Personalized medicine - Future Medicine

Perspective For reprint orders, please contact: [email protected]

’Personalized medicine’: what’s in a name?

Over the last decade genomics and other molecular biosciences have enabled new capabilities that, according to many, have the potential to revolutionize medicine and healthcare. These developments have been associated with a range of terminologies, including ‘precision’, ‘personalized’, ‘individualized’ and ‘stratified’ medicine. In this article, based on a literature review, we examine how the terms have arisen and their various meanings and definitions. We discuss the impact of the new technologies on disease classification, prevention and management. We suggest that although genomics and molecular biosciences will undoubtedly greatly enhance the power of medicine, they will not lead to a conceptually new paradigm of medical care. What is new is the portfolio of modern tools that medicine and healthcare can use for better targeted approaches to health and disease management, and the sociopolitical contexts within which these tools are applied.

Anna Pokorska-Bocci*,1, Alison Stewart1, Gurdeep S Sagoo1, Alison Hall1, Mark Kroese1 & Hilary Burton1 PHG Foundation, 2 Worts Causeway, Cambridge, CB1 8RN, UK *Author for correspondence: Tel.: +44 01223 761 900 anna.pokorska-bocci@ phgfoundation.org 1

Keywords: genomics • individualized medicine • patient empowerment • personalized medicine • precision medicine • stratified medicine

Genomics has led to new capabilities that many claim are revolutionizing the practice of medicine and healthcare by enabling diagnosis and disease management to be more accurately targeted at the individual patient. A variety of terms have been used to describe this concept: personalized medicine, stratified medicine, individualized medicine, precision medicine and so on. Such terms have attempted to define or describe these new dimensions of practice, and around them have been built programs of research, funding streams, new journals and even whole institutes, as well as new clinical practice guidelines and new business models in the healthcare industry. Although this article is only intended to be exploratory in its objective, to contextualize these terms and their conceptual meanings, we have followed a systematic approach to literature searching, reporting and analysis. The review of the published literature has allowed the capture of these concepts as proposed by various authors and allowed a detailed nar-

10.2217/PME.13.107 © PGH Foundation

rative synthesis and understanding of the origins of these terms. We discuss whether genomic and other biomedical knowledge and technologies are, as some have claimed or implied, leading to a conceptually different approach to medicine – one that is encapsulated in terms such as ‘personalized medicine’ – or whether their contribution is to expand and refine the tools available for the diagnosis, management and prevention of disease but without changing the fundamental paradigm of clinical practice. Methods Two databases, MEDLINE (via PubMed) and EMBASE, were searched by two independent reviewers on 8 May 2012. The search strategy involved the following terms: ‘stratified medicine’, ‘personalis(z)ed medicine’, ‘precision medicine’, ‘P4 medicine’, ‘personalis(z)ed healthcare’, ‘individualis(z) ed medicine’ AND ‘stratified healthcare’. The phrase ‘stratified healthcare’ was not found in either database.

Pers. Med. (2014) 11(2), 197–210

part of

ISSN 1741-0541

197

Perspective Pokorska-Bocci, Stewart, Sagoo, Hall, Kroese & Burton Three selections were then performed, based first on the publication titles and then on abstracts and applying exclusion criteria. The criteria used and the justification are presented in Box 1. The selection criteria were applied in order to identify the most critical and influential publications although it was impossible to rule out the possibility of bias as selections relied on some subjective judgements. The risk of bias was minimized by the incorporation of influential reports from the grey literature where possible and appropriate. The searches of both databases were then combined, and, after elimination of duplicate publications, resulted in 524 articles, which were briefly screened. A further selection based on perceived relevance to the discussion of ‘What is personalized medicine?’ led to a final list of 52 articles, which were then used for detailed analysis and to inform the main discussion of this paper. The rationale behind this final selection was to eliminate papers duplicating original opinion papers and to choose publications describing original definitions, fundamental concepts and influential commentaries. The terms ‘pharmacogenomics’ or ‘genomic medicine’ were not included in the search. This may result in a bias against the identification of early publications

containing concepts relating to personalized medicine that did not explicitly use our search terms. Following initial analysis of selected papers, results were incorporated into a narrative with particular focus on the key characteristics and major implications of each term. Results A total of 7194 papers were identified by the search strategy before the application of the exclusion criteria. Analysis of search results by year of publication showed an exponential increase in the number of publications for all search terms from the year 2000 (Figure 1) . For individual search terms, Table 1 shows that personalized medicine/healthcare is by far the most commonly used term followed by individualized medicine, with stratified medicine and P4 medicine having emerged later and being less commonly discussed. Scope, content & contexts for terms in the personalized medicine arena

Using our shortlist of 52 key articles, we looked in more detail at the scope, contexts and content of the discussion for the six different terms describing this ‘novel’ medicine: individualized medicine, precision

Box 1. Exclusion criteria and their justification. Exclusion 1 based on titles • Specific therapy for specific condition • We tried to discuss general concepts and not specific examples of therapies

Alternative medicine • We concentrated on mainstream medicine and healthcare

Patents • The technical aspects of patents, although informative, were deemed less relevant for the conceptual discussion

Languages other than English, French, Italian or Polish • Only languages read by authors were included for pragmatic reasons and resource limitation. Less than 5% of the overall articles identified (318 of 7194) were non-English and we note that several of the authors of these articles subsequently also published in the English language • Exclusion 2 and 3 criteria: further selections of papers were performed in order to retain the most informative and influential papers. Given that the article is very conceptual in its attempt, publications describing specific technical aspects of therapies and technologies were discarded at this stage

Exclusion 2 based on titles • Clinical trials • Informatics tools • Purely sequencing papers • Personalized therapy for a specific condition • Titles with no mention of personalized/stratified/individualized/precision/P4 medicine and its variations (but enigmatic titles were retained) • Purely pharmacogenetics/pharmacogenomics

Exclusion 3 based on abstracts • Experimental strategies/targeted therapies/taxonomy • Specifically on next generation sequencing • Industrial/business models • Biomarkers • Genetic counseling

198

Pers. Med. (2014) 11(2)

future science group


Perspective

3000

2516 2500

Publications (n)

2000 1643 1500

1000

809

451

500

0

1

1

1

1

1

0

10

16

47

19

65

91

129

198

302

1971 1976 1979 1990 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 Year

Figure 1. Number of publications per year 1971–2011.

