Protein Biomarker Research in UK Hospital Clinical Biochemistry Laboratories: A Survey of Current Practice and Views

Abstract

Background:

With the increasing drive for more and better disease biomarkers to underpin the stratified or personalised medicine agenda, clinical biochemistry laboratories should be ideally placed to play a major role in their translation into clinical practice. However, little is known about the current extent of biomarker-related research activity in UK National Health Service clinical biochemistry departments.

Methods:

In December 2010, an online questionnaire was sent to active UK members of the Association for Clinical Biochemistry (ACB) to determine the extent of their current research activity and involvement in protein biomarker discovery and translation, including an assessment of the awareness of proteomics.

Results:

A total of 198 eligible responses (19% response rate) was received from across the UK. Of a further 50 eligible people who responded to a follow-up for initial non-responders, most cited insufficient knowledge about the topic as the reason for non-response (24% total response rate). The results illustrate the highly skilled nature of the workforce with many having experience in a research environment (75%) with postgraduate qualifications. However, more than half spend <10% of their time undertaking research in their current role, and many (61%) would like to be more research active. Encouragingly, approximately a third were involved in biomarker discovery activities, even though for <10% of their time, with slightly more reporting involvement in biomarker translation.

Conclusions:

Although there are people with the necessary skills and desire to be involved in biomarker research in clinical biochemistry departments, their involvement is small, predominantly due to issues with capacity and resources. It is likely that the majority of biomarker programmes will therefore continue to be carried out by a small number of academic groups, hopefully with collaborative input from hospital laboratories.

Introduction

There is an increasing demand for cost-effective healthcare provision with any potential new biomarker requiring a strong evidence-base, showing improvements in patient outcome before it can be introduced into routine use.¹ Substantial resources have been focussed on the discovery of novel biomarkers since the advent of ‘omic’ technologies. However, as yet this has not been reflected in a commensurate introduction of new diagnostic protein biomarkers, for example into clinical practice.^2–4 Over the last 15 years, an average of only 1.5 new protein analytes per year have been approved by the US Food and Drug Administration.³

The biomarker pipeline can be viewed as essentially a three-stage process comprising (1) discovery; (2) qualification/validation; and (3) clinical evaluation, with each having different challenges and a high attrition rate (Figure 1). The progress of a novel biomarker through the validation phase seems to be the main limiting factor.⁵ The reasons for this have been extensively debated but include poor study design, technical variability, lack of standardisation in the collection and storage of clinical samples (all of which have resulted in some discovery results arising as a result of bias or artefact), an inconsistent approach to biomarker qualification, lack of samples from appropriate patient cohorts to facilitate study comparison and shortage of funding.^2,4,6–9 This disjointed approach can lead to underpowered studies that are not comparable between institutions, with insufficient evidence of biomarker clinical effectiveness and limited funding to drive biomarkers beyond the academic stage.¹⁰ There is also still a degree of uncertainty about the infrastructure/mechanisms needed to allow the rapid demonstration of sufficient clinical utility to support clinical acceptance.

Figure 1.

The biomarker pipeline.

Following the selection of suitable clinical samples to answer a clinical question, the biomarker discovery stage begins. This can include some of the ‘omic’ technologies and usually multiple platforms are used. Candidate biomarkers are then chosen for their novelty and relevance and validated by independent methods such as antibody arrays, multiple reaction monitoring or ELISA. This follows with the development and validation of biomarker assays and qualification of the biomarker for its intended clinical endpoint. The clinical evaluation phase includes clinical trials prior to translation into clinical practice. Throughout the pipeline the number of samples required for testing increases and the number of biomarkers in the pipeline decreases.

Ideally, biomarker discovery and/or translation programmes would be collaboratively pursued by staff within clinical biochemistry laboratories, with their insights into diagnostic testing, analytical expertise and access to clinical samples.^11–13 In the UK National Health Service (NHS), clinical biochemistry laboratories are staffed by healthcare scientists. These consist of two major groups of staff: Clinical Biochemists who may have a scientific or clinical background (this latter group may also be termed Chemical Pathologists) and Biomedical Scientists.¹⁴ In an ideal position to interact with clinicians treating patients to determine the usefulness of new and existing tests, they are experts in test result interpretation and analytical quality with, for example, an awareness of interference and cross-reactivity. In the past, most routine clinical biochemistry laboratories in the UK had substantial research capability, particularly in assay optimisation¹ but much of this capability has been lost, as highlighted in a review of NHS Pathology services in 2008.¹⁵ There may be a number of reasons for this. In the UK, population expansion and advances in medicine have resulted in an ever-increasing workload for routine services resulting in less time being available for research. The daily workload in Leeds Teaching Hospitals NHS Trust, for example, averaged 8700 test requests per day in November 2009 and this has steadily increased year-on-year by ∼8–10%, despite the pressures to make cost savings across the Trust. In addition, biomarker research is becoming more restricted by the drive towards automation, with single managed contracts being the preferred model¹⁶ and only limited or no support for the acquisition of expensive research-specific equipment. With the decline of academic pathology posts,¹¹ the opportunity to undertake research is even more limited and there may also be a lack of awareness of the ongoing work in relatively new research areas such as clinical proteomics.

