The innovation journey of genomics and asthma research
Address for correspondence: Lise Bitsch, Department of Science, Technology and Policy Studies, University of Twente, Drienerlolaan 5, PO Box 217, Enschede 7500 AE, The Netherlands e-mail: firstname.lastname@example.org
Abstract This article concerns the transformative potential of medical genomics for common disease research. We analysed 13 review articles in asthma research in the period 1999 to 2008. Our aim was to understand how genomics has emerged in this research field, and the attendant changes. Motivated by Lippman’s geneticisation thesis, we use the concept of an ‘innovation journey’ to trace how expectations of improved understanding, prevention, diagnosis and treatment structure a dynamic co-evolutionary process through which a genome-based discourse emerges. We show how the asthma researchers involved continuously struggle to define their contribution to asthma research, as well as to clinical practice. Along the way, the researchers propose changes to both the definition and the aetiological model of asthma, thus highlighting gene–gene and gene–environment interactions. It is, however, difficult to characterise this discourse as one of geneticisation. With increasing attention being given to epigenetics, metabolomics, proteomics and systems biology, the emerging picture suggests an important, but much less deterministic, role for genes.
Genomics and asthma research
Genomics, often illustrated with reference to the Human Genome Project (HGP), promises a transformation of our understanding, treatment and prevention of common multifactorial conditions. The aim of this article is to explore the transformative potential of genomics for common disease research, by mapping the changing ways in which genomics and asthma have discursively been framed in research on asthma. The genetisicisation thesis introduced by Abby Lippman (1991, 1992) is an influential theory shaping debates on the transformative potential of genetic discourse and practice (see, for example, Hedgecoe 1998, 2001, 2004, Kerr 2004). The thesis asserts a growing tendency to define disorders and behaviour as genetic and to distinguish between people on the basis of genetics. The concept of geneticisation has been instrumental in initiating a debate in social science on the transformative potential of genetics.1 However, it has also been criticised for allowing only negative interpretations of geneticisation and for having a weak empirical basis (Hedgecoe 1998, 2001). Although the rise of genetic discourses for common conditions has been demonstrated in the literature (Cox and Starzomski 2004, Hall 2004, 2005, Hedgecoe 2001, 2002, Lock 2005, Weiner and Martin 2008), the way in which these discourses have been influenced by the introduction of genomics is little documented.
In a detailed empirical study of the presence of genetic discourse in the field of coronary heart disease, Weiner and Martin (2008) have shown that, in addition to what could be seen as deterministic genetic discourses, several other discourses co-exist in this field. In our analysis of the asthma field we show that several discourses can be found not only to co-exist at a certain moment in time, but also co-evolve over a period of time. This is demonstrated through an analysis of how, over time, genomics and asthma are framed in a variety of ways. Specifically, we analyse how asthma researchers link up with genomics with the expectation that it will contribute to understanding a genetic component of asthma. Our analysis shows the emergence of a clear-cut genomic agenda for studying the genetic component of asthma. However, this agenda does not emerge in a vacuum. Lacking a better definition, asthma-genome researchers use the current definition of asthma in their work while simultaneously criticising it as being problematic for genome-based studies. Existing knowledge is thus, however reluctantly, part of the creation of new knowledge. Over time a gradual movement takes place, from efforts to frame genetic knowledge as an addition to existing hypotheses of the aetiology of the disease, to framing the exploration of a genetic component of asthma through genome-based methods as a true reconstitution of the field. Consequently, the group of researchers involved in genome-based asthma research produce suggestions that may change the very definition and explanatory models of asthma. It would be somewhat premature to predict that the outcome, if they are successful, will result in a prioritisation of genetic explanations of asthma cause and development. Instead, a complex process of co-evolution between many different scientific epistemologies, all involved in the exploration of asthma, is taking place.
The process undertaken by a group of asthma researchers in creating expectations and agendas linking asthma and genomics is conceptualised as an innovation journey. Following Rip and Schot (2002), an innovation journey represents ‘the underlying phenomena of emerging path-dependencies’ (Rip and Schot 2002: 157). An emerging innovation journey thus reflects increasing stabilisation. Emerging expectations and agendas guide and provide direction and roles to the asthma researchers involved in genome-based asthma research (Brown et al. 2000, Van Lente 1993). They constitute signposts for an emerging innovation journey. We have identified these initial signs of the innovation journey of asthma and genomics through an analysis of 13 scientific review articles. We argue, that while the introduction of genomics has not resulted in geneticisation as such, genomics has become part of an ‘omics’ based, complex, multi-level approach to asthma. The goal towards which these research efforts are directed is the identification of (susceptible) subgroups at which new clinical applications can be directed. Such a development implies new forms of disease management based on surveillance (screening) and increased individual responsibility for health. However, genetics will only be one factor in this development and perhaps not even a determining one at that.