medicine, stratified medicine, P4 medicine, personalized medicine and personalized healthcare. Table 2 provides a summary of the definitions introduced by the main proponents and the context in which the term was introduced. Individualized medicine

Individualized medicine was first mentioned in the published literature in 2003, when it was used to refer to individual drug metabolism in the context

of pharmacogenomics [3] . It was subsequently used by commentators interchangeably with ‘personalized medicine’ in describing the application of knowledge of genetic differences and physiological information to improve diagnostics and better tailor treatment to an individual patient’s needs [12,13] . More recently a second context was introduced, referring to therapeutic approaches that utilize an individual’s own cellular material to develop a treatment that is unique for the patient from whom the

Table 1. Number of publications and year of first appearance. Search term

Total publications (n) †

First publication year

‘Precision medicine’

30

1997 (1979 but relating to acupuncture)

‘Personalis(z)ed medicine’

5925

1999 (1971 but relating to family doctors’ relationship with patients)

‘Personalis(z)ed healthcare’

114

2000

‘Individualis(z)ed medicine’

2161

2003 (1976 but relating to doctor’s dialog with each patient)

‘Stratified medicine’

53

2007

‘P4 medicine’

17

2008

Some publications include information on multiple search terms.


www.futuremedicine.com

199

Perspective Pokorska-Bocci, Stewart, Sagoo, Hall, Kroese & Burton

Table 2. Summary of main definitions for included terms. Term

Main definition

Conceptual introduction or discussion of this definition

Individualized medicine

Therapies involving patient’s own cells in diseases where there is no efficient drug treatment available, e.g., stem cell therapies and cancer vaccines

Baker [1] , Gravitz [2] Clinical research

Precision medicine

Integration of molecular research with clinical data from individual patients to develop a more accurate molecular taxonomy of diseases to enhance diagnosis and treatment and tailor disease management

US National Academy of Sciences report [4]

National Academy Wasi [5] of Sciences

Stratified medicine Matching therapies with specific patient population characteristics using clinical biomarkers

Trusheim et al. [6]

Industry

Trusheim et al. [6]

P4 medicine

Clinical application of tools and strategies of systems biology and medicine to “quantify wellness and demystify disease for the well-being of an individual”

Hood [7]

Systems biology approach

Hood [7]

Personalized medicine

Application of genomic and molecular data to better target the delivery of healthcare, facilitate the discovery and clinical testing of new products, and help determine a person’s predisposition to a particular disease or condition

Abrahams et al. [8]

Definition of Personalized Medicine Coalition, USA

Langreth and Waldholz [9]

Personalized healthcare

Tailoring of medical management and patient care to the individual characteristics of each patient

Teng et al. [10]

Personalized healthcare white paper, USA

Abidi and Goh [11]

material has been derived. The contexts for these therapies included stem cell therapies [1] , cancer vaccines [2] and complex cancer types with individually different molecular profiles [14,15] . These therapies have the potential to revolutionize treatment of particularly challenging clinical conditions where no effective treatment is currently available. However, they are far from being ready for acceptance by regulators and for routine adoption in the clinic. Precision medicine

The term precision medicine was used by Boguski et al. in 2009 and was described as having three essential features: an understanding of what causes a disease; the ability to detect the presence of these causal agents/elements; and the ability to treat the root causes effectively [16] . In 2011, a more specific meaning was coined by the National Research Council of the US National Academies in their report ‘Toward precision medicine’ [4] . The report focused on reclassification of disease based on molecular data and proposed a framework

200

Pers. Med. (2014) 11(2)

Context

First publication referring to the term Srivastava [3]

for new data networks that would integrate molecular research with clinical data from individual patients. The report recognized the potential of genomics but also of other emerging new technologies in the investigation of molecular features of disease and the establishment of a new, more accurate taxonomy of illnesses. The scientific work that led to this report had been going on for many years. As early as 1999, for example, the National Cancer Institute Director put forward a challenge to the scientific community to harness the power of new molecular analysis technologies to move the basis of tumors classification from morphological to molecular [17] . Other commentators developed the discourse around precision medicine further. Mirnezami et al. acknowledged that state-of-the-art molecular profiling had to be used together with established clinical–pathological indices to create diagnostic, prognostic and therapeutic strategies precisely tailored to an individual patient’s requirements [18] . Robinson introduced the concept of ‘deep’ phenotyping, by which he meant precise and comprehensive analy-



sis of phenotypic abnormalities [19] . He postulated that all medically relevant disease subclassifications could be considered as having different phenotypes and recommended that analysis and exchange of phenotypic data were prerequisites for the development of precision medicine. Stratified medicine

The term ‘stratified medicine’ was introduced in 2007 by Trusheim and colleagues [6] , who defined it as being “where therapies are matched with specific patient population characteristics using clinical biomarkers”. They suggested the term ‘stratified medicine’ on the basis that each patient can be matched to a cohort (or subpopulation) exhibiting a specific therapeutic response, by using a biomarker specifically associated with that response. The result of the biomarker test would then indicate a preferred treatment for the patient subpopulation. A good example of such a clinical biomarker is the BCR-ABL-positive tyrosine kinase genotype used to select patients with chronic myeloid leukemia who are likely to respond to imatinib (Gleevec®), an inhibitor of this kinase [6] . Stratified medicine was thus said to be dependent on the development of medically acceptable clinical biomarkers and the availability of multiple treatment options. The clinical biomarker could be one of a number of different types including a genotype, a biochemical biomarker, a physiological biomarker or an imaging biomarker. In the UK Blair et al. recognized the crucial role of genomic medicine in the development of stratified medicine for cancer but also noted the important contribution of other scientific disciplines, including imaging techniques, especially in disease areas such as lung disease, infectious diseases or neurodegenerative disorders [20] . The best examples of stratified medicine to date can be found in oncology [21] , with the development of cancer drugs tailored to specific molecular targets rather than clinical types of the disease. Thus, identification of key molecular changes in tumors can define a subpopulation with a high probability of response to a particular treatment. This approach is also emerging in the fields of inflammatory disorders, cardiovascular disease and mental disorders. For most commentators stratified medicine combines a biomarker, diagnostic test, treatment and health outcome into one linear pathway to be applied to a well-defined subpopulation. P4 medicine