To examine the capacity and level of activity relating to protein biomarker discovery and translation in NHS clinical biochemistry laboratories in the UK, we carried out a survey based on an online questionnaire circulated to the active UK ACB membership. The results and the issues raised concerning the practicalities and limitations of incorporating biomarker research into the NHS, from a staff perspective, provide useful information for planning and developing the strategy needed to meet the ever-increasing demand for new biomarkers, by highlighting some of the key aspects needing to be addressed and suggesting how the academic and NHS laboratories can potentially work together to deliver in this area.

Questionnaire Design and Study Participants

The pilot and final questionnaires were implemented using Bristol Online Surveys software (©University of Bristol, Bristol, UK). The questionnaire consisted of twenty questions covering three main areas: demographics of the respondents and research experience/current research activity, biomarkers and proteomics. Definitions of ‘biomarker’¹⁷ and ‘proteomics’ were provided for clarity. Most questions required categorical answers to be selected from predefined lists, although the option of selecting ‘other’ and providing a free text entry was also offered in most cases. A copy of the final questionnaire, following piloting as described below, is shown in Appendix 1.

The questionnaire was piloted at four hospitals before the final survey was initiated on 21 December 2010, with a 3-month time window for completion. Data from the pilot studies was not used in the final analysis. In order to target the most relevant population and enable the production of response rate statistics, the focus of the main survey was the membership of the ACB, as the vast majority of Clinical Biochemists are members. A targeted email was sent by the ACB Office to all active UK-based members with an electronic link to the survey. The survey was additionally advertised in the ACB News (January 2011 edition) and on the ACB Mailbase,¹⁸ a professional email discussion list. The eligibility criteria for completion included a request that respondents were scientifically or medically trained, laboratory or managerial, and based in UK clinical biochemistry laboratories. A follow-up message was sent out on 23 March 2011 to ask non-respondents to identify the main reason for not participating (open for a 1-month time window) with 74 people responding, 50 of whom fitted the eligibility criteria.

Only the responses from the final questionnaire were used in the data analysis which was performed using Microsoft Excel 2007 (©Microsoft, Redmond, USA). Responses are reported as absolute numbers or percentages of the overall number of respondents. Eight questions generated multiple responses from most participants and these were reported as percentages of responses or of respondents as indicated. The results are presented as observations and summaries of data.

Results

Demographics and Research Experience/Activity

A total of 205 individuals responded to the final survey. Seven of these were excluded because they worked outside the NHS (in industry or overseas). Of the remainder, 186 were active Clinical Biochemists (Table 1) providing an estimated response rate of 18.9% of active Clinical Biochemists as determined from the 2010–2011 UK ACB workforce statistics and ACB member database. A follow-up survey asking about reasons for non-response drew a response from a further 50 eligible members, with the majority indicating that this was due to lack of knowledge of the area, giving a total response rate of 23.7%. The response rate in this study compares favourably to that observed in other healthcare-based surveys.^19,20 Responses were received from 10 different grades of staff (Table 1) and from 91 of 167 UK Acute NHS Trusts. The response rate from scientifically and medically trained Clinical Biochemists was 17% and 18% respectively. Nearly half (44%) were of consultant grade. Only five responses (2.5%) were received from Biomedical Scientists, which would be expected given their limited membership in the ACB at the time of the survey and the staff grouping of another seven (4%) could not be determined.

Table 1.

Professional groupings of respondents^*

Professional Group	Number	Percentage of respondents
Biomedical Scientist	5	2.5
Trainee Clinical Biochemist	25	13
Higher Specialist Trainee Clinical Biochemist	4	2
Senior Clinical Biochemist	30	16
Principal Clinical Biochemist	32	16
Consultant Clinical Biochemist	47	24
Consultant Chemical Pathologist	39	20
Chemical Pathology Registrar	5	3
Metabolic Medicine Registrar	4	2
Other	7	4

^*The total number of Clinical Biochemists eligible to respond was 1046, of whom 785 had a scientific background and 261 a medical background. Figures show number of eligible respondents and % in each staff category. A total of 186 Clinical Biochemists completed the questionnaire with 72% being scientifically qualified and 26% being medically qualified. Of a further 12 respondents, 5 were Biomedical Scientists and 7 could not be categorised.

The respondents were based in 119 different hospitals, with the majority being from teaching hospitals (74%) and most (89%) being full-time and in permanent posts (80%). All indicated they worked in diagnostic laboratories with 42% having more than 20 years experience and 75% indicated that they had prior research experience, most for a period of 1–5 years (Figure 2). This was consistent with the level of postgraduate experience amongst the study population, where 90% had at least one postgraduate qualification (higher degrees and/or diplomas). Just over half (59%) had gained this experience in academic laboratories with the remainder having gained experience in specific research institutes and industry.

Figure 2.

Experience.

Period of time (years) respondents have worked in a diagnostic laboratory (dark shading) versus years in a research laboratory during previous employment (light shading).