Expectations and agendas as the first signs of an emerging innovation journey
Originally, the concept of innovation journey was developed through an analysis of innovation processes in firms (Van de Ven et al. 1999). Rip and Schot (2002) developed the concept from a sociological perspective. In their conceptualisation, innovation journeys are non-linear processes that are part of the overall co-evolutionary process between technology and society. The central dynamic in the progress of an innovation journey is the emergence of path-dependency, signifying direction and increasing stabilisation (Rip and Schot 2002). Path-dependencies are the outcome of the actions of actors taking part in the innovation journey. During the lifecycle of an innovation journey, different groups of actors may join in or jump off the ‘train of innovation’. At the very early stages of innovation journeys, expectations of future outcomes are central in guiding actors’ imaginations and activities (Borup et al. 2006, Brown et al. 2000, Van Lente 1993). Actors will create so-called prospective structures, which they will then attempt to fill in through dedicated efforts (Van Lente and Rip 1998).
Collective expectations represent prospective structures to which large and diverse groups of actors subscribe (Konrad 2006, 2010). They can exert pressure and invite uptake in agendas through a number of mechanisms. One such mechanism is a rhetorical strategy in which an emerging development is compared with an already stabilised structure. As a result, collective expectations take the form of ‘perceived established structures’ (Konrad 2006: 434), although they are clearly not structures in the sense of organisational or material structure. Collective expectations can also govern actors through ‘image pressure’ (Konrad 2006: 434). Actors consciously relate to the expectations of others, and thus subscribe to collective expectations in order to live up to (imagined) peer expectations. Finally, by virtue of their interpretive flexibility, collective expectations play an important role in cycles of hype and disappointment. As long as collective expectations are held in high regard, the plausibility and function of the promised innovation tend to be evaluated in a positive light. However, if such collective expectations begin to de-stabilise, outcomes are judged in a negative light. Unless new resources are found, and collective expectations can again be strengthened, this often implies the end of the innovation journey (Konrad 2006, Van de Ven et al. 1999).
These three mechanisms are clearly illustrated in the way in which researchers who aim to link asthma and genomics base their expectations on finding genes for asthma on past discoveries of major genes using ‘traditional’ approaches, their collective references to the HGP and their shifting evaluative criteria of the success of these efforts as the innovation journey proceeds. The comparison with past achievements provides the justification for exploring a genetic component to asthma as well as a research structure. The HGP functions as a common collective reference for the expectations of others with which the participants would like to align. Finally, the participants manage to keep the specifications of the eventual outcome of the innovation journey vague enough so as to incorporate the finding that many genes, together with environmental influences, present a critical challenge to their initial visions of opportunities for predicting susceptibility to asthma as well as to develop ways of preventing asthma. We can thus trace emerging innovation journeys by following the process in which actors produce and subscribe to collective expectations and agendas.
In order to trace this process of emerging stabilisation, we build on a method developed in a 2005 article by Van Merkerk and Van Lente. These authors argue that ‘emerging irreversibilities’ are central to understanding the eventual build-up of a technological path in (nano) technology developments. They conceptualise emerging irreversibilities as those elements that make it easier to do something or ‘make it more difficult (or less easy) for actors to do something else’ (Van Merkerk and Van Lente 2005: 1096). Further they show how such irreversibilities can be identified on the basis of statements relating to expectations and agendas in a body of literature selected from the field of study (Van Merkerk and Van Lente 2005). Accordingly, for the purposes of our study we used a particular body of literature, defined by the links made in it between asthma and genomics, to trace an emerging genomics innovation journey in asthma research. Our analysis is a first step towards getting to grips with the implications of efforts to link asthma and genomics, for asthma research in particular, but also for asthma clinical practice more generally.
A historical narrative of the meeting of genomics and asthma
A number of authors have analysed genetic discourse in disease research using journal accounts (see for example, Hedgecoe 2001, 2002, 2003, Kerr 2000, Weiner and Martin 2008). The aim of our analysis is to trace a dynamic process denoting an innovation journey of the genomics and asthma in asthma research. In particular, we explore whether and how an emerging irreversibility is created in terms of how a genetic component to asthma should be researched.