The term ‘P4 medicine’ was introduced by Leroy Hood in 2008 based on the development of systems


Perspective

biology [7] . Standing for predictive, personalized, preventive and participatory, the term P4 medicine denoted a paradigm shift in the practice of healthcare from reactive to preventive. Hood claimed that systems biology, in which the analysis of relationships in an entire system in response to genetic or environmental perturbations is undertaken, would enable a more holistic approach. Firstly, molecular insights into disease mechanisms would allow stratification of complex diseases and new types of drug discovery. Secondly, increasing availability of information through the internet and the development of devices that measure personal information would enable more active participation of patients and consumers in their healthcare: they would then become participatory rather than passive recipients of medical advice. Hood and Flores postulated that in the future everyone will have their genome sequenced and reviewed yearly for new actionable variants; organspecific blood proteins will be analyzed throughout life; and the systems approach to the immune system will enable the prediction of potential future occurrences of ‘disease-perturbed networks’ in patients, and therefore render the whole of medicine more preventive [22] . Overall they suggested that the practice of medicine will become personalized in the common sense of the term: each person will be treated as a unique individual and individuals will “serve as their own controls” to determine their bodies’ transition from health to disease. Other commentators add a ‘fifth P’ to the equation – but not always the same one. Pravettoni and Gorini discuss P5 in the context of cancer medicine and the fifth P refers to the psychocognitive aspects to be considered in the personal profile of a patient in order to recognize him not only as a biological and genetic entity but also as a person with specific needs, values, hopes and fears [23] . Then, recently it has been argued by Khoury et al. that in order for P4 medicine to fulfill its promise, the ‘population perspective’ – the ‘fifth P’ – must be integrated into each of the other four components [24] . Personalized medicine

The term ‘personalized medicine’ appeared in literature as early as 1971 in a paper by Gibson [25] . It discussed the changing role and place of the family doctor in the modern world of advanced science and healthcare delivered by teams of highly specialized experts. Gibson reminded his audience about the art of medicine and importance of the personalized approach. He warned about modern medicine seeing a patient more as a ‘condition’ than as a


201

Perspective Pokorska-Bocci, Stewart, Sagoo, Hall, Kroese & Burton human being, and advocated that the physician of the future should retain “the old traditions of the service and trust” supplemented by current scientific (and technological) tools. When ‘personalized medicine’ began to reappear in the literature 30 years later it referred to the application of pharmacogenomics in clinical practice (e.g., Gupta et al. [26]). Since then, the term personalized medicine has covered a wide variety of concepts, although most authors use it in the context of “human genome deciphering and the resulting sudden increase in the understanding of disease causes and therapeutic options” (e.g., Burke and Psaty [27]). As the most commonly used term (see Figure 1), personalized medicine seems to have emerged as an umbrella term for the whole concept. In 2005, the Personalized Medicine Coalition was formed in USA and defined personalized medicine as “the application of genomic and molecular data to better target the delivery of healthcare, facilitate the discovery and clinical testing of new products, and help determine a person’s predisposition to a particular disease or condition” [8] . Many commentators emphasized the potential of the genomic revolution to lead to a much more refined molecular classification of diseases [28] . For example, Moon et al. defined personalized medicine as “the use of genomic signatures of patients in a target population for assignment of more effective therapies as well as better diagnosis and earlier interventions that might prevent or delay disease” [29] . More recently and encompassing the more holistic lifelong approach, Meyer envisioned personalized medicine broadly as a comprehensive, prospective approach to preventing, diagnosing and treating diseases in order to achieve the optimal result for an individual [30] . He distinguished four stages of personalized medicine: assessment of an individual’s disease risk to allow early diagnosis and/or preventative measures; increase of diagnostic precision by better defining diseases and describing phenotype; tailoring of treatment to the individual characteristics of each patient with breakthrough molecular diagnostic tools and pharmacogenomics advances; and evaluation of objective and subjective clinical outcomes and tangible benefits to the individual patient. Hong and Oh also described four components of personalized medicine but from a more practical point of view [31] . They looked at the type of tools a physician requires to practice personalized medicine: standard health risk assessment tools to evaluate the likelihood of an individual developing a disease; family health history described as a complex combination of shared genetic, environmental and lifestyle risk fac-

202

Pers. Med. (2014) 11(2)

tors; integration of information on the genome and its ‘derivatives’ such as the transcriptome, proteome and metabolome; and clinical decision-support systems. Another group of commentators have discussed personalized medicine in a wider context, going beyond technologies and genetic diversity, and taking into account a number of additional data management, societal and practical aspects [32–34] . Fierz conceptualized personalized medicine in six dimensions: the set of individual characteristics of disease and associated personal risk profiles; the infectious environment and its varying properties; genes and molecular traits; drug development and pharmacogenetics; the healthcare process including patient education, privacy and regulatory issues, and patient empowerment; and information and data management [35] . Harvey summarized the range of current approaches to personalized medicine expressed by a number of key opinion leaders interviewed by the European Science Foundation (see also European Science Foundation project report [36,37] . Some focused on science and new molecular understanding paving the way to more targeted prediction, prevention and treatment of illness. For others, the driving force behind personalized medicine was the current sociopolitical context and ideals of personalization that lead to new paradigms in healthcare revolving around the needs and wants of an individual. Finally, another dimension was discussed by Cribb and Owens, who addressed the role of personalization within the welfare state [38] . They explored the political context of personalization and discussed its place on UK health policy agendas, warning against using personalization as a convenient inclusive label for many concepts such as consumer and citizen power, and as an easy response to a range of critiques of the welfare state. They noted that personalization can mean different things to different audiences and addressed this by distinguishing between personalized medicine, “which deals with the solid scientific business of adapting medicine to individual needs” and is responsive to biological needs of individuals patients, and personalized healthcare, which “deals with the humanistic and fuzzier business of catering to people’s preferences about those things that surround medical interventions, such as appointment times, hospital or consultation styles”. According to those commentators, the development of personalized medicine reflects that, even though the basic needs of patients are still treated as universal (which is compatible with a continuing emphasis upon maintaining equitable access to the same services within a welfare state), there is sufficient variation in clinical need between individuals to justify a range of responses.



