An Online Tool for Improving Biospecimen Data Element Reporting

Abstract

Detailed documentation of the experimental materials and methods is essential for the validation of scientific papers. Human biospecimens are increasingly utilized as materials in cancer research and information about the biospecimens used is a component of this documentation. We hypothesized that previously reported biospecimen data are inadequate for accurate replication and/or validation of a substantial proportion of studies. To examine this issue, we analyzed biospecimen reporting in a representative cross section of publications over the past 12 years (1998, 2004, 2010) in the journals, Cancer Research (CR, n=46) and Clinical Cancer Research (CCR, n=73). We assessed biospecimen data in relation to the standards outlined as the Tier 1 recommended data elements from the Biospecimen Reporting for Improved Study Quality (BRISQ), in addition to ethics criteria. These data elements encompass features of biospecimens influenced by the patient, medical procedure, and biospecimen acquisition, handling and storage processes. Analysis found that while there was a significant increase in the reporting of ethics board approval status (p<0.008) and name of the ethics board (p<0.0001), there were no significant differences between these journals or over this period in reporting other biospecimen-related data elements. Of the 15 Tier 1 data elements assessed in CR and CCR, the data elements commonly obtained from the “Clinical Chart” (8/15 elements) were significantly better reported than elements that would typically be obtained from the “Biobank” (p<0.0001). Our findings demonstrate that reporting of biospecimen-related data elements has been incomplete. As one part of the solution to this issue, we propose the use of an online data-elements reporting tool (www.biobanking.ca) by biobanks. This BRISQ Report tool aims to help biobanks provide the relevant biospecimen-related data as a structured report, and to promote its inclusion as supplementary material in publications to improve the quality of future research studies.

Introduction

In order for scientific research published in articles to be properly assessed, sufficient data should be available within the description of materials and methods to enable validation of the experiments. Biospecimens are increasingly used as important material components in cancer research. Without proper annotation and the reporting of relevant biospecimen information, it is easy for unrecognized bias to be introduced into research projects using biospecimens.¹ For example, variability in biospecimen collection time can impact molecular data and gene expression,² while prefixation, fixation, processing, and storage have all been shown to influence immunostaining of tissues and the subsequent research conclusions.³ There is currently significant variation, and what we believe are deficiencies in the information provided by publications on biospecimen-specific variables. This perception of the need for better information about the biospecimens has stimulated several groups to make recommendations on relevant clinical, biospecimen, and biobanking variables to report. These recommendations have included the ‘Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK)’ guidelines,⁴ the Standard PREanalytical Code (SPREC),⁵ and the Biospecimen Reporting for Improved Study Quality (BRISQ)⁶ criteria, among several examples listed on the EQUATOR project site,⁷ which provides an extensive listing of guidelines for health research. In particular, the BRISQ criteria have provided comprehensive recommended data elements applicable to biobanking and application in research using either solid or fluid biospecimens.

We hypothesized that previously reported biospecimen data in cancer research publications are inadequate for accurate replication and/or validation of studies, and that the adoption of improved reporting standards requires the development and dissemination of an electronic tool to aid biobanks and researchers to compile and report the information.

Our short term goals were to 1) conduct a baseline assessment of the current reporting status of biospecimen information in research publications, and 2) to develop an open access online tool based on the criteria listed in the BRISQ manuscript that could allow biobanks to generate a “BRISQ Report”. Ideally, this report would accompany material (biospecimen and data) released to researchers and become part of the supplementary data accompanying article submission for publication. Our future goal is to assess the impact of such a tool on the adoption of BRISQ reporting recommendations.

Our analysis of biospecimen reporting in two scientific journals over the past decade shows that there are indeed variations and deficiencies in the reporting of elements determined to be important in the BRISQ manuscript. These deficiencies have an impact on the ability to validate studies, and new approaches are needed to improve the standards for reporting of biospecimen-related data elements.