A minority of respondents (20%) indicated that they took the lead in research projects and many specified that research was not departmentally funded (88%) with multiple other sources of funding reported. However, 52% indicated links with or being part of a university, presumably allowing collaborative research where possible. As illustrated in Table 2, the main research activities taking place in diagnostic laboratories included academic projects, method development and optimisation, assay evaluation and clinical trials. There was, however, a major difference between what respondents thought should be happening in their laboratory compared to what was actually happening (Figure 3a), with 49 % thinking that the lab should have a research role but only 9% indicating that their laboratory had this role. Many (61%) indicated a desire to be more active in research, with 55% of respondents only spending 1–10% of their time on research projects in their current role and 12% not doing any research at all (Figure 3b).

Table 2.

Types of research activity undertaken by each respondent’s department^*

Research Activity	Number	Percentage of respondents
MSc/other academic projects	148	74.7
Method development/optimisation	132	66.7
Evaluating existing assays	126	63.6
Clinical trial work	111	56.1
Developing diagnostic strategies	76	38.4
Process management/optimisation	60	30.3
Evaluating new technologies	59	29.8
Evaluating new biomarkers	55	27.8
Biochemical mechanisms of disease	32	16.2
Biomarker discovery	19	9.6
Other	13	6.6
None	8	4.0

^*Each respondent was asked to tick each activity that applied or free text others which were then grouped into similar activities as indicated. A total of 839 departmental research activities were indicated by the 198 respondents.

Figure 3.

Roles and research time.

(a) The role of clinical biochemistry currently (dark shading) and what the future role of clinical biochemistry should be (light shading). Since respondents could choose any number of statements the data is shown as percentage of respondents who selected each possibility.

(b) Percentage time spent on research in current position.

Biomarkers

Respondents were asked about their understanding and experience of biomarker discovery and translation. Staff identified that they maintained their knowledge by keeping up to date with relevant literature regarding novel biomarkers (67%) and existing biomarkers in routine use (84%). Seventy percent said that they read articles at least weekly/monthly to keep abreast of developments. However, more than half said they spent no time on biomarker discovery/translation programmes, even though most (85%) had indicated that they thought they should play a role in translational research (see Figure 1).

Approximately one-third of respondents indicated that they had direct involvement in biomarker discovery projects although most of these spent <10% of their time on this activity and, when asked in more detail about biomarker activities, <10% reported a current role in biomarker discovery per se and 24% in evaluating new biomarkers. A greater number (45%) were involved in biomarker translational research. Those who confirmed the latter were mostly involved in assay validation, clinical trials and assessing the evidence base (total 81%), which are all key elements in introducing a new biomarker. Almost two-thirds (59%) said they only had an average interest in learning more about biomarker programmes, although an enthusiastic 24% showed considerable interest. In contrast, 81% were of the opinion that a subsection of their laboratory should be devoted to biomarker discovery/translation, but most indicated that this was not viable due to limited staffing and financial resources.

A majority (79%) of respondents took the view that diagnostic laboratories should be involved in identifying a need for a new biomarker, with 85% supporting involvement in biomarker translation (validation/qualification) and two-thirds supporting banking samples for biomarker research. Only one-third took the view that diagnostic laboratories should play a role in the discovery stage. Of note, two respondents identified that assay development in collaboration with In Vitro Diagnostic (IVD) companies, and auditing the clinical utility of new tests was also essential (free text entries).

Proteomics

Few respondents indicated any more than a superficial/theoretical knowledge of proteomics, with only 2% indicating an extensive experience and a further 7% having been involved in at least one proteomics-based project. Most (42%) indicated only an awareness of the term. Few of the respondents had practical experience of the more specialised mass spectrometry-focussed proteomics techniques either from working in a diagnostic laboratory (Figure 4a), or in previous research employment (Figure 4b).

Figure 4.

Experience of proteomic techniques.

(a) In a diagnostic laboratory

(b) In previous research employment.

Respondents who had never worked in a research laboratory are excluded in the calculation of % respondents in (b).

Key: 1D-GE = 1D-gel electrophoresis; 2D-GE = 2D-gel electrophoresis; TMS = tandem mass spectrometry; MRM = multiple reaction monitoring; MALDI = matrix-assisted laser desorption/ionisation mass spectrometry; SELDI = surface-enhanced laser desorption/ionisation mass spectrometry; HPLC = HPLC for protein purification; Protein array = protein array technology.

Despite the lack of expertise and experience, the majority of respondents thought that it would be feasible to introduce proteomics techniques into clinical biochemistry laboratories for diagnostic use (59%) or for biomarker research (67%) within the next 10 years.

Feedback and Comments

A total of thirty respondents left additional comments at the end of the survey (Appendix 2). The main themes were: a need for clinical laboratories to be more closely integrated with research and development (R&D) at all levels; collaboration with academia/industry is essential; ethics and R&D approval is frustrating and time-consuming; and the inhibitory effect of the current financial climate for improvements in this area.

In the follow-up to the survey, where the ACB membership was requested to identify reasons for non-response, 74 individuals replied. The main reasons given were that they did not know enough about the topic (24) or that clinical biochemistry was not their speciality and so were not eligible to respond (24).