Personalized healthcare

The term ‘personalized healthcare’ appeared in the analyzed literature in early 2000s in the context of new information tools and platforms. Abidi et al. focused on customized healthcare information available through the internet as the principal tool of patient empowerment [11,39] . Zhang et al. provided a definition of a new global healthcare system that “links healthcare service providers to an individual’s personal and physical spaces” [40] . Jang et al. discussed the use of ‘intelligent gadgets’ that would be wearable, reconfigurable, scalable platforms processing information, in personalization of healthcare [41] . Other commentators have used the term interchangeably with other terms to cover concepts of genetics-based interventions, personal genomics services and their role in the tailoring of patient care [10,42] and in the context of global systems biology [43] . Simmons et al. bring many of these ideas together, defining personalized healthcare as a “coordinated strategic approach to patient care” that broadly encompasses P4 medicine and uses technological and other means to deliver care across the health continuum from promotion and prevention to detection and treatment of disease [44] . A critical feature of this is the personalized healthcare plan, which is customized for each individual. They argue that the rational introduction of such personalized healthcare across the entire system will require the creation of infrastructure, validation, regulation and reimbursement mechanisms. They warn that personalized healthcare should not be equated with genomic medicine, and suggest that many elements of personalized healthcare can be introduced into practice now, while most genomic applications still await the accumulation of sufficient supporting evidence. Our results regarding these key concepts and how they relate to each other can be summarized as shown in Figure 2. Discussion Many terms are used to describe the new advances and approaches in medicine and healthcare. Our literature survey shows that ‘personalized medicine’ is the most often used and that it is an umbrella term covering many concepts. Since ‘personalized medicine’ was first introduced, many other terms have been coined, but although originally covering specific, well-defined ideas, their meanings have become looser and even interchangeable over time. Two examples are ‘stratified medicine’ and ‘precision medicine’. ‘Stratified medicine’ was initially used in commercial settings to define medical achievements where therapies are matched with specific patient pop-


Perspective

ulation characteristics using clinical biomarkers. Over time it has started to appear on many policy agendas and has been used more holistically, describing the endeavor to take account of individual characteristics and needs when implementing clinical research and healthcare reforms. In the minds of many, the boundaries between ‘stratified’ and ‘personalized’ medicine have become very blurred, which could mean that the original focus on specific health developments may be diluted or lost. ‘Precision medicine’ was coined by the National Research Council of the US National Academies in their report entitled ‘Toward precision medicine’. In this report the term ‘precision’ is used to mean both ‘accurate’ and ‘precise’, and describes the potential development of new tailored treatments based on integration of molecular and clinical research data and on a molecular reclassification of diseases. The term has been widely used since, and very often interchangeably with ‘personalized medicine’. We would argue, however, that accuracy and precision in diagnosis, and the molecular characterization of diseases, constitutes only a subset of the issues contained within the term ‘personalized medicine’. The situation becomes even more confusing when commentators use ‘individualized medicine’ to describe any endeavor to cater to specific health and medical needs of a single individual. Although this usage accords with the general semantic meaning of the term, when originally formulated it was much narrower in focus, applying specifically to therapies based on a person’s own cells or tissues. ‘P4 medicine’ is another interesting concept that initially described an almost visionary approach to ‘wellness’ and that again subsequently began to be used interchangeably with ‘personalized medicine’. Our analysis suggests that the term ‘personalized medicine’ has emerged as most prevalent in recent years. It has been used in many different applications; indeed, the range of its definitions has widened and changed over the last decade. It is a term that has been used (and is still used by some commentators) in a narrow way as synonymous with the application of pharmacogenomics data in healthcare, particularly in the early 2000s following the explosion of newly available genomic information. At the other end of the spectrum, the term encompasses a broad range of medical, scientific, technological and sociopsychological aspects affecting current and future medicine and healthcare. The breadth and composition of the concept varies extensively and depends on the type of audience and type of healthcare and political agenda for which it is used. Given this lack of consistency in understanding the breadth and focus of ‘personalized


203


Individualized medicine Using patients’ own cells Precision medicine Molecular taxonomy of diseases P4 medicine Predictive, preventive, personalized and participatory

Stratified medicine Biomarkers + diagnostics + drugs

Personalized medicine Genomics + information technology + patient empowerment

Personalized healthcare Tailoring of medical management and patient care

Figure 2. How the key concepts of personalized healthcare, personalized medicine, stratified medicine, precision medicine, individualized medicine and P4 medicine all relate to one another.

medicine’, it becomes crucial to define the term when used in a particular context and for particular purposes. We would advocate using ‘personalized medicine’ in the broad sense of the term as it is understood today by most audiences, and to keep the usage of the other terms for more specific purposes. However, for the purposes of policy development, it is very important that there is clarity about the nature and scope of the definition of ‘personalized medicine’ being used in a particular context and by a particular audience. Use of terminology to imply novelty

The literature search confirmed that the burgeoning discourses around personalized medicine occurred in parallel with the emergence of genomic knowledge and technologies, and considerations of how these would be used. We noted that discussions of personalized medicine, as well as the use of other terms, often reflected commentators’ attempts to categorize the new capabilities that had been enabled by genomics, and make claims about the added value that genomics could bring, while also acknowledging and integrating the impact of other innovations. Based on these descriptions and augmented by our own knowledge and experience, it is important to examine what the contribution of genomics has been and whether genomics has brought anything fun-

204

Pers. Med. (2014) 11(2)

damentally new to medicine. We re-examined some of the concepts advocated under the general rubric of personalized medicine in relation to processes of disease classification, diagnosis, prevention, clinical management and the role that genomic and other new technologies might play. The changing classification & diagnosis of disease

