Importance The terms “personalized oncology” and “precision oncology” have increased in usage and have generated considerable traction in terms of public attention and research funding. To our knowledge, no prior study has as thoroughly documented the use of the “precision oncology” terminology over the last decade.
Objective To determine how the use of the terms “personalized oncology” and “precision oncology” have changed over time.
Design A retrospective literature analysis using two databases (PubMed and Scopus) over 10 years was performed. Manuscripts using either term “personalized oncology” or “precision oncology” were collected. Manuscripts published in 2011, 2013, 2015, 2017 and through 30 June 2019 were pulled for text analysis. Common reasons for exclusion were if the search term appeared in the institution name only, the search term appeared only in keyword or publication title, or the search term was used to justify the relevance or application of research with no clear definition.
Setting Manuscripts published and catalogued in PubMed or Scopus.
Results In our study, we analysed 399 unique manuscripts published over the last decade. Over time, the terminology has shifted from “personalized oncology” to “precision oncology”. Targeted therapy, molecular biomarker-guided tumour profiling and next generation sequencing (ie, “omics-guided tumor profiling”) are the three most common definitions of the term. While these definitions are somewhat overlapping in concept, over the decade we observed an increase in the number of distinct interpretations of “precision oncology”, ranging from structural biology to clinical practice.
Conclusions and relevance We have observed that the phrase “precision oncology” is shifting, overlapping and expanding in definition. This all-encompassing approach to defining “precision oncology” ironically renders the term imprecise. Our analysis highlights the inherent challenges in defining novel movements in medicine.
- cancer genetics
- gene therapy
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
Strengths and limitations of this study
To our knowledge, this study is the first to document and quantify the use of “precision oncology” over the past decade.
Other strengths include a large and systematic assessment of literature and use of figures to succinctly communicate the ambiguity that the term “precision oncology” has adopted in the literature.
Limitations of the study include the choice of search term selected (eg, “precision oncology” and “personalized oncology”) which may influence the results returned, the omission of even-numbered years in our decade-wide search and the acknowledgement that the interpretation of the use of terms may be susceptible to user bias.
Over the last decade, the terms “personalized oncology” and “precision oncology” have generated considerable traction in the media coverage of medicine and research funding. Some commentators have argued the profession has entered a new ‘era’ of oncology care.1
Formal definitions of “precision” and “personalized” oncology exist but are broad. In the USA, the Precision Medicine Initiative defines it as ‘an emerging approach for disease treatment and prevention that takes into account individual variability in genes, environment, and lifestyle for each person’.2 According to the UK Medical Research Council, ‘precision’ and ‘stratified’ approaches are ‘based on identifying patients or subgroups of patients with distinct mechanisms of disease, or particular responses to treatments’.3 Lastly, in Germany, at the well-established German Cancer Research Center, “precision” or “personalized” oncology is described functionally, as it ‘aims to offer individualized treatment to each cancer patient by applying a comprehensive molecular, cellular, and functional analysis of tumors’.4 Even with these institution-specific definitions, what precisely precision medicine means within the discipline of oncology has not been previously described in the literature.
Therefore, we sought to ask: how are the terms “precision” or “personalized” oncology used in the biomedical literature? Has the definition changed over time, and if so, how? To our knowledge, there has been no empirical analysis on the use of the terms “precision oncology” and “personalized oncology”. A small prior survey outside the peer-reviewed literature sampled 50 articles over 3 years to determine trends in the use of the term “precision oncology”, but did not use multiple search terms and provided only limited analysis.5 Here, we sought to provide a broad, updated analysis of these terms in the biomedical literature. Our work aimed to infer the authors’ interpretation of the term, even if it was not explicitly offered as an official definition. If the term itself is vague, it suggests that the field of “precision oncology” itself may be unstructured, nascent or still coalescing. Our analysis sheds light on the way in which a novel movement in medicine defines itself over time.
Literature search and use-of-term analysis
We sought to assemble a systematic collection of articles that used the terms “personalized oncology” and “precision oncology”. To do so, the terms “personalized oncology” and “precision oncology” and British spellings were searched on PubMed and Scopus literature databases between 1 January 2011 through 30 June 2019. We then attempted to characterise the definition of precision oncology over time. We prespecified that we would perform this analysis in odd years only, which we felt would be sufficient to document time trends and remain tractable. Manuscripts from 2011, 2013, 2015, 2017 and 2019 were obtained and stored on EndNote. We set a cut-off date of 30 June 2019 for article inclusion. Redundant entries were removed. All articles not originally published in English were translated with Google Translate. Articles that could not be obtained through the library were excluded from analysis. Publications that were errata, replies to letters, full-length books and book reviews were excluded from analysis.