Discussion

With increasing technological developments enabling proteomic biomarker discovery there is real optimism that novel biomarkers will begin to emerge. Ideally this would present opportunities for staff in clinical biochemistry laboratories to participate in research involving either biomarker discovery or the translation of research findings into clinical practice, based on evidence-based progression. However this national survey suggests that many UK clinical biochemistry staff do not have the time and resources to carry out research in their current roles. Whilst these results are of no real surprise, it was important to obtain a staff perspective on the issues of biomarker research in the NHS which may be useful in developing future strategic initiatives in this area.

In order to determine a response rate and therefore gain some idea of the representative nature of the responses, the key workforce members were targeted using the ACB membership list. However, although this covers the majority of Clinical Biochemists it is not complete and, additionally, a relatively small number of staff from groups such as Biomedical Scientists and Clinical Scientists in specialities other than clinical biochemistry, such as immunology and microbiology, are members. Responses from Clinical Scientists in other specialities were excluded and the eligible population was based on the ACB membership list and workforce statistics. The response rate obtained is therefore likely to be an underestimate as it is difficult to establish from the list who is active, for example. Another potential flaw of any survey of this type is a potential for bias towards scientists with a particular interest in the questionnaire topic although the fact that more than half of the respondents admitted to spending <10 % of their time on research suggests that this was not the case.

A majority of the individuals who responded to this survey indicated that research was still a vital part of their role, but few had the opportunity in their current employment. Indeed, many indicated experience and willingness to drive research activities, but the presence of academic groups within healthcare science is diminishing in the UK with efficient, cost-effective routine diagnostic testing becoming the main objective due to ever-increasing service pressures. In 2002, Wierzbicki and Reynolds²¹ investigated the publication activity of Clinical Biochemists in the UK between 1995 and 1999. They found that 80% of research in publications was performed in teaching hospitals and only 1% of Clinical Biochemists published research in journals with an impact factor of >4 each year, and most of these were senior members of the profession. Nearly half of the staff studied did not publish any journal articles in the five-year study period. Long-term research activity appeared to correlate with the possession of a higher research degree or early research opportunities, therefore they suggested that research training is required at the earliest opportunity. Retrospectively, it would have been useful to have included questions in our survey about research activity currently in terms of publications, both broadly and specifically in the area of biomarkers. Such information would have been useful in providing a benchmark against which future activity levels could have been judged.

In the UK there has been a lack of specific research training for trainee Clinical Biochemists beyond attainment of a Masters degree in Clinical Biochemistry. However, hopefully this will be addressed by the Modernising Scientific Careers curriculum.²² The curriculum contains modules for research methods training as well as an extensive research project although no direct reference is made to the need for specific analytical and methodological skills. Of course it may be some time before any changes to the curriculum are reflected in workforce expertise and, in the interim, provision of workshops on biomarker development methodology by bodies such as the ACB and Royal College of Pathologists may provide some training and support.

It was encouraging to find that approximately a third of respondents were involved in biomarker discovery projects although this was <10% of their research time. Slightly more reported involvement in biomarker translation projects. However, since only ∼10% of respondents reported that their departments were involved in biomarker discovery and that some departments were potentially represented by more than one respondent, this may indicate that much of the biomarker research is undertaken collaboratively rather than being led by Clinical Biochemists. This is also supported by ∼40% reporting their current role in research to be collaborative. In terms of the role of the diagnostic laboratories, it was interesting that the vast majority indicated that the main roles should be in identification of the need for new markers and their validation/qualification, with only about a third advocating roles in biomarker discovery. This aligns with their area of expertise and current involvement, in that very few indicated in-depth knowledge or involvement with proteomic technologies, and is actually realistic given the often different equipment base and high costs/intensive nature of such activities, which would make this difficult in the majority of routine diagnostic laboratories unless through academic activities and collaboration.

A surprising number of respondents did have practical experience of specialised mass spectrometry techniques (tandem mass spectrometry or multiple reaction monitoring) in a diagnostics laboratory, although the survey was ambiguous in that it did not specify experience in the area of protein biomarkers and this may reflect use for example in inborn errors of metabolism, toxicology or steroid endocrinology. It illustrates an important skill set however, particularly given the use of mass spectrometry multiple reaction monitoring (MRM) assays in high-throughput biomarker screening for prioritisation for stringent assay development following initial discovery. However, the use of MRM for serum proteins is challenging and unlikely to be used routinely.²³ As the focus of the questions on technologies and practical skills was really about proteomic approaches, immunoassays per se were not included in the listing. However, it would have been useful to have included within the survey a section identifying the current level of involvement in this area specifically, both in terms of new assay development skills and evaluation of existing assays, as this is clearly a major area of potential contribution from diagnostic laboratories in biomarker testing and development. We also did not ask our respondents about wider aspects of biomarker studies such as their awareness of guidelines such as STARD²⁴ and REMARK²⁵ which arguably are equally as important as technological skills in terms of ensuring good study design and assay characteristics/performance.