The earliest classifications of disease were based on gross anatomical and physiological signs and symptoms. Developments in microscopy and biochemistry added to and in some cases changed these classifications by introducing information gained from observation and measurements at the cellular and subcellular level. Genomics and other molecular biosciences are now further extending our ability to classify disease to a more fundamental, basic biological level that in some cases is related to an underlying, causal molecular lesion. In the last decade for example, genomic approaches have led to the definition of an increasing number of inherited cardiac conditions and genetic tests are now available for many [45] . Groups of cardiac arrhythmias can be classified by symptomatology and characteristic ECG changes and, increasingly, can be further subdivided according to the particular pathogenic mutations (e.g., genes associated with potas-



sium or sodium ion channels in long QT syndrome). Knowledge of the underlying mutation may be important for prognosis, management and identification of at-risk family members. In many cases genomic knowledge has led to a new taxonomy of disease, with disease states being classified according to a detailed description of clinical and pathological features (so-called ‘deep phenotyping’) coupled with state-of-the-art molecular profiling. In some cases, molecular information may add discrimination to existing clinical classifications, while in others it may lead to new classifications altogether. For example, a detailed genetic mapping of breast tumors has shown recently that this cancer is better considered as a group of ten different diseases according to which genes are mutated, overexpressed or silenced. Those results provide a novel molecular subdivision of the breast cancer population, derived from the impact of somatic mutations and copy number aberrations on the transcriptome [46,47] . Genomics does not, however, change the fundamental raison d’être of classification, which is to enable clinical diagnosis: the attempt to match the symptoms and signs of an individual patient to a specific category within the classification. In other words, clinical diagnosis remains a process of placing an individual within a population subgroup with whom they share a specific set of clinical features. The term ‘precision medicine’ may be helpful in indicating that the basis for defining the diagnostic subgroup will be more finely grained and more closely related to underlying molecular pathology, but it runs the risk of raising expectations by implying complete certainty in diagnosis – a level of ‘precision’ that is unlikely to be attained in any foreseeable future. Disease management

As with diagnosis, information from genomics and other biomedical technologies is increasingly being used to inform decisions about the clinical management of disease. The terms ‘personalized medicine’ and ‘individualized medicine’ (Table 2) imply that it will, in the future, be increasingly possible to provide a treatment and management regime that will be optimal for a specific individual. However, except in cases where the treatment has been designed specifically for one patient, for example, by using his or her own cells, this claim may be overstated. Clinical management decisions are made on the basis of accumulated evidence about the response of populations or subpopulations of patients to different therapeutic regimes. These subpopulations are often related partly to the diagnostic/taxonomic categories discussed above, but also depend on other features of


Perspective

the patient such as age, sex, race, comorbidities, BMI, comedications, treatment preferences and likelihood of treatment adherence. We argue that it is unlikely that the conceptual basis of this clinical management decision-making process will change in the genomic era: management decisions will still be taken on the basis of attempting to match an individual to a population subgroup that behaves similarly clinically and whose average response to a treatment is known. However, the nature and amount of information used to define the subgroup may change radically, and in some cases the introduction of genomic information may supersede previously used characteristics, for example race. The genetic association between HLA-B*5701 and adverse drug effect for abacavir (Ziagen®) is a good example of such a case [48] . Clinical management may also be modified in the light of initial treatment response. Management of some cancers provides examples where genomic technologies will improve the ability to track treatment responses and disease progression. For example, next-generation sequencing technologies allow identification of cell-free cancer DNA, which is used as a biomarker to quantify the disease burden and monitor response to treatment [49,50] . Here, again, there is no paradigm shift, but the facilitation of more finely tuned disease management with improved clinical outcomes. The incorporation of genomic information will lead to an increase in the number of different management subgroups and the size of the population subgroups used as reference might change but there is likely to be a limit to this process of differentiation: the minimum size of a subgroup will be determined by the feasibility of establishing a sufficient evidence base for the response of that subgroup to treatment, as we discuss further below. However, there are examples of conditions where clinical decision-making processes will be highly influenced by genomic technologies and this process of subgroup differentiation might go much further than in traditional diagnostics and disease management. Examples include some very rare conditions and intermediate disease markers in inherited metabolic disorders where genomic technologies might allow clinical decisions to proceed in the absence of a sufficient evidence base concerning average response to treatment, provided that interventions are associated with possible positive health outcomes outweighing an acceptable level of risk [2,51–52] . Emergence of ‘stratified’ medicine

As noted above the emergence of the term ‘stratified’ medicine within oncology, inflammatory diseases, infectious diseases and cardiovascular conditions [6,53] reflected both a dissatisfaction with the granularity


205

Perspective Pokorska-Bocci, Stewart, Sagoo, Hall, Kroese & Burton implied by terms such as ‘personalized’ or ‘individualized’ medicine, which was felt to be somewhat impractical and commercially challenging, together with the idea that treatment decisions should be made on the basis of classifying or ‘stratifying’ patients by measuring a clinical biomarker(s) associated with a specific therapeutic response. Although the process of basing treatment decisions on the detection or measurement of biomarkers or other clinical characteristics (e.g., x-ray or ultrasound findings) is not conceptually new, what may be different is the very tight relationship between a specific biomarker measurement and a specific drug or intervention such that the intervention is only used when indicated by the biomarker (the so-called ‘companion diagnostic’). This represents a new diagnostic-intervention model from the regulatory perspective where the drug or intervention is only licensed for use in one specific context. Disease risk estimation & preventive medicine

A different use of the term ‘stratified’ has emerged in the field of disease risk estimation and population disease prevention [54] . Here, the proposition is that genomic and other factors (e.g., age and sex) are used to ‘stratify’ a population into a series of subgroups based on quantitative estimation of their risk for a specific disease, such as a cancer. The implication is that screening and preventive measures can be targeted to the subgroups at highest risk. It has been suggested that the utility for the individual arises from minimization of overtreatment [52] and an increased likelihood of identifying aggressive rather than indolent tumors [55] . Again, the idea is that genomics adds discriminatory power to more traditional risk estimations based on age (e.g., in population-screening programs for breast and bowel cancer), or on combinations of physical, biochemical and lifestyle features (e.g., BMI, blood lipid profile and smoking status for estimating coronary heart disease risk). Once again, the clinical process is one of placing an individual within a population subgroup – in this case, one that is at a similar level of risk and for which there is evidence of an effective preventive intervention. Conceptually, the process is the same whether genomic information is involved or not. The only potentially significant difference is that germline genetic information is stable throughout an individual’s lifetime, whereas phenotypic and lifestyle information is likely to change. Genomic information can also be accessed at an earlier age than many lifestyle factors. The use of genomic factors brings, however, potential concerns about discrimination and/or stigmatization in areas such as insurance that may limit its uptake within population screening programs [56] . It is important to note that genomic