Review articles belong to a particular group of academic articles, since they represent an interpretive effort on the part of the authors to create an overview of the past, present and future of a scientific field (Myers 1990, 1991). In review articles authors highlight what they see as the history, that is, the past, present and future of their field. Review articles are therefore particularly helpful in providing an insight into how discursive links between specific expectations and agendas are created. In sum, they are collections of attempts to create and maintain collective expectations, as well as summaries of past and future agendas. Following this changing dynamic therefore provides insight into how researchers give shape to the relationship between genomics and asthma and thereby the innovation journey. In the following analysis, we describe how genomics and asthma were variously framed in three periods of time. The analysis is based on 13 review articles, which were found through a bibliographical search2 in the ISI Web of Knowledge using the key words ‘asthma OR bronchial hyperresponsiveness’ AND ‘genome OR genomics’. The first reviews matching these criteria appear in 1999. The review articles on which the analysis is based thus derive from a large dataset generated within the predefined categories of the ISI Web of Knowledge. The initial set contained 131 reviews. The main criteria used for narrowing down our set was that the review appeared in a high-ranking journal, had a general title and belonged to one of the four ISI categories dominant in the set. These four categories were immunology, allergy, genetics and heredity and respiratory research. In reaching our final selection we consulted a recognised expert in asthma genetics research. The possible bias introduced by such a consultation was minimised by the simultaneous application of our own selection criteria. The reviews are listed in Appendix 1 according to year of publication.
In addition to the reviews, 10 one-hour semi-structured interviews were held with asthma and genome researchers. Extracts of the interview material are included here, as we draw on it in our concluding section. An important finding from our analysis of the review articles centres on how the reference to clinical impact and importance plays a role in discursively framing asthma and genomics. Such findings per se say little about actual contributions to or changes in clinical practice or how expectations to clinical impacts are perceived outside the realm of the review articles. However, with reference to the interview material, we show that (some) framings of genomics and asthma in the review articles seem to create a distance rather than a closer connection to clinical practice and concerns. We revisit this discussion in the concluding section.
The innovation journey of genomics and asthma research
Rising expectations (1999)
In 1999 ongoing efforts taking place under the heading of the HGP constitute the common reference point for collectively shared expectations that genes for asthma can actually be found (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999):
The Human Genome Project is working hard to sequence the human genome and will probably detect new candidate genes implicated in the pathogenesis and severity of asthma. The identification of asthma susceptibility genes will probably provide more insight into the identification of individuals at risk of asthma in the near future and open the way for early prevention. (Los et al. 1999: 1222)
Results from the HGP project are expected to greatly aid asthma researchers in their efforts to unravel and understand the genetic component of asthma. While there is a long-established agreement among asthma researchers that a genetic component must play a role in the aetiology of asthma, it is the flow of information from the HGP which raised expectations that genes for asthma can actually be found. Simultaneously, the authors of the reviews conceive of a future healthcare structure in which more specific knowledge about genes plays an important role. Within this future structure, the focus is on the identification of susceptible individuals and the optimisation of diagnosis and treatment (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999). The model that is conceived for the future thus contains a novel element in terms of prevention, but it also adds to established healthcare practices of diagnosis and treatment:
Genetic studies of asthma continue to provide insight into the pathophysiological mechanisms of the disease. This should ultimately lead to new and more effective therapeutic interventions, new diagnostic tools for presymptomatic diagnosis, the development of strategies for disease prevention in susceptible individuals, and delineation of the interaction between genotype and the response to specific treatments (pharmacogenetics). (Wiesch et al. 1999: 900)
In the above extract, genetic studies of asthma are linked to continuous efforts to understand the pathophysiological mechanisms of the disease. By framing genetic studies as an activity that has already contributed to the knowledge base of asthma, these studies take the form of ‘perceived established structures’, which create a sense of stability and continual progress. In turn, this move justifies collective expectations of the way in which growing insights in the genetic component of asthma will lead to the ultimate aim of disease prevention and the optimisation of diagnosis and treatment.
The link to ongoing efforts is laid down by presenting genetic factors as the key to solving principal and vexing questions in current asthma research. In particular, asthma researchers are struggling to understand the environmental factors that may explain the development of asthma, as well as the question as to why only some develop asthma in response to the presence of these factors (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999). In this context, genetics is presented in the reviews as an essential element contributing to the established understanding of the importance of environmental factors:
The widely accepted paradigm is that environmental factors are important to the development of asthma, but one must be genetically predisposed to respond to environmental differences. (Los et al. 1999: 1210)
Accordingly, the reviews not only present and discuss a list of the most relevant environmental factors that contribute to the aetiology of asthma, but also contribute evidence for its hereditariness, and thus for its genetic component on the basis of observations made as long ago as 1650. Asthma, however, does not seem to be transferable from parent to child in a straightforward Mendelian manner (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999). The search for a genetic component to asthma is thus presented both as a new challenge and as part of established research efforts in asthma research. In taking up this challenge, two different agendas are sketched in the reviews. One suggestion by Los et al. (1999) is that already established methods in studying the genetic component to asthma will be appropriate:
The latter (twin studies) will be particularly helpful in studying the relation of the genetic and environmental variances contributing to a trait. The classical twin study, the parental twin design and the cotwinn control study are suitable for this purpose. (Los et al. 1999: 1223)
On a second agenda is the claim that a new direction has to be taken with regard to the basis of emerging technology, although it might prove a more uncertain pathway:
Chip and array technology already allows the expression of thousands of genes to be measured simultaneously. ... Genome-wide association studies by the typing of multiple SNPs [single-nucleotide-polymorphisms] have been proposed as a more powerful method of detecting genetic effects than linkage mapping. The number of SNPs to be typed in such an exercise may exceed 500,000. This is just feasible with current technology but might be prohibitively expensive. (Moffatt and Cookson 1999: 608)
As genomics and asthma were linked together for the first time, collective expectations of improved understanding, prevention, diagnosis and treatment of asthma were introduced with reference to the HGP. Care was taken to connect these expectations to the established structures of received knowledge on the importance of environmental factors and on continuous efforts of understanding the pathophysiological mechanisms of asthma. However, a prospective structure was also introduced, with a focus on the identification of susceptible individuals with the aim of preventing the disease and optimising its diagnosis and treatment. Two items for the further exploration of a genetic component of asthma were also suggested, and thus filling out the prospective structure. One was to follow known methods, while the other was to break with traditional methods in order to apply new, more risky technologies. The latter proposal to implies the formation of a new branch on the innovation journey of asthma research. It suggests a move away from characterising phenotypes on the basis of observational features like bronchial hyper-responsiveness and towards identifying phenotypes on the basis of genomic information. In the next section, we show how this branch continued to be shaped as asthma researchers framed the connection of genomics and asthma as an opportunity to discover novel pathways of the disease.