In addition to the constraints of increasing workloads and the drive to increase cost effectiveness, other principal issues hindering research in the NHS, which were identified by this survey were staff availability and financial resources. The lack of staff resource is illustrated by the dramatic (9%) fall in the number of UK Clinical Biochemist posts in clinical biochemistry seen over the last two years (ACB Workforce Advisory Committee), falling from a total of 829 in 2009 to 754 in 2011. Of note, consultant grade posts have fallen by 23% in this period and the results of this survey and the literature²¹ indicate that their involvement in research outweighs that of other staff members. There is still a perception in the NHS that research is not a core activity, with clinicians expecting it to occur in universities or diagnostic companies.²⁶ Currently, diagnostic companies spend large sums of money bringing new biomarkers to the market, often with little evidence of clinical utility.¹ In this survey, a number of respondents felt that translational research would need to be organisationally separate and externally funded as most Trusts no longer have a real research budget applicable to anything other than immediate clinical improvements. This is the case now more than ever as large elements of the health service are being privatised and expertise will suffer as diagnostic companies select profitable but clinically less important tests.

It would have been interesting to have asked respondents their opinions as to what could be done to increase their research involvement and expertise. In light of increasing awareness about some of the issues described here, it may be that professional bodies such as the ACB and the Royal College of Pathologists have a role in organising research training workshops, for example. Conversely in terms of increasing opportunities for involvement, models such as that proposed by Sturgeon and Selby¹² may be the way forward, with a four-way multidisciplinary team comprising research laboratories, clinical laboratories, the diagnostics industry and clinicians. Ideally projects would be multi-centre or multi-national, with the creation of large banks of clinical samples to reduce the heterogeneity in sample selection and assembly. Open collaborations would speed up the process of producing data and could enable a model to be developed for future projects. An increase in collaborative projects between academic research groups and NHS laboratories may also provide a platform for staff to gain knowledge and experience in this rapidly developing and closely linked discipline, and allow knowledge and expertise to be utilised appropriately. Currently, these joint NHS and university posts are rare but future efforts to increase them would be a promising step forwards. Recently introduced initiatives such as the NIHR Diagnostic Evidence Cooperatives²⁷ exemplify these types of partnerships in the latter phases of biomarker evaluation and translation and emphasise the involvement of pathology laboratories. Such funding streams may offer further opportunities.

Conclusions

Our survey has gained a unique view of the issues surrounding participation of Clinical Biochemists in biomarker discovery and translation in the NHS. At the time of our data analysis, the Medical Education England Healthcare Science Programme Board was conducting a survey regarding clinical academic careers for healthcare scientists to quantify the contribution healthcare scientists make to research, education and teaching. It will be interesting to gauge the agreement between the two survey reports.

Appendix 1. Questionnaire

BIOMARKER DISCOVERY & TRANSLATION IN CLINICAL BIOCHEMISTRY LABORATORIES

Dear User

We would appreciate it if you would spend 5–10 minutes answering the following 20 questions. The questions have been designed to assess the current level of awareness and/or experience Clinical Biochemistry staff have regarding protein biomarker discovery and translation programmes (in UK NHS hospitals).

All data collected in this survey will be held anonymously and securely. No personal data is asked for or retained.

Cookies, personal data stored by your Web browser, are not used in this survey.

• 1My job title is:

  • Trainee Biomedical Scientist

  • BMS 1

  • BMS 2

  • BMS 3

  • BMS 4

  • Trainee in a Clinical Science discipline

  • Higher Specialist Trainee

  • Senior Clinical Scientist

  • Principal Clinical Scientist

  • Consultant Clinical Scientist

  • Consultant Chemical Pathologist

  • Chemical Pathology registrar

  • Metabolic Medicine registrar

  • Other: (blank box)

  1. And my employment is:

     □ Full-time    □ Part-time    □ Job share

     □ Other (please specify):

  2. In a position that is:

  3. □ Permanent    □ Fixed-term    □ Locum

     □ Other (please specify):

• 2I work at the following hospital: (please enter in free text below)

  The hospital I work in is a teaching hospital

  □ Yes □    No □    Do not know

• 3I have the following post-graduate degree(s):

  (select all that apply)

• • □ MSc

  • □ MPhil

  • □ MD

  • □ PhD

  • □ None

  • □ Other (please specify):

• 4I have worked in a diagnostic laboratory for the following number of years (range <1 to >50 years):

  1. In addition, I have previously worked as a research scientist in a research laboratory for the following number of years (range never to >50 years):

  2. If you have previously worked in a research laboratory please specify the type(s): (Optional)

     (select all that apply)

     □ Academic

     □ Government

     □ Industrial

     □ Research Institute

  3. □ Other (please specify)

• 5The diagnostic laboratory I currently work in has links to the following academic research institute/department (if your answer is ‘none’ or ‘do not know’ please enter in free text box below):
• 6I currently spend the following percentage of my time on research (0–100%):
• 7The research activity undertaken in my department generally involves:

  (select all that apply)

• • □ MSc/other academic projects

  • □ Clinical trial work

  • □ Method optimisation

  • □ Evaluating new technologies

  • □ Evaluating new biomarkers

  • □ None

  • □ Do not know

  • □ Other (please specify):

• My current role in research is usually:

  □ Not applicable    □ Employer-driven    □ Collaborative    □ Lead

  1. In the following area(s):

     (select all that apply)