Materials and Methods

Literature review and inclusion criteria

We set out to analyze biospecimen reporting in publications using human biospecimens over the past decade. The selection criteria for these journals were as follows: must be well-established; must be listed on PubMed, the leading biomedical citation database; must have a recent impact factor in the top 25% of oncology journals (>4.5 in 2010); and must have an editorial policy encouraging a wide variety of cancer research areas. The two journals were selected on the basis that together they represent a broad spectrum of basic, Cancer Research (CR), and translational and clinical cancer research, Clinical Cancer Research (CCR).

Using 2010 as a recent year preceding the initiation and publication of the SPREC (2010) and BRISQ (2011) articles, and for which we had full access to all papers in both CR and CCR journals, three publication years at 6-year intervals were selected to cover more than a decade: 1998, 2004, and 2010. For each year, we identified and analyzed all publications using biospecimens within the first issue (January 1 issue). Biospecimens were defined as any human biospecimens or biological material comprised of whole solid tissues, cells isolated from solid tissues, and fluids or blood.⁸

Analysis of reported data elements

The accuracy and reporting frequency of 15 elements from the BRISQ criteria that comprised the first tier of elements that were “recommended to be reported” was assessed. The 15 Tier 1 BRISQ data elements were previously categorized by the BRISQ Committee as information that might influence the integrity, quality, and/or molecular composition of biospecimens; they related to five phases based on the lifecycle of a biospecimen. We modified and further defined the Shipping temperature element number 14 as two phases; 14A) from the patient to location of preservation or research use, and 14B) from storage to location of research use. Three additional non-BRISQ elements, which are not directly relevant to biospecimen quality but are relevant to use of biospecimens, related to reporting of research ethics information were also included in our analysis to provide a comparison with other data elements intrinsic to most biospecimen research studies (Table 1). The reporting status for each of these 15 BRISQ and 3 non-BRISQ data elements was determined using the following definitions: 1) Not reported, 2) Unclear, inferred, or assumed, 3) Reported, or 4) Not applicable (Table 2). Specific definitions and guidelines for each data element and each reporting status were developed (Appendix Table A1 with a small pilot set of articles and through group discussion by the following three individuals described. The review and analysis was performed by two trained individuals, one Bachelor of Science graduate with 3 years' experience as a summer student in a tumor biobank and research laboratory, and one Master of Science graduate with 2 years' experience in a research laboratory and 1 year in biobanking (AC and SC). Any disagreements in classification or difficult-to-interpret studies were reviewed and arbitrated by a physician with a specialty degree in anatomic pathology and over 20 years' experience in cancer research and tumor banking (PW).

Table 1.

List of Biospecimen Data Elements Examined ^*

Category		Data Element	Examples
I. Preacquisition
	1	Biospecimen type	Tissue, ascites, blood
	2	Anatomical site	Right breast
	3	Disease status of patients	Diabetic, healthy control
	4	Clinical characteristics of patients	Age, gender, treatment information, pre-menopausal breast cancer patients
	5	Vital state of patients	Postmortem
	6	Clinical diagnosis of patients	Breast Cancer
	7	Pathology diagnosis	Her2-negative intraductal carcinoma
II. Acquisition
	8	Collection mechanism	Fine needle aspiration, biopsy, surgery, operation, surgical resection
III. Stabilization/preservation
	9	Type of stabilization	Heparin, on ice
	10	Type of long-term preservation	FFPE, frozen, fresh
	11	Constitution of preservative	10% neutral-buffered formalin, 10 USP heparin units/ml, FFPE, OCT
IV. Storage/transport
	12	Storage temperature	−80°C, 20°C–25°C
	13	Storage duration	8 days, 5 to 7 years
	14A	Shipping temperature A – from patient to preservation or research use	−170°C to −190°C, room temperature
	14B	Shipping temperature B – from storage to research use	−170°C to −190°C, room temperature
V. Quality assurance measures
	15	Composition assessment and selection	Minimum 80% nuclei & maximum 50% necrosis
Ethics Criteria^**
	16	Documentation of consent	Consent was obtained from all patients
	17	Documentation of Research Ethics Board (REB) or Institutional Review Board (IRB) approval status	The trial and study received appropriate ethics approval
	18	Name of REB/IRB	The study was approved the University of British Columbia REB

Adapted from BRISQ Recommended List.⁶

Not part of Tier 1 BRISQ Recommended List.