206

Pers. Med. (2014) 11(2)

information is likely to remain one of a number of factors when establishing disease risk. Role of the patient

The advent of the internet and the development of userfriendly information technologies (e.g., mobile phone applications or devices for home health monitoring) have created opportunities for patients to access more health-related information and apply it to their own health and medical care. Web-based initiatives such as, for example, PatientsLikeMe [57] , allow individuals to identify others with similar health problems and to pool knowledge and advice. Such groups and resources are particularly useful and powerful for those with rare diseases. This unprecedented access to health information is linked to increasing health literacy and enthusiasm of individuals to understand and manage their own health. It also influences changes in sociopsychological aspects of medicine and of the medical professional–patient relationship by enabling individuals to take a more proactive part. Many patient-centric technologies are in development. For example, socalled patient-centered initiatives [58] include tools and processes providing an interactive information technology interface that empowers patients and research participants and places them at the center of decisionmaking. Although developed to facilitate longitudinal data collection, at the same time, they help to build a long-term public trust in biomedical advances and operate over increasingly blurred boundaries between research and clinical practice. Both patient and citizen empowerment also come high on current sociopolitical agendas. This and the increase in awareness of public health and ethical issues all provoke changes in the priorities and perceived obligations of a modern welfare state. An important aspect of ‘personalization’ of medicine is the attitude of treating each patient/individual as a ‘whole person’ and staying respectful of their wishes, views and lifestyle, which might include wishes to take or not to take greater responsibility for managing their own health [59] . All these factors provide a mechanism for moving from population- to individual-focused interventions and initiatives. On the other hand personalization can also be seen as a provision of healthcare similar to the provision of other goods or commodities in response to consumer demand and as such subject to similar consumer law, policy and sovereignty principles [59] . Finally, globalization, availability of international networks and initiatives, and easy access and connection through social networks play an important role in the development and potential implementation of personalized medicine.



Critical to this implementation will be the ability of all stakeholders to understand and act upon the more complex information available and assessing its quality and reliability. Both patients and their clinicians face a challenging future in this healthcare scenario. Conclusion Over the last decade or so, claims about the impact of genomics and other molecular biosciences have been associated with a range of terminologies, including ‘precision’, ‘personalized’, ‘individualized’ and ‘stratified’ medicine. We have charted the emergence and utilization of these terms in the published literature and found that they have provided an opportunity to crystallize and discuss important concepts in patient management. However, they are often inconsistently applied, thus obscuring the debate that needs to underpin the systematic introduction of personalized medicine into healthcare practice [44] . The most common inconsistency is to use the term personalized medicine to discuss solely the introduction of genomic technologies [25,28] . The other terms, such as individualized or precision medicine, are also used inconsistently, not always in line with their initial scope [12–13,16] . As discussed in this article, these inconsistencies can lead to misunderstanding among different stakeholder groups, which may significantly hinder the adoption of personalized medicine approaches. We have rejected the implied presumption inherent in many of these themes that genomics and other molecular biosciences have been instrumental in leading to a conceptually new paradigm of medical care, broadly falling under the rubric of personalized medicine. Genomic medicine will not cause a conceptual shift in the basic paradigm of medical care, although it will greatly enhance its power. Most clinical medicine will continue to be a process of attempting to choose the best care for an individual through a process of matching to a defined population subgroup. Advances in the biomedical sciences will change, refine and more finely divide the subgroups, with opportunities for more effective prevention and clinical care. The practice of medicine will be ‘personal’ in the sense that the experienced clinician will draw upon objective information from tests, evidence-based treatment options and clinical examination, together with subjective judgments informed by an understanding of an individual patient’s preferences and personal circumstances. What is new is the combination of powerful bioscience-based and information tools, and the rapidly changing sociopolitical contexts within which they are used, including increasing globalization, the bur-


Perspective

geoning use of social networks, and the emergence and use of ‘big data’ [60] . Together these developments may indeed revolutionize healthcare and the role of patients in that endeavor. New technologies and the vast amounts of biological and medical information they convey pose substantial organizational and financial challenges. Development and integration of flexible and responsive patient pathways that are effective and cost effective, while also being ethical and acceptable to professionals and patients alike, will be a major challenge. The continuing enthusiasm for the concept of ‘personalized medicine’ is a testament to the power and range of opportunities offered by the use of molecular science and technological innovation in combination with traditional medical practice. If the term continues in use, as seems likely, it is perhaps best understood not as a technocratic replacement for clinical judgment mostly based on advances in genomic medicine, but as a complex endeavor with a holistic lifetime approach in which both healthcare providers and individual healthcare users, in active partnership, take full advantage of the portfolio of modern tools and information available. Ultimately the success of this new phase of ‘personalized medicine’ will depend on the nature and level of benefits it can provide, both to individuals and to society more generally. Future perspective In the coming years molecular science and technological innovation will offer increasing opportunities for the establishment of a new phase in healthcare practice. Personalized medicine will continue to be used as an umbrella term to cover the ever-increasing range of modern tools and information as well as the evolving practice of healthcare providers and demands of healthcare users. In order to facilitate policy development, and to avoid confusion, we recommend that the other terms are used, in a consistent manner, for the specific applications described in this paper. The framework presented in this paper could be used as a useful tool to distinguish between the various terms, and facilitate the whole personalized medicine endeavor. Financial & competing interests disclosure The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties. No writing assistance was utilized in the production of this manuscript.


207

Perspective Pokorska-Bocci, Stewart, Sagoo, Hall, Kroese & Burton Executive summary Background • Genomics has led to new capabilities impacting the practice of medicine and healthcare. This phenomenon has been described by many terms such as ‘personalized medicine’, ‘stratified medicine’, ‘individualized medicine’ or ‘precision medicine’.

Methods • We reviewed the published literature surrounding ‘personalized medicine’ to determine its scope and conceptual basis. In order to achieve a good account of the concepts and their evolution over the years, we followed a systematic approach to identify the published literature and undertook a narrative synthesis.