A number of centres or institutes use the terms “precision” or “personalized” in their title. Search engines produce these articles in the search results, even if the term may not be used in the manuscript text. As such, we excluded articles where the search term appeared only in the manuscript title without further definition, only in the name of the associated research institution (eg, DKFZ-Heidelberg Center for Personalized Oncology), only in introductions or discussion paragraphs without further definition, or when the search term appeared in the manuscript without any attempt at definition.
Manuscripts were assessed for their opinion towards the field of precision oncology and relevant quotes surrounding the search term “precision oncology” or “personalized oncology” that supported our assessments of either enthusiasm or scepticism were collected.
Data organisation, statistical methods and RStudio analysis scripts
Data analyses were conducted using Microsoft Excel and R/RStudio, package Tidyverse. The analysis was conducted between 24 June 2019 and 31 August 2019.
Patient and public involvement
The development of the research question stemmed from a patient-centred approach to care. Specifically, as multiple new “precision oncology” drugs become available through the pipeline, and as media coverage increases its communication about the “precision oncology” concept, we felt it was worth exploring how the field defined “precision oncology” treatments to quantify and assess the consistency of the term. This study was not submitted for institutional review board approval as it involved not involve personally identifiable data, and all data are publicly available. No patients were involved in the design, analysis or interpretation of this study.
From 1 January 2011 to 30 June 2019, there were a total of 1547 publications using either term “precision oncology” or “personalized oncology” (PubMed (n=823) and Scopus (n=724)). Over this decade, the “precision oncology” phrase was more commonly used, and the number of publications that used either term has steadily increased, with “precision oncology” rising with a steeper slope than “personalized oncology”. These trends are shown in figure 1.
We analysed content in prespecified years (2011, 2013, 2015, 2017 and 2019). In these years, there were a total of 717 publications from PubMed (n=410) and Scopus (n=307). Removing duplicate publications, there were 565 unique entries retrieved between the two literature databases in our selected years.
“Precision oncology” and “personalized oncology” were used in 306 unique journals. The journal with the most publications containing either search term was Oncotarget, with 22 entries within our search. Other journals that frequently included articles using our search terms included Cancer Research (n=13), Clinical Cancer Research (n=13), NPJ Precision Oncology (n=10) and Annals of Oncology (n=8).
Characteristics of the search results, by year, are summarised in figure 2. Review articles were the most common form of article retrieved in 2011 (60% or 6 articles of 10), while original reports constituted 30% (3 of 10). By 2019, original reports were most prevalent, comprising 45% of articles, and reviews represented 41% of articles.
Of the 565 unique entries, we identified 399 eligible articles eligible for the use-of-term analysis: 2011 (n=8 of 10, 80%), 2013 (n=25 of 31, 80.6%), 2015 (n=74 of 85, 87.1%), 2017 (n=169 of 240, 70.4%), and 2019 (n=123 of 199, 61.8%), respectively, as shown in figure 3. Figure 3 also shows reasons for exclusion. The total number of articles excluded due to using the search term without offering a definition is 25.4% (n=144 of 565). The other common reasons for exclusion were due to the search term appearing only in the name of the institution without appearing in the manuscript text (n=89, 15.8%), the search term appearing in the manuscript text without an explicit attempt at definition, explanation or context (n=20, 3.5%), and the search term appearing only in the keywords without appearing in the manuscript text (n=12, 2.1%).
Across all years of interest, “precision oncology” is most commonly defined as using next generation sequencing and ‘omics’ data to identify an actionable mutation (n=6 (75%) in 2011; n=13 (52%) in 2013%; n=46 (62%) in 2015; n=115 (68%) in 2017; n=76 (62%) in 2019). However, many publications define the term with more than one definition; in addition to omics-guided tumour profiling, manuscripts also referenced other interpretations of “precision oncology”. These definitions included but were not limited to: (A) targeted therapy drugs; (B) use of a molecular biomarker to delineate subgroups, laboratory platforms for basic research (eg, high-throughput drug screens), (C) omics (genomics, transcriptomics, epigenomics, etc) to guide therapy or genome sequencing to identify actionable targets; (D) radiology and clinical imaging-based tumour profiling, (E) patient-derived xenografts and genetic engineering of mouse models, mouse or fly avatars to test drugs; (F) chemical drug screens and lab microarrays, (G) laboratory imaging and immunofluorescence; (H) immunotherapy and checkpoint inhibitors; (I) using circulating tumour DNA in the bloodstream to identify cancer recurrence and (J) prognostic clinical algorithms. As such, we mapped the overlapping and distinct uses of the terms, which is shown in figure 4.