Table 2.

Reporting Status of Data Elements

Reporting Status	Explanation
Not reported	No mention of the data element
Unclear, inferred or assumed	Information relating to the data element is unclear, can only be indirectly inferred by an expert or requires an assumption by the reader
Reported	The data element was clearly stated
Not applicable	The data element was not applicable to the biospecimen type or application

Statistical analysis

Statistical analysis of literature review results was conducted using Chi-squared tests for significance. All statistical tests were performed using GraphPad Prism 5 (GraphPad Software Inc., La Jolla, CA).

Development of a BRISQ report tool

We developed an online tool that provides a structured and comprehensive checklist for reporting BRISQ data elements. The BRISQ Report tool was developed using an open source software/web-enabled database, MySQL (Oracle Corporation, Redwood Shores, CA) and is hosted by the University of British Columbia (UBC) Faculty of Medicine IT network. The software stacks that make the core of the system include Apache and PHP in Windows 2008 R2 platform. The BRISQ Report tool is designed to allow users to selectively enter the biospecimen-related data elements recommended in the BRISQ guidelines, with tabs separating the elements by each of the three tiers. Users are not required to fill in all fields but can populate specific fields relevant to their biospecimens. Some fields consist of drop-down lists to facilitate standardized data entry where possible. Upon completion of the form, submission of the data to the database enables generation of a summary of the all data elements entered and an option to download a portable document format (PDF) of the report. The values in the dropdown list (or controlled data) are taken from examples in the BRISQ and/or SPREC articles.

Results

Proportion of articles using biospecimens

In total, 234 articles from both journals were reviewed from the three selected publication years and 51% (119/234) were found to have used human biospecimens. The breakdown of articles using biospecimens was 36% (46/127) in CR and 68% (73/107) in CCR (Table 3).

Table 3.

Publications Using Biospecimens

Publication Year	Cancer Research	Clinical Cancer Research
1998	44% (12/27)	60% (21/35)
2004	29% (17/58)	69% (27/39)
2010	40% (17/42)	76% (25/33)
Total	36% (46/127)	68% (73/107)

Reporting of ethics criteria relevant to use of biospecimens

All 119 articles were assessed for 3 ethics-related data elements. There was a significant trend in better reporting of the ethics criteria over the period studied, with increasing frequency of documentation of Research Ethics Board (REB)/Institutional Review Board (IRB) approval status (Fig. 1B, p<0.008) and the name of the REB/IRB (Fig. 1C, p<0.0001), but there was no significant change in documentation of consent (Fig. 1A). Analysis of all three ethics criteria combined showed a significant improvement in reporting frequency over time (Fig. 1D, p<0.0001).

FIG. 1.

Reporting status of ethics criteria data elements in Cancer Research and Clinical Cancer Research journals from 1998-2010. (A) Documentation of consent; (B) documentation of Research Ethics Board approval; (C) name of Research Ethics Board; and (D) all elements combined. Differences in reporting status over the time period were significant in B, C, and D, p<0.05. A color version of this figure is available in the online article at www.liebertpub.com/bio.

Reporting of Tier 1 BRISQ elements relevant to use of biospecimens

The reporting of the 15 Tier 1 BRISQ elements was assessed in both CR (Fig. 2A) and CCR (Fig. 2B) journals. Overall, we found that there was no change across this period in the reporting frequency of the data elements in either journal. The CCR articles had overall higher reporting numbers than CR across most of the criteria examined. For both journals, the three most frequently reported data elements were Biospecimen type (100% and 100%, in CR and CCR, respectively), Disease status (98% and 100%), and Clinical diagnosis (89% and 97%). In CR, the criteria that were reported infrequently (40%–60%) were Clinical characteristics at 41% (19/46), Pathological diagnosis at 48% (22/46), and Type of long-term preservation at 48% (22/46). Elements that were rarely/never reported (<20%) were Storage duration at 15% (7/46), Type of stabilization at 9% (4/46), and Shipping temperatures at 0%. In contrast, CCR had high reporting rates for Clinical characteristics at 84% (61/73), Pathological diagnosis at 67% (49/73), and Type of long-term preservation at 70% (51/73). Similarly, elements rarely/never reported included Storage temperature at 18% (13/73), Type of stabilization at 14% (10/73), and Shipping temperatures at 0%.