Results • Individualized medicine: often used interchangeably with ‘personalized medicine’, but also refers to therapeutic approaches that utilize an individual’s own cellular material. • Precision medicine: a term used by the US National Research Council in a report referring to a reclassification of disease and providing a framework for new data networks integrating molecular research with clinical data. • Stratified medicine: a combination of a biomarker, diagnostic test, treatment and health outcome into one linear pathway to be applied to a well-defined subpopulation. • P4 medicine: stands for predictive, personalized, preventive and participatory, and denotes a paradigm shift in the practice of healthcare from reactive to preventive. • Personalized medicine: an umbrella term for the whole concept. Encompasses the application of genomic and molecular data to better target the delivery of healthcare and more holistic lifelong approaches. • Personalized healthcare: used interchangeably with other terms to cover concepts of genetics-based interventions, and their role in the tailoring of patient care.

Discussion • Our analysis suggests that ‘personalized medicine’ has emerged as the umbrella term in recent years. There is a lack of consistency in understanding the breadth and focus of ‘personalized medicine’ and it becomes crucial to define the term when used in a particular context and for particular purposes. • The use of terminology to imply novelty: it is important to examine what the contribution of genomics has been and whether genomics has brought anything fundamentally new to medicine. • The changing classification and diagnosis of disease: genomics and other molecular biosciences are now extending our ability to classify disease to a more fundamental cellular and molecular level but not changing the fundamental raison d’être of classification. • Disease management: in the genomic era management decisions will still be taken on the basis of attempting to match an individual to a population subgroup. However, the nature and amount of information used to define the subgroup may change radically. • Emergence of ‘stratified’ medicine: treatment decisions should be made on the basis of classifying or ‘stratifying’ patients by measuring a clinical biomarker(s) associated with a specific therapeutic response. • Disease risk estimation and preventive medicine: genomic and other factors are used to ‘stratify’ a population into a series of subgroups based on quantitative estimation of their risk for a specific disease, such as a cancer. The implication is that screening and preventive measures can be targeted to the subgroups at highest risk. • The role of the patient: the unprecedented access to health information and increasing health literacy generate an enthusiasm of individuals to understand and manage their own health.

Conclusion • We have charted the emergence and utilization of different terms used to describe the concept of personalized medicine. • We have rejected the presumption that genomics and other molecular biosciences have been leading to a conceptual shift in the basic paradigm of medical care. • What is new is the combination of powerful bioscience-based and information tools, and the rapidly changing sociopolitical context within which they are used. • The concept of ‘personalized medicine’ is best understood as the art of medicine combined with the portfolio of all the modern tools available to healthcare providers and users.

References

Papers of special note have been highlighted as: • of interest; •• of considerable interest

208

1

Baker M. Reprogramming Rx. Nat. Med. 17(3), 241–243 (2011).

2

Gravitz L. A fight for life that united a field. Nature 478(7368), 163–164 (2011).

Pers. Med. (2014) 11(2)

3

Srivastava P. Drug metabolism and individualized medicine. Curr. Drug Metab. 4(1), 33–44 (2003).

4

Committee on a Framework for Development a New Taxonomy of Disease, National Research Council. Toward Precision Medicine: Building a Knowledge Network for Biomedical Research and a New Taxonomy of Disease. The National Academies Press, DC, USA (2011).



••

The specific meaning of the term ‘precision medicine’ was coined in this report referring to a new molecular classification of diseases.

5

Wasi P. Human genomics: implications for health. Southeast Asian J. Trop. Med. Public Health 28(Suppl. 2), 19–24 (1997).

6

Trusheim MR, Berndt ER, Douglas FL. Stratified medicine: strategic and economic implications of combining drugs and clinical biomarkers. Nat. Rev. Drug Discov. 6(4), 287–293 (2007).

••

Introduces the term ‘stratified medicine’.

7

preventive, personalized and participatory. N. Biotechnol. 29(6), 613–624 (2012). 23

Pravettoni G, Gorini A. A P5 cancer medicine approach: why personalized medicine cannot ignore psychology. J. Eval. Clin. Pract. 17(4), 594–596 (2011).

24

Khoury MJ, Gwinn ML, Glasgow RE, Kramer BS. A population approach to precision medicine. Am. J. Prev. Med. 42(6), 639–645 (2012).

25

Gibson WM. Can personalized medicine survive? Can. Fam. Physician 17(8), 29–88 (1971).

Hood L. A personal journey of discovery: developing technology and changing biology. Annu. Rev. Anal. Chem. (Palo Alto Calif.) 1, 1–43 (2008).

26

Gupta R, Kim J, Spiegel J, Ferguson SM. Developing products for personalized medicine: NIH Research Tools Policy applications. Pers. Med. 1(1), 115–124 (2004).

••

Introduces P4 medicine based on the development of systems biology.

27

Burke W, Psaty BM. Personalized medicine in the era of genomics. JAMA 298(14), 1682–1684 (2007).

8

Abrahams E, Ginsburg GS, Silver M. The Personalized Medicine Coalition: goals and strategies. Am. J. Pharmacogenomics 5(6), 345–355 (2005).

28

Woodcock J. The prospects for “personalized medicine” in drug development and drug therapy. Clin. Pharmacol. Ther. 81(2), 164–169 (2007).

9

Langreth R, Waldholz M. New era of personalized medicine: targeting drugs for each unique genetic profile. Oncologist 4(5), 426–427 (1999).

29

10

Teng K, Eng C, Hess CA et al. Building an innovative model for personalized healthcare. Cleve. Clin. J. Med. 79 (Suppl. 1), S1–S9 (2012).

Moon H, Ahn H, Kodell RL, Baek S, Lin CJ, Chen JJ. Ensemble methods for classification of patients for personalized medicine with high-dimensional data. Artif. Intell. Med. 41(3), 197–207 (2007).

30

Meyer UA. Personalized medicine: a personal view. Clin. Pharmacol. Ther. 91(3), 373–375 (2012).

11

Abidi SS, Goh A. A personalised healthcare information delivery system: pushing customised healthcare information over the WWW. Stud. Health Technol. Inform. 77, 663–667 (2000).

31

Hong KW, Oh B. Overview of personalized medicine in the disease genomic era. BMB. Rep. 43(10), 643–648 (2010).

32

Brand A. Public Health Genomics and personalized healthcare: a pipeline from cell to society. Drug Metabol. Drug Interact. 27(3), 121–123 (2012).

33

Carlberg C. The need for education in personalized medicine. Pers. Med. 9(2), 147–150 (2012).