A promise is reinforced (2002–2004)
In the reviews from the period 2002–2004, the exploration of a genetic component to asthma was formulated for the first time as contributing towards a profound change in received understandings of asthma. This way of framing the exploration of a genetic component to asthma is connected to the connected to the data from the HGP as well as the successful identification of genes for asthma. In addition, the explicit attention paid to environmental factors in the 1999 reviews has disappeared during this period. Instead, ongoing efforts in asthma research were framed in terms of discovering novel pathways of disease, understood as regulated through the model of gene–gene and gene-environment interactions.
As data from the HGP arrived, it appears that there were far fewer genes than originally expected (Vercelli 2002). This challenged ideas that there was a linear relationship between genes and the expression of a trait like asthma. For asthma, this ‘unprecedented flood of data’ (Vercelli 2002: 15) led to the hypothesis that, in addition to genes having a more subtle and complex influence on the disease, gene–gene and gene–environment interactions were of central concern. Contrary to previous reviews, environmental factors were no longer explicitly presented or discussed. Instead they were now factored into sentences such as:
More sophisticated analytical approaches that identify gene–gene and gene–environment interactions on a genome-wide level are required to fully elucidate the genetic risk factors for asthma and atopy, and to understand the pathogenesis of these common conditions.(Hoffjan and Ober 2002: 714)
In conclusion, we envision a scenario in which a constellation of small quantitative variants in critical genes fine-tunes innate and adaptive immune responses and the way in which they interface with the environment, resulting in a wide spectrum of phenotypes.(Vercelli 2002: 19)
The hypothesis of subtle and complex gene–gene and gene–environment interactions signifies a shift away from a research agenda primarily concerned with adding to existing knowledge on the importance of environmental factors. Rather than describing environmental factors in their own right, they are reduced to factors that can be understood through the study of genes (Kauffmann 2004, Vercelli 2002, Wills-Karp and Ewart 2004). The inclusion of traditional environmental factors in the models of gene–environment interaction creates a perceived established structure for the exploration of this model as it adds a sense of contributing to past research. However, the implications of interpreting the findings of the HGP and the identification of genes in terms of interactive models led to an increased emphasis on the space between the genotype and the phenotype and hypothesis-free research strategies (genome-wide scans). The identification and interpretation of the first genes for asthma therefore represent a decisive moment for genomics in asthma research. Finding actual genes is interpreted as proof that a genetic component to asthma exists, but is also clashes with the understanding of the linear relationship between genes and the environment implied by earlier research. The outcome of this tension is to frame this finding as introducing a new era in asthma genetics research, leading to radically changed understandings of the disease:
Although great strides have been made in the past decade, the next decade will almost certainly yield pronounced changes in our understanding of asthma susceptibility that, it is hoped, will translate into improved diagnosis, prevention and therapeutic strategies for this ever-increasing disease. With the recent successes, we might predict a time in the not too distant future when these strategies will be tailored to the individual on the basis of their genotype at a few key loci and their unique environmental exposure history.(Wills-Karp and Ewart 2004: 386)
The interpretation of the identification of asthma genes as the first step towards ‘pronounced changes in our understanding of asthma susceptibility’ opens the way for so-called ‘hypothesis-free’ research methodologies. Because the genes are seen as pointing to novel insights, researchers scramble to find a research methodology with which this novelty can be explored and understood. Available methodologies suffer from two critical shortcomings: they are either problematic in terms of producing statistically significant results or they are not suitable for discovering novel pathways and genes (Hoffjan and Ober 2002, Kauffmann et al. 2004, Vercelli 2002, Wills-Karp and Ewart 2004).