FIG. 2.

Reporting frequency of Tier 1 BRISQ criteria in (A) Cancer Research and (B) Clinical Cancer Research from 1998–2010. The criteria were divided into the two source categories based on the most common source of data for these elements. Within the “Clinical Chart” source category, the data elements BRISQ categories were: I. Preacquisition and II. Acquisition. Data elements associated with the “Biobank” source category consisted of III. Stabilization/preservation, IV. Storage/transport, and V. Quality assurance measures. A color version of this figure is available in the online article at www.liebertpub.com/bio.

Comparison of data elements typically obtained from the “Clinical Chart” versus the “Biobank”

We classified the data elements into those that are typically obtainable from the “Clinical Chart” (I. Preacquisition and II. Acquisition categories) and those typically only obtainable from the biospecimen collection process or the “Biobank” (III. Stabilization/Preservation, IV. Storage/Transport, and V. Quality Assurance). The reporting status of data elements from the Clinical Chart were 80% (757/952) Reported, 8% (72/952) Unclear, inferred, or assumed, and 13% (123/952) Not reported. Biobank-obtained data elements had reporting frequencies of only 39% (328/850) Reported, 1% (12/850) Unclear, inferred, or assumed, and 60% (510/850) Not reported. Using these two distinctions for comparison, it was found that data elements commonly obtained from the Clinical Chart were reported at a significantly higher rate than those elements obtained from the Biobank (p<0.0001) (Table 4).

Table 4.

Reporting Status of Data Elements Grouped by Typical Source from “Clinical Chart” or “Biobank”

Reporting Status	Clinical Chart	Biobank
Reported	757^*	328
Unclear, inferred, or assumed	72	12
Not reported	123	510
Total	952	850

Data elements typically obtained from the Clinical Chart are reported at a higher rate than those from the Biobank, p<0.0001.

Discussion

Over the past 2–3 decades, there has been a rapid expansion of research using human biospecimens such that these are now essential materials in all levels of research from discovery to clinical validation. Replication and validation are essential steps in all research; one important example is the specific process of translating basic research findings in cancer research into prognostic and predictive biomarkers that can impact cancer care. Recognition of the need for reporting standards around patient cohorts used for biomarker studies that could facilitate replication of results and minimize cohort selection bias, stimulated the development of the REMARK criteria and the recent publication of several biospecimen quality-focused articles.^9–12 Subsequently, appreciation of the potential influence of additional biospecimen variables on creating bias in biospecimen-based research raised the issue of developing complementary additional reporting standards focused on clinical and biobanking variables relevant to biospecimens. In light of this, the BRISQ list of recommended elements was generated through the deliberations of an expert committee that included members from the National Cancer Institute Office of Biorepositories and Biospecimen Research (OBBR), the International Society for Biological and Environmental Repositories (ISBER), and the REMARK group. As detailed in the original BRISQ article and listed on the EQUATOR site,⁷ potentially relevant criteria were identified and separated into three tiers, with Tier 1 elements recommended to report, Tier 2 considered beneficial to report, and Tier 3 considered as other potentially relevant elements. In this study, focused only on reporting of Tier 1 elements, we have found that reporting of biospecimen-related data elements used in cancer research studies is frequently incomplete and has not improved over a period of more than a decade, despite a greater than 2-fold increase in use of biospecimens in this period.⁸ Of the 15 Tier 1 BRISQ data elements considered to be essential for validation of research findings involving biospecimens,⁶ we found a significant deficiency in reporting of those elements that would typically be obtained from the biobank or biobank processes.