12

Hall JG. Individualized medicine. What the genetic revolution will bring to health care in the 21st century. Can. Fam. Physician 49, 12–17 (2003).

13

Hoffman MA, Williams MS. Electronic medical records and personalized medicine. Hum. Genet. 130(1), 33–39 (2011).

34

Moldrup C. Beyond personalized medicine. Pers. Med. 6(3), 231–233 (2009).

14

Graham-Rowe D. Overview: multiple lines of attack. Nature 480(7377), S34–S35 (2011).

35

15

Humphries C. Genetics: profiling a shape-shifter. Nature 480(7377), S50–S51 (2011).

Fierz W. Challenge of personalized health care: to what extent is medicine already individualized and what are the future trends? Med. Sci. Monit. 10(5), RA111–RA123 (2004).

16

Boguski MS, Arnaout R, Hill C. Customized care 2020: how medical sequencing and network biology will enable personalized medicine. F1000 Biol. Rep. 1, 73 (2009).

17

Director’s challenge: toward a molecular classification of tumors (1999). http://grants.nih.gov/grants/guide/rfa-files/RFACA-98--027.html

18

Mirnezami R, Nicholson J, Darzi A. Preparing for precision medicine. N. Engl. J. Med. 366(6), 489–491 (2012).

19

Robinson PN. Deep phenotyping for precision medicine. Hum. Mutat. 33(5), 777–780 (2012).

20

Blair ED, Clarke BR, Oneill T. Sustaining development of stratified medicines in the UK healthcare system: a commentary. Pers. Med. 8(5), 517–521 (2011).

21

Shaw EC, Johnson PW. Stratified medicine for cancer therapy. Drug Discov.Today 17(5–6), 261–268 (2012).

22

Hood L, Flores M. A personal view on systems medicine and the emergence of proactive P4 medicine: predictive,


Perspective

••

Conceptualizes personalized medicine in a wider context.

36

Personalised Medicine for the European Citizen – Towards More Precise Medicine for the Diagnosis, Treatment. European Science Foundation, Strasbourg, France (2012).

•

Summarizes current approaches to personalized medicine and identifies needs of all stakeholders.

37

Harvey A. Opinion leaders’ views on issues in personalized medicine. Pers. Med. 9(2), 127–131 (2012).

38

Cribb A, Owens J. Whatever suits you: unpicking personalization for the NHS. J. Eval. Clin. Pract. 16(2), 310–314 (2010).

39

Abidi SS, Han CY, Abidi SR. Patient empowerment via ‘pushed’ delivery of personalised healthcare educational content over the internet. Stud. Health Technol. Inform. 84(Pt 2), 1425–1429 (2001).

40

Zhang D, Yu Z, Chin CY. Context-aware infrastructure for personalized healthcare. Stud. Health Technol. Inform. 117, 154–163 (2005).


209


210

41

Jang H, Kim S, Bae C. Personalized healthcare through intelligent gadgets. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2008, 3308–3311 (2008).

51

Bonham JR. Impact of new screening technologies: should we screen and does phenotype influence this decision? J. Inherit. Metab. Dis. 36(4), 681–686 (2013).

42

Brand A. Integrative genomics, personal-genome tests and personalized healthcare: the future is being built today. Eur. J. Hum. Genet. 17(8), 977–978 (2009).

52

Sasinowski FJ. Quantum of Effectiveness Evidence in FDA’s Approval of Orphan Drugs. National Organization for Rare Disorders, CT, USA (2011).

43

Nicholson JK, Holmes E. Global systems biology and personalized healthcare solutions. Discov. Med. 6(32), 63–70 (2006).

53

Tang WH. The ongoing search for a stratified medicine approach in heart failure. J. Am. Coll. Cardiol. 51(24), 2310–2312 (2008).

44

Simmons LA, Dinan MA, Robinson TJ, Snyderman R. Personalized medicine is more than genomic medicine: confusion over terminology impedes progress towards personalized healthcare. Pers. Med. 9(1), 85–91 (2012).

54

Burton H, Sagoo GS, Pharoah P, Zimmern RL. Time to revisit Geoffrey Rose: strategies for prevention in the genomic era? Ital. J. Public Health 9(4), e8665-1–e8665-9 (2012).

55

45

Noseworthy PA, Newton-Cheh C. Genetic determinants of sudden cardiac death. Circulation 118(18), 1854–1863 (2008).

Esserman LJ, Thompson IM, Reid B. Overdiagnosis and overtreatment in cancer: an opportunity for improvement. JAMA 310(8), 797–798 (2013).

56

46

Baird RD, Caldas C. Genetic heterogeneity in breast cancer: the road to personalized medicine? BMC Med. 11, 151 (2013).

Burton H, Chowdhury S, Dent T, Hall A, Pashayan N, Pharoah P. Public health implications from COGS and potential for risk stratification and screening. Nat. Genet. 45(4), 349–351 (2013).

47

Curtis C, Shah SP, Chin SF et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature 486(7403), 346–352 (2012).

•

Role of stratified medicine in disease risk estimation and prevention.

57

•

Good example of use of sequencing technologies to fine-tune disease classification and management.

PatientsLikeMe®. www.patientslikeme.com

58

48

Phillips EJ, Mallal SA. Pharmacogenetics of drug hypersensitivity. Pharmacogenomics 11(7), 973–987 (2010).

Kaye J, Curren L, Anderson N et al. From patients to partners: participant-centric initiatives in biomedical research. Nat. Rev. Genet. 13(5), 371–376 (2012).

49

Leary RJ, Kinde I, Diehl F et al. Development of personalized tumor biomarkers using massively parallel sequencing. Sci. Transl. Med. 2(20), 20ra14 (2010).

59

Medical Profiling and Online Medicine: the Ethics of ‘Personalised Healthcare’ in a Consumer Age. Nuffield Council on Bioethics, London, UK (2010).

50

McBride DJ, Orpana AK, Sotiriou C et al. Use of cancerspecific genomic rearrangements to quantify disease burden in plasma from patients with solid tumors. Genes Chromosomes Cancer 49(11), 1062–1069 (2010).

••

A report discussing in detail the ethics of personalized healthcare.

60

Larson EB. Building trust in the power of “big data” research to serve the public good. JAMA 309(23), 2443–2444 (2013).

Pers. Med. (2014) 11(2)