On the one hand then, the identification of asthma genes has allowed for an enormous boost of confidence and a strengthening of the prospective structure implied in expectations to future clinical improvements of better therapy, diagnosis and prevention. On the other hand, this way of framing the contribution of genetics to asthma implies the opening up of the space between the genotype and the phenotype. It also focuses on gene-centred explanations of asthma. Instead of environmental factors acting as central touchstones in explaining asthma cause and development, genes have become the central explanatory unit in terms of which environmental factors are considered. The resulting challenge now lies in connecting these two levels with each other.
A new research agenda (2006–2008)
The 2008 reviews saw the emergence of a new research agenda for pursuing the prospective healthcare structure of improved understanding, diagnosis, treatment and options for preventing asthma. Framing the genetic component in asthma as a matter of gene–gene and gene–environment interactions has led to a demand for improved means of exploring the connections between the genotype and then phenotype. The genome-wide association study (GWAS) methodology was introduced as a solution to this challenge. However, it has become increasingly clear that the effect of genes on asthma is minor and that no single gene or set of genes is strongly predictive of asthma. This prompts a reframing of the role of genes in interactive models, which opens the way for additional ‘omic’ approaches, in addition to systems biology and epigenetics.
Following the identification of the first genes for asthma, many regions thought to harbour asthma genes were identified, but, as the list of susceptibility genes expands, new challenges emerge. The challenge for the asthma researchers involves elucidating the role of the suspected genes and their interaction with each other and the environment. Researchers also find it difficult to replicate and validate their findings of susceptibility genes and none of the susceptibility genes can be said to be strongly predictive for asthma (Ober and Hoffjan 2006). This challenges the expectation of being able to predict who is susceptible to developing asthma in order to improve its prevention and treatment. Instead, the hypothesis of more genes with small effects has been strengthened:
On the basis of this review we suggest that the total number of genes that contribute to risk may exceed 100 and that the individual effect of any of these genes on disease effect is quite small. (Ober and Hoffjan 2006: 98)
So far, the search for a genetic component to asthma has been driven by the expectation that the HGP will facilitate the discovery of genes, and once genes had been found, improved understanding, diagnosis, treatment and the possibility of prevention through the identification of susceptible individuals would follow. Seeing that the effect of the genes suspected to be involved in asthma is likely to be ‘quite small’, asthma researchers have to figure out how to interpret these findings. They could signal an end to the expectations of future applications for predicting asthma. Instead, a new cycle of collective expectations has been introduced with the emergence of a new research methodology of GWAS. One example of this in an introduction in the review articles is the following remark:
The year 2007 has witnessed a quantum leap in genotype and phenotype association analyses with the publication of the first GWA study for an asthma trait (childhood asthma). (Vercelli 2008: 178)
A heading in a work by Moffatt (2008) introduces GWAS thus: ‘New approaches and the future for disease gene identification’ (Moffatt 2008: 414). This method is moulded as the solution to a specific problem: the exploration of an association between the genotype (the genetic make-up of an organism) and the phenotype (its physical expression). Furthermore, the method introduces a specific collaborative structure:
This is an exciting time to be studying the genetics of asthma, with large-scale, collaborative, whole-genome association studies under way in the United States and Europe. These are expected to yield results within the next year, and epidemiologists may quickly be able to translate genetic associations into a better classification of the disease. In addition to this, further advances are already occurring with the application of genome-wide expression to understand complex disease. Add to this developing areas of proteomics, metagenomics and metabolomics and the so called ‘systems biology’ approach, researchers in the field of complex disease genetics will find themselves deluged with an abundance of relevant data. This ultimately will result in a deep understanding of the aetiology of previously mysterious diseases such as asthma, leading to the real potential of prevention and cures. (Moffatt 2008: 416)
The GWAS method does not stand alone: proteomics, metagenomics, metabolomics and systems biology are lined up to complement it. All these approaches are directed at bringing interacting systems and their products in the body into view. Vercelli (2008) adds epigenetics as a promising new agenda in addition to GWAS. Characteristic of all these approaches is the turning of attention away from genes as a determining factor in these systems.