Since 2011, there has been an increase in the number of distinct interpretations of “precision oncology”, represented by circles (from 4 in 2011 to 9 in 2019). The number of overlaps, representing a single article that uses multiple concepts to define “precision oncology”, has also increased from one overlap (eg, between targeted therapy and omics-guided tumour profiling) in 2011, to three overlaps in 2013, to five overlaps in 2017. The most common triple overlap is between targeted therapy, molecular biomarkers to delineate subgroups and identify actionable mutations, and omics-guided tumour profiling (n=1 in 2011, 12.5%; n=3 in 2013, 12%; n=9 in 2015, 12%; n=24 in 2017, 14%; and n=16 in 2019, 13%) was consistently seen throughout all years.
Manuscripts were also assessed for their opinion towards the field of precision oncology. Of the articles that took a stand about current or future prospects (n=280), 80.3% (n=225) were enthusiastic without any reservation, and 19.6% espoused some scepticism or reservation (n=55). Selected quotes from enthusiastic and critical publications have been collated to demonstrate the language surrounding precision oncology, as seen in table 1 (full collection of quotes available in online supplementary table 1). Whether enthusiastic or critical, publications commonly use phrases such as the “promise”, “potential”, “paradigm” and “era” of precision oncology.
Both precision and personalised oncology have gained popularity over the last decade, with the term precision oncology more frequently used. This is perhaps in part attributable to Dr. Harold Varmus’ preference for the latter term, exemplified by his statement indicating that his father received personalised care before the era of genomics.6 7 Care, in his mind, has long been personalised, unique to the patient in front of the physician; thus, a new term—precision oncology—is a better description of a novel effort.
Our analysis suggests that while the term “precision oncology” is currently most commonly understood to mean genomics-based tumour profiling, over the last decade its definition has shifted, overlapped and expanded to connote a variety of research fields that span basic research, clinical research and clinical practice.
Shifting definitions of precision oncology may reflect the natural scientific process where ideas evolve and change while concepts are emerging and undergoing clarification. At the same time, the shift may result in uncertainty among practitioners and the public as to what exactly precision oncology means. Moreover, claims about the ability of precision oncology to result in future advances in therapy are contingent about what precisely constitutes precision oncology. Standardisation and specification as to what precisely constitutes precision oncology may provide assistance in the ongoing debates about the potential of the field.8–11
Interestingly, we find in 2011, when precision oncology was more nascent, there was a high ratio of reviews over original reports (60% and 30%, respectively). This is interesting because the conventional wisdom is that reviews serve as the culmination and summary of decades of work, but perhaps reviews occurring earlier in scientific movements allow researchers to debate, ponder and shape the ideas yet to come. Future research may study this for other fields and topics.
Strengths and limitations
Our study has three strengths worth mentioning. First, this study is the first formal publication to document the in-depth use of “precision oncology” terminology and to quantify its use and definition over the past decade. Second, our approach is a large and systematic assessment of the literature. Third, our figures aim to succinctly communicate the complex ambiguity that the term “precision oncology” has adopted in the literature.
In terms of limitations, we also note at least three. First, the concept of precision oncology is recognised to have interchangeable terminology; “precision oncology” can also be described as personalized oncology, personalised oncology, personalized cancer medicine, precision oncology, precision cancer medicine, precision medicine in cancer or precision therapy. With our limited search parameters, and choice of search terms selected, we acknowledge that there are many papers we may have overlooked due to the limitations of our choice of search terms. Second, we omitted even-numbered years, for the sake of time, though we think their inclusion is unlikely to change the general time trends observed here. Lastly, interpretation of terms may be susceptible to user bias, and other readers may interpret these articles differently. We encourage others to examine these papers.
Even with these limitations, it is clear that the terminology we use to describe progress of “precision oncology” work has expanded, overlapped and shifted—in other words, precision oncology has at times been used imprecisely. And yet, the field of “precision oncology” is a consequential field and has gained momentum and considerable interest, judging by the growing number of articles in the biomedical literature on this topic (figure 1). If the term itself is vague, it suggests that the field of “precision oncology” itself may be unstructured, nascent or still coalescing.
Precision and personalised oncology are important concepts gaining increasing recognition and discussion in the peer review literature. The definitions used to describe these terms have been numerous, have overlapping concepts and have shifted over time. Our analysis sheds light on the way in which a novel movement in medicine defines itself over time.
Contributors AT, TC and VP contributed to the design and implementation of the research, to the analysis and to the writing of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests VP reports (Research funding) Arnold Ventures. (Royalties) Johns Hopkins Press, Medscape. (Honoraria) Grand Rounds/lectures from universities, medical centers, non-profits and professional societies. (Consulting) UnitedHealthcare. (Speaking fees) Evicore. (Other) Plenary Session podcast has Patreon backers. AT and TC have no competing interests.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article, publicly available or uploaded as supplementary information.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.