The impact of failure to report data clearly related to some of the specific BRISQ elements can be debated. The relevance of some elements may vary with the focus of the research; some that are based on stable biological factors such as DNA mutation may be less important to document in replicating findings than those derived from RNA expression profiles. For example, we have noted that the Tier 2 BRISQ element Collection time can differentially influence expression of some breast tumor biomarkers, and that different biobanks can have very different distributions of collection times across their stock.² Therefore, both collection time for individual biospecimens and overall biobank of origin for cohorts can be factors to consider in assessing and validating studies, but only in some circumstances. Definition of what constitutes a “lack of clarity” in reporting detail may also relate to the expertise of the reviewer, but it should be noted that the authors here encompass a range of experience from biobank technician to graduate student to established investigator—all with knowledge of biospecimens. The feasibility of detailed documentation of some factors also needs to be taken into consideration.

Our study supports previous findings of poor reporting of other recommended elements relevant to tumor biomarker studies,^13,14 and confirms that replication of experiments in a significant proportion of articles in leading cancer research journals may be hindered by incomplete reporting of data relevant to biospecimens: the underlying issues are multifactorial. Refinement of which elements are most important to report based on experimental data is important to validate or modify the initial BRISQ criteria, which are largely based on empirical existing standards, which provide a necessary starting point. Awareness of these elements and reporting standards should be communicated widely among researchers, journal reviewers, and editors. Review and dissemination of protocols to improve BRISQ criteria should reduce the effort involved and facilitate and promote the adoption of better standards. Finally, although it is anticipated that the impact of improved reporting will result in greater success in the translation and validation of research in the future, processes to measure and validate this assumption will also be required.

One approach to promote the delineation of the BRISQ elements by researchers is for biobanks that support research users to also provide a detailed summary of the relevant biospecimen data. The BRISQ Report tool was created to facilitate reporting quality by enabling biobanks to provide a structured and standardized report of biospecimen data based on all tiers of the BRISQ recommended list of elements. The first release of this tool is openly accessible for any biobank, and available online at www.biobanking.ca (Fig. 3). In this first iteration, users must enter all data on each occasion to generate the report. Plans for the second release of the tool include additional features such as user account and login capability authenticated by username/password, individual data storage for each biobank user, and user-defined templates allowing reduction of data entry time for similar multiple data releases. We anticipate that biobank users will value the ability to create a PDF BRISQ data element report that can then be provided to research users as an additional ‘product’ (along with typical products such as biospecimens, extracted nucleic acids, patient and clinical data reports, etc.). Once the research study has been completed and submitted for publication, this BRISQ Report can be included with other supplementary data as an appendix. This may facilitate review by the journal by clearly delineating biospecimen factors and, perhaps, assist in successful publication.

FIG. 3.

The BRISQ Report tool. Screenshot of part of the first page of an online tool to allow users to enter biospecimen-related information based on the BRISQ recommended list and generate a summary report. Placement of the cursor into the data field box displays the definition of the data element.

In conclusion, our literature review and analysis demonstrates that reporting of Tier 1 BRISQ recommended elements obtained from biobanks has been poor. To overcome this, we have created a simple tool for biobanks to report biospecimen-related data that will offer a complete and informative representation of the biospecimens used in research. Better reporting of biospecimen information allows researchers to have a clearer representation of studies conducted, improve reproducibility of results, and ultimately increase the overall quality of biospecimen-based research.

Footnotes

Acknowledgment

The authors would like to acknowledge the BRISQ committee for their contribution to defining important biospecimen variables for reporting which was the basis for development of the BRISQ Report tool.

Author Disclosure Statement

The authors have no conflicts of interest of financial ties to disclose.

Appendix Table A1.

Data Element Reporting Guidelines

		Reporting Status
	Data Element	Not Reported	Reported	Unclear, Inferred, or Assumed	Not Applicable
1.	Biospecimen type	No mention of biospecimen type used	Biospecimen type clearly stated
2.	Anatomical site	No mention of anatomical site biospecimen taken from	Clearly states where the biospecimen was taken from. Must be specific enough that another scientist could accurately replicate the projectNOTE: If article uses both tissue and blood biospecimens, and gives the anatomical site for the tissue biospecimen, but not blood, status will be “Reported”.If