The above approaches have been introduced as ‘quantum leaps’ that will finally allow a ‘deep understanding’ of the ‘previously mysterious’ condition of asthma and lead to ‘the real potential of prevention and cures’. Thus, the connection between asthma and genomics is made through a discourse implying a break with previous understandings and approaches to asthma. The new innovation journey is thus taking shape through a number of emerging irreversibilities. There is growing confidence about asthma genetics as a mature field, with a history of past achievements pointing to a promising future. GWAS was introduced as a method marking a new era in which links between asthma and genetics may be established in more elaborate and concrete ways. However, this link must be made through complex interactive systems consisting of many different factors. The shape of the ‘claims of transformation’ of this prospective structure – improved treatment, diagnosis and prevention – remains a moving target throughout the journey. As the prospective structure of the innovation journey is maintained by the introduction of GWAS and accompanying ‘omic’ methods and epigenetics, asthma is reframed as a mysterious condition in which knowledge of genes and other complex internal mechanisms are an essential aspect. It is therefore now too simple to portray what is going on as geneticisation. Instead, asthma genetics researchers have created a new (gen)omic space in asthma research where the exploration of the space between genotype and phenotype is modelled on the interactive model of gene–gene and gene–environment interactions. With this reframing of the genetic component of asthma research as one that should be explored on a genome-wide, and even on multiple ‘omic’ levels, the innovation journey is taking a new twist. In the following section, we elaborate on the shape of the conceptual emerging irreversibility in the changing definition as well as on the aetiological model of asthma.
Transforming the field of asthma research
Genomics has taken shape in asthma research through the proliferation of collective expectations on improved understanding, prevention, diagnosis and treatment. Specific expectations of the possibility of finding genes related to asthma were fulfilled in the period 2002–2004. Some aspects, however, did not develop as expected. Instead of a few major genes involved in the development of asthma, a yet-to-be-determined number of genes involved in gene–gene and gene–environment interactions have emerged as explanatory factors (Anderson 2008, Koppelman et al. 2008, Moffatt 2008, Vercelli 2008). Asthma genome researchers went from shaping their research agenda in terms of contributing to existing knowledge to that of creating new knowledge. This implied hypothesis-free investigations and an opening for the GWAS method, together with epigenetics, metabolomics, proteomics and systems biology. In the preceding sections we followed this innovation journey by tracing the dynamic process in which links were created between asthma and genomics. How does this emerging innovation journey relate to other ongoing endeavours in asthma research? Our analysis does not necessarily imply that gene-centred research is transforming asthma research in general, or that clinical practice is changing through the incorporation of genetic information. What we do show in the above sections is how expectations of future changes in clinical practice play a role in the discursive strategies of creating prospective structures. This group of researchers view their own efforts as making a potentially major contribution to reconstituting the general field of asthma research by a redefinition of the aetiological model of asthma. Indeed, a wider uptake of such research results would imply the beginning of a genome-based transformation in asthma research.
In the 1999 reviews asthma was mainly described in terms of clinical manifestations. It was defined by the obstruction of the airway and chronic inflammation and subdivided into early-onset and late-onset asthma, allergic asthma, asthma without evidence of allergy, occupational and exercise-induced asthma. Patient questionnaires and clinical observations were usually used to establish the diagnosis of the disease. Asthma could be directly observed via the obstructed airway and the causal explanation referred to the clinical phenomenon of airway inflammation (Anderson and Cookson 1999, Los, Koppelman, and Postma 1999, Moffatt and Cookson 1999, Wiesch et al. 1999).
However, for researchers interested in finding genes this definition was problematic. A common complaint centres on the difficulty of defining the asthmatic phenotype. The definition, with its focus on the obstructed airway and the role of inflammation, is criticised for being too general; it mixes together different clinical entities. The diagnostic methods are too subjective, as they are left to the assessment of individual doctors and patients’ own judgment. Intermediate markers (also called intermediate phenotypes) are introduced as more objective units for measurement (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999). These intermediate markers are imagined in closer and even direct, causal connection with the genes searched for.
In reviews from 2008 asthma was redefined as a condition resulting not from inflammation and an obstructed airway but from unknown mechanisms of gene–gene and gene–environment interactions (Anderson 2008, Koppelman et al. 2008, Moffatt 2008, Vercelli 2008). Clinical characteristics were still used to describe the condition but no longer featured as the cause of asthma. The subtle change in the definition of asthma implies a major shift in research attention. While the obstructed airway and inflammation were key to understanding asthma according to the 1999 definition, gene–gene and gene–environment interactions were at the centre of such understanding in 2008.
Towards a new aetiological model of asthma
In addition to these attempts at a redefinition of asthma, the researchers in the innovation journey of asthma and genomics also challenged the aetiological model of asthma. In the 1999 reviews, researchers refer to two environmentally based hypotheses of the aetiology of asthma: the allergen and the hygiene hypothesis. A western lifestyle is generally argued to account for the rise in the observed incidence of asthma. According to the allergen hypothesis, exposure to allergens such as house dust mites, smoke and diet is responsible for sensitising individuals to an extent that asthma results (Anderson and Cookson 1999, Los, Koppelman, and Postma 1999, Moffatt and Cookson 1999, Wiesch et al. 1999). Similarly, the hygiene hypothesis suggests that, due to heightened hygiene standards, the adaptive immune response does not receive adequate stimulus to develop properly. As a consequence it overreacts when exposed to allergens, and asthma results (Wiesch et al. 1999). An explanation is, however, lacking as to why only some people develop asthma. Thus, in the 1999 reviews, genetic susceptibility was highlighted as a determining factor in triggering responses to these environmental factors. Genetic susceptibility is double-faced, as it can both increase and decrease the sensitivity of the individual (Anderson and Cookson 1999, Los et al. 1999, Moffatt and Cookson 1999, Wiesch et al. 1999).
As more and more susceptibility genes for asthma were found, a so-called endophenotype model developed. This model is based on an understanding of a non-linear pathway from genotype to phenotype. Groups of susceptibility genes, found in various physiological systems in the body, are imagined to lead to a broad range of asthmas (Anderson 2008, Vercelli 2008). These different endophenotypes each represents a subset of the disease, with a discrete pathogenic pathway connected to a group of genes. This model is more complex than the clear-cut model of the gene–environment interaction, with differentiated internal pathways interacting amongst themselves and the environment. In this construction, attention is also drawn to the role of epigenetics (Vercelli 2008). In the reviews, a shift has taken place from talking about asthma to talking about asthmas. As in other cases, the identification of genes (mutations in genes) has led to a fragmentation of the definition of asthma (Rabeharisoa and Bourret 2009). In addition, the proposal an aetiological model of asthma based on complex non-linear relations has also fragmented asthma into asthmas and hence seemingly endless opportunities for sub-characterisations of the disease. This reframing of the aetiological models of asthma is central to understanding the necessity of shaping the link between asthma and (gen)omics as activities aimed at unravelling novel disease pathways and creating novel knowledge about asthma, now framed as a mysterious disease. It is an aetiological model essentially different from models such as the hygiene or allergen hypothesis because it takes molecular biological mechanisms as its reference for explaining phenotypic expressions. In the exploration of a genetic component to asthma, it represents an emerging irreversibility, as aetiological models (like the hygiene hypothesis) and methods of classification (like questionnaires) are framed as both dogmatic and imprecise.
Our analysis shows the emergence of an explicitly genomic agenda for studying the genetic component of asthma. Expectations of acquiring improved understanding and the prevention, diagnosis and treatment of asthma provide a stable prospective future for the activities of the asthma researchers over time. The agenda for realising this future changed from finding major genes and contributing to existing knowledge to discovering many genes with small effects, gene–gene and gene–environment interactions and creating novel knowledge. As the journey progresses, it is increasingly shaped by a number of emerging irreversibilities. The discovery of genes for asthma increases confidence in a genetic research direction for asthma. The emergence of new genome-wide methods for exploring the genetic component to asthma introduces expectations of profound changes in the understanding of asthma. In addition, epigenetics, metabolomics, proteomics and systems biology imply that the focus is moving away from genes as the sole factor of interest.
The co-evolutionary character of this process is illustrated through efforts of, on the one hand, incorporating existing definitions and aetiological models of asthma into the search for a genetic component, and on the other hand, attempting to change the definition and aetiological models. The emerging definition emphasises gene–gene and gene–environment interactions as causal factors. This change could be interpreted as prioritising genetic explanations, in that it focuses attention away from environment and the organ level. However, the accompanying emerging aetiological model focuses on non-linear discrete pathways, leading to a diversity of asthmas, and the necessity of understanding cellular as well as molecular processes. Furthermore these pathways are conceived as being intimately connected to environmental stimuli.
Returning to the transformative potential of genomics for asthma research and its implications for the thesis of geneticisation, it is clear that the researchers who take part in genome-based explorations of asthma research aim to transform asthma research in general. Their expectation is that a profound change will result from (gen)omic explorations of a genetic component to asthma. However, this framing is not uncontested or unproblematic. In 10 semi-structured one-hour interviews with asthma and genome researchers in 2009 and 2010, we encountered uncertainty as well as direct opposition to genome-based research as the solution to developing prevention, improved diagnosis and treatment for asthma. The bone of contention is the very prospective structure of the clinical value of the knowledge created through genome-based research. The issue was framed as a communication challenge between ‘people in the lab’ and clinicians:
You should always try to get to stay in contact with each other, but it’s a constant threat that the lab people are doing things that are not relevant for daily practice, and the clinicians do not want to hear about it or do not understand it, and there is autonomous development of research looking for things that might not really be the most relevant things. For as long as I remember that has been a problem for clinical departments. (Asthma6 2009)3
This accusation levied at laboratory researchers operating outside actual clinical interests came after the interviewee had discussed how research on patients’ behaviour and interaction with patients would be more beneficial to improving asthma clinical practice than genomics research aimed at discovering the biological mechanisms of asthma. This perceived missing link to practice formed the basis of another interviewee’s critical comments:
I think that genomics may become useful in certain diseases, but in highly complex diseases, such as asthma, where there are, well, many, many genes involved, and many environmental influences, and many interactions between genes and these environmental influences, I think it will be impossible to find a batch of genomic testing that will reliably identify one phenotype or another. (Asthma9 2010)
Here, the key to interpreting genomics as contributing to a profound change is not its ability to produce data and so-called basic knowledge but its ability to contribute to practice. Evaluations of the prospective structure of clinical changes in terms of improved diagnosis, therapy and possibilities of prevention, on which (gen)omic investigations of asthma rely, show a tension between this prospective structure and the agenda of genome-based methodologies as the means to fill out this structure. Rabeharisoa and Bourret (2009) examined clinical practice for cancer genetics and psychiatric genetics and identified the introduction of genes, as a possible element in arriving at a diagnosis, to have changed the arenas for clinical work, as well as its content and object and finally, the way in which clinical work is carried out. However, for asthma, no clinical space exists for medical genetics. These findings point to an asynchronous development between asthma research and clinical practice, and raise an important question, for us, as well as for the asthma researchers. How should we view the fact that genomics has not had immediate clinical implications?
In this article we open up the complex co-evolutionary process through which genomics is taking shape in asthma research. By conceptualising this development as an innovation journey, we draw attention to the non-linear character of this journey, the choices that have to be made, and the emerging irreversibilities that will eventually add up to a path-dependency for the interpretation and use of genomics in research on asthma. Further research should investigate this process for other areas of complex disease research, in order to understand how genomics is taking shape there.
1 The ISI Web of Knowledge lists 229 citations that cite Lippman (1991) or Lippman (1992). Search performed on 16 July 2010.
2 The bibliographical search was carried out in the spring of 2009.
3 ‘Asthma’ refers to the interviewee as part of the asthma researchers interviewed, and the number refers to the order in which the interviewees were interviewed.
We thank Dr Gerard Koppelman, University Medical Center Groningen, The Netherlands, for his generous support and invaluable help in helping us get our bearings on the innovation journey of asthma and genomics. This research is part of a PhD project supported by the Dutch Centre for Society and Genomics. A draft version of this article was presented at the workshop ‘Technoscientific and Social Dynamics of Health and Healthcare’, 8–9 October, 2010, as well as at the international conference ‘Tentative governance in Emerging Science and Technology’, 28–9 October 2010 at the University of Twente, The Netherlands.
Appendix: Review papers
|1999||Anderson, G.G. and Cookson, W.O.C.M.||Recent advances in the genetics of allergy and asthma|| Molecular Medicine Today ||Indexed in biochemistry and molecular biology, cell biology and medicine, research and experimental|
|1999||Los, H., Koppelman, G.H. and Postma, D.S.||The importance of genetic influences in asthma|| European Respiratory Research ||3/34 in respiratory research|
|1999||Moffatt, M.F. and Cookson, W.O.C.M.||Genetics of asthma and inflammation: the status|| Current Opinion in Immunology ||8/119 in immunology|
|1999||Wiesch, D., Meyers, D.A. and Bleecker, E.R.||Genetics of asthma|| Journal of Allergy and Clinical Immunology ||1/17 in allergy and 9/119 in immunology|
|2002||Hoffjan, S. and Ober, C.||Present status on the genetic studies of asthma|| Current Opinion in Immunology ||8/119 in immunology|
|2002||Vercelli, D.||The functional genomics of CD14 and its role in IgE responses: an integrated review|| Journal of Allergy and Clinical Immunology ||1/17 in allergy and 9/119 in immunology|
|2004||Kaufmann, F. and the Post Humane Genome Respiratory Epidemiology Group||Post-genome respiratory epidemiology: a multi-disciplinary challenge|| European Respiratory Journal ||3/34 in respiratory research|
|2004||Wills-Karp, M. and Ewart, S.L.||Time to draw breath: asthma susceptibility genes have been found|| Nature Reviews. Genetics ||2/132 in genetics and heredity|
|2006||Ober, C. and Hoffjan, S.||Asthma genetics 2006: the long and winding road to gene discovery|| Genes and Immunity ||28/119 in immunology and 37/132 in genes and heredity|
|2008||Anderson, G.P.||Endotyping asthma: new insights into key pathogenic mechanisms in a complex, heterogeneous disease|| Lancet ||2/100 in medicine, general and internal|
|2008||Koppelman, G.H., Te Meerman, G.J. and Postma, D.S.||Genetic testing for asthma|| European Respiratory Journal ||3/34 in respiratory research|
|2008||Moffatt, M.F.||Genes in asthma: new genes and new ways|| Current Opinion in Allergy and Clinical Immunology ||Indexed in allergy and immunology|
|2008||Vercelli, D.||Discovering susceptibility genes for asthma and allergy|| Nature Reviews. Immunology ||2/119 in immunology|