     □ My own MSc/other academic project

     □ As a supervisor of an MSc/other academic project

     □ Clinical trial work

     □ Evaluating new methods

     □ Evaluating new technologies

     □ Evaluating new biomarkers

     □ Not applicable

     □ Other (please specify):

  2. And this research is funded by (if your answer is ‘not applicable’ or ‘do not know’ please enter this into the free text box below):

  3. I would like to be:

     □ More research active    □ Less research active    □ Undertaking the same amount of research as I do now

• 9I think the direction of diagnostic clinical laboratories IS CURRENTLY:

  (select all that apply)

  □ To only provide a diagnostic service

  □ Diagnostics and undertaking independent research

  □ Diagnostics and undertaking research in collaboration with in-vitro diagnostics companies

  □ Diagnostics and undertaking research in collaboration with academic institutes

  □ Diagnostics and undertaking research in collaboration with biotech/pharmaceutical companies

  □ Diagnostics and undertaking research in collaboration with government-funded research departments

  □ Do not know

  □ Other (please specify):

  I think the future of diagnostic clinical laboratories SHOULD BE:

  (select all that apply)

  □ To only provide a diagnostic service

  □ Diagnostics and undertaking independent research

  □ Diagnostics and undertaking research in collaboration with in-vitro diagnostics companies

  □ Diagnostics and undertaking research in collaboration with academic institutes

  □ Diagnostics and undertaking research in collaboration with biotech/pharmaceutical companies

  □ Diagnostics and undertaking research in collaboration with government-funded research departments

  □ Do not know

  □ Other (please specify):

Biomarkers

Biomarkers (‘biological markers’) are characteristics used as an indicator of a biological state. They are objectively measured to indicate normal biological or pathogenic processes, or pharmacologic/biologic responses to an intervention. There are inadequate biomarkers for many diseases, whether for diagnosis, prognosis, treatment selection or monitoring. Examples of biomarkers include commonly measured analytes like troponin, TSH, tumour markers etc, as well as novel biomarkers such as BNP or N-terminal telopeptide.

Biomarker discovery is the process by which a ‘marker’ is linked to a particular biological condition. The recent interest in biomarker discovery is due to the advances in proteomic techniques, among others, which promise to find relevant markers rapidly. Achieving the full value of biomarkers hinges on the ability to successfully translate them for clinical use, that is, moving them from the research laboratory to the clinic. This involves validation of the biomarker assay, as well as qualification of the biomarker for its intended clinical endpoint.

Ideally, biomarker discovery and/or translation programmes would be pursued within hospital clinical biochemistry laboratories, with their expertise in routine biomarker analysis and access to clinical material. However, with service pressures and the expense of research technology acquisition, this rarely occurs and the research is increasingly carried out by a small number of academic research groups.

• 10.I regularly read current biomarker-related reviews/original articles in the literature regarding:

  (select all that apply)

  □ Existing biomarkers and their clinical use

  □ Biomarker discovery processes

  □ Novel biomarkers and their prospective clinical use

  □ Current methodology for biomarker discovery

  □ Current methodology for routinely measured analytes

  □ Novel methodology for biomarker discovery

  □ Novel methodology for routinely measured analytes

  □ None

  □ Other (please specify):

  1. On a:

     □ Daily basis    □ Weekly basis    □ Monthly basis    □ Quarterly basis

     □ Six-monthly basis    □ Annual basis    □ Very rarely    □ Never

  2. I select the article(s) I read based on:

     (select all that apply)

     □ A subscription to a relevant journal(s)

     □ Email alerts from relevant journal(s)

     □ Access to a relevant journal(s) within my department

     □ Articles being brought to my attention by a colleague

     □ Not applicable

     □ Other (please specify):

• 11My interest in learning more about biomarker programmes is:

  (please select the most appropriate)

• □ Considerable    □ Average    □ Minimal    □ Non-existent

• 12.I participate in biomarker discovery projects the following percentage of my time at work (0–100%):

  1. I participate in biomarker translation projects the following percentage of my time at work (0–100%):

  2. If you have been involved in biomarker translation what has your involvement been? (Optional)

     (select all that apply)

     □ Clinical trials of new biomarkers

     □ Assay validation for new biomarkers

     □ Other (please specify):

• I think that diagnostic laboratories should get involved in the following stages of biomarker programmes:

  (select all that apply)

  □ In identifying a clinical need for a biomarker

  □ In sample banking

  □ In the biomarker discovery phase

  □ In the biomarker translational phase

  □ None of the above

  □ Other (please specify):

  If you selected the ‘translational phase’ above, which of the following does this relate to:

  □ Not applicable    □ Assay validation    □ Qualification of clinical utility    □ Both validation and qualification

• I think having a subsection of our diagnostic laboratory devoted to biomarker discovery/translation programmes is a:

  □ Good idea    □ Bad idea

  And it: (please select the most applicable option)

  □ Is not viable at all

  □ Is not viable, but would be if the financial resources were available

  □ Is not viable, but would be if the staff time was available

  □ Is not viable, but would be if the financial resources and staff time were available

  □ Not sure

  □ Other (please specify)

Proteomics

Proteomics is the study of the proteome, which is all of the proteins expressed, including post-translational modifications, within a particular fluid, cell, tissue, organism etc. Studying the proteome utilises a number of specialist technologies, some of which are time-consuming, technical, and/or expensive.

