and goals of modern biomedical research, wherein anyone with a good idea and the skills necessary to realize it has the potential to add value to the enterprise and benefit to society.

AUTHOR CONTRIBUTIONS

J. Flier and J. Loscalzo both conceived of the ideas presented, and jointly wrote and edited this article.