• 15.My experience/understanding of proteomics is:

  (please select the most appropriate answer)

  □ Extensive

  □ Previous involvement in at least one proteomics project

  □ A broad understanding of the theory/methods involved

  □ Reading of original research that has utilised proteomic techniques

  □ Being aware of the term ‘proteomics’

  □ None prior to this questionnaire

  □ Other

  In relation to the above, I have had this experience for a period of (0 to >30 years):

• 16.
  I have had practical and/or theoretical experience of (includes reading or learning about) the following techniques DURING MY EMPLOYMENT IN A DIAGNOSTIC CLINICAL LABORATORY:

  	Experience:
  	Practical & Theoretical	Theoretical only	None
  a. 1D-gel electrophoresis	□	□	□
  b. 2D-gel electrophoresis	□	□	□
  c. Tandem mass spectrometry	□	□	□
  d. Multiple reaction monitoring (MRM)	□	□	□
  e. Matrix-assisted laser desorption/ionisation mass spectrometry (MALDI)	□	□	□
  f. Surface-enhanced laser desorption/ionisaton mass spectrometry (SELDI)	□	□	□
  g. HPLC for protein purification	□	□	□
  h. Protein array technology	□	□	□

• 17.
  I have had practical and/or theoretical experience of (includes reading or learning about) the following techniques DURING MY PREVIOUS EMPLOYMENT IN A RESEARCH LABORATORY:

  	Experience:
  	Practical & Theoretical	Theoretical only	None
  a. 1D-gel electrophoresis	□	□	□
  b. 2D-gel electrophoresis	□	□	□
  c. Tandem mass spectrometry	□	□	□
  d. Multiple reaction monitoring (MRM)	□	□	□
  e. Matrix-assisted laser desorption/ionisation mass spectrometry (MALDI)	□	□	□
  f. Surface-enhanced laser desorption/ionisation mass spectrometry (SELDI)	□	□	□
  g. HPLC for protein purification	□	□	□
  h. Protein array technology	□	□	□

• 18.In my opinion the feasibility of introducing proteomic technologies into specialist clinical biochemistry or blood science laboratories FOR DIAGNOSTIC USE IS: (please select most appropriate answer for each technique)

  MALDI: Matrix-assisted laser desorption/ionisation mass spectrometry

  SELDI: Surface-enhanced laser desorption/ionisation mass spectrometry

  Viability:
  Viable	Not viable, due to financial constraints	Not viable, due to staff time availability	Not viable	Not sure
  a. 1D-gel electrophoresis	□	□	□	□	□
  b. 2D-gel electrophoresis	□	□	□	□	□
  c. MALDI	□	□	□	□	□
  d. SELDI	□	□	□	□	□
  e. Protein array technology	□	□	□	□	□

• 19.In my opinion the feasibility of introducing proteomic technologies into specialist clinical biochemistry or blood science laboratories FOR RESEARCH USE IS: (please select most appropriate answer for each technique)

  MALDI: Matrix-assisted laser desorption/ionisation mass spectrometry

  SELDI: Surface-enhanced laser desorption/ionisation mass spectrometry

  Viability:
  	Viable	Not viable, due to financial constraints	Not viable, due to staff time availability	Not viable	Not sure
  a. 1D-gel electrophoresis	□	□	□	□	□
  b. 2D-gel electrophoresis	□	□	□	□	□
  c. MALDI	□	□	□	□	□
  d. SELDI	□	□	□	□	□
  e. Protein array technology	□	□	□	□	□

• 20.I envisage the time-frame for integration of proteomic technologies (e.g. MALDI) into specialist clinical biochemistry or blood science laboratories for BIOMARKER DISCOVERY RESEARCH to be (now to never or don’t know):

  1. I envisage the time-frame for integration of proteomic technologies (e.g. MALDI) into specialist clinical biochemistry or blood science laboratories for THE ROUTINE MEASUREMENT OF BIOMARKERS to be (now to never or don’t know):

• 21.If there are any further comments you wish to add relating to this questionnaire please enter them in the text box below: (Optional)

Appendix 2: Feedback Comments

Feedback comments
1	Clinical Laboratories need to be closely integrated with R&D and translation as technology develops and treatments change. Because of access to clinical samples, we have shown that collaborative working with academic/ industrial and NHS partners enable developments in new biomarkers.
2	Collaboration with academic research departments is essential for NHS Departments to become involved in biomarker research but IP issues seem to hinder early stage translation.
3	Diagnostics labs are now too busy fighting for survival and being directed towards becoming glorified District General Labs providing GP results as a Hub within a Hub and spoke system spending less and less time on research and development. The number of labs with research aspirations is shrinking and service is dominating even so called University/Teaching Hospital Labs. The future looks grim. Introduction of anything that apparently increases expense for blood sciences labs is not very likely in the next 10 years. A completely new approach is required.
4	Getting proposed research through R&D committees and then Ethics Committees has been very frustrating and time consuming. Required forms seem to change every few months (or less). I feel this has had a negative effect on performing relatively simple research projects. Indeed our MSc students look for projects which will not need ethics approval because they do not have time to go through the process. Hospital laboratories used to do lots of good research but I feel this has almost dried up because of the barriers (ethical, financial and staffing) that are now in place.
5	Have any of the newer ‘omic’ approaches lead to the development of a clinical ‘biomarker’ that outperforms our current biomarkers?
6	I am now nominally retired, and much of this questionnaire is no longer applicable, but have recently been asked to collaborate on a research project relating to diagnostics funded by the ***** (which was impossible due to loss of appropriate lab facilities), and continue to be requested to publish reviews or comment on recently published diagnostic technologies. My main experience of clinical chemistry has been as the ******** at a time prior to the development of diagnostic assay kits and automatic analyzers, when assays were performed ‘by hand’ using reagents obtained in-house by highly trained personnel. SAS laboratories were all originally established within university departments directed by academic scientists with a background in research, and who were the source of several new assay technologies. Current ACB laboratories are largely staffed by personnel who lack this background; also either the time, ability, funding or motivation to develop new methods. Hence very few if any new concepts, techniques or applications have emerged from routine diagnostic departments in the UK in the past 30 years. This is an important loss since clinical chemists have contact with patients and clinicians, and are therefore ideally placed to perceive when new assay methods are required, or to evaluate the clinical utility of methods developed by diagnostic companies.
7	I hear SELDI is really bad, if evidence shows it is good then I may change my answers from not viable. Analysis of glycosylation as a biomarker are much further away (in my opinion) than other biomarker projects. I think more should be taught about glycosylation in biochemistry MSc courses.
8	I would envisage candidate biomarkers being isolated by a research lab with the necessary knowledge base and MS equipment (e.g. orbitrap, MALDI-TOF, TOF/TOF etc) and once candidate protein(s) are isolated a specialist centre lab being involved in screening large numbers of samples (by tandem MS or immunoassay) to get the necessary sensitivity / specificity data. Once that is generated then a commercial immunoassay can be developed to enable the test to enter any routine lab.
9	I would really like to be involved in biomarker research / translation but don’t have the opportunity in my current position.
10	My own experience has been to witness a progressive loss of research staff and support for academic clinical biochemistry in the setting of a diagnostic laboratory. My academic involvement is now minimal as a result of the need to support the diagnostic service as we have shed academic staff (from 7 to 3).
11	Note I have been in both a diagnostic and research environment for 35y. Question 18 and 19 miss out MRM based peptide analysis by electrospray MSMS. This is the technique already in routine use!
12	Proteomics has been around for a great number of years with many millions spent on research - have yet to see anything that is of any true pragmatic use that would add anything to the value of already existing diagnostics - the research findings are generally greatly exaggerated but with little true value that never gets anywhere. Mainstream diagnostic labs need to focus on proper clinical evaluation of existing markers thus allowing a clearer picture of which tests we should be using, how we should be using them and indeed which we should stop using. Multi-marker technology research is hugely flawed and very dependent on the population used - the danger is that some labs/clinicians with more money than sense buy the propaganda that these biomarkers are of use - thankfully in the UK we have more sense or at least financial constraint that only lets tests through that really deliver – i.e. even BNP has not completely broken through because the evidence that it actually does what it says on the tin is actually lacking - with these barriers in place, the introduction of biomarker array/proteomic markers getting through to frontline medicine is unlikely. Sorry for warbling on but I do get frustrated when such a lot of money is being spent on biomarker technology that will never be of any use.
13	Questions 18 and 19 impossible to answer Yes they are viable but only if finance, staff time and expertise is available.
14	Some of the questions presuppose erroneously that none of the technologies listed are in current use in diagnostic laboratories.
15	The current financial climate is likely to have a big impact on the work undertaken by clinical diagnostic laboratories. Now more than any other time during my 7 years working in clinical diagnostic labs the emphasis is on routine diagnostic work. Research and even method development for new biomarkers is ‘discouraged’ unless a strong financial incentive for the laboratory can be demonstrated.
16	The staff- and finance-related pressures on the diagnostic laboratories that I am familiar with are already great. They are likely to become greater. It is inefficient to dissipate research of this type across numerous laboratories in which it is likely at best to be a minor interest. It should be concentrated in academic and commercial laboratories until translational research is required, when diagnostic laboratories may have a role.
17	There is a broad spectrum of Clinical Biochemistry laboratories in the UK. Only a few of these will have the staff and instrumentation to perform the research necessary to develop proteomics based assays.
18	Viability will depend on finance - also sometimes early biomarkers may seem a good idea only to be proven not so good later [e.g. the inventor of PSA now recognises it has been a BAD thing where it is used in screening].
19	You have assumed that biomarkers are a proteomic issue. Actually, all the ‘omics’ are relevant. Of these the most important and relevant to us is metabolomics dependent on NMR and MS.

A total of 30 respondents made additional free text comments. Those relating to technical issues such as having entered multiple responses or any information which would allow the identification of the responder have been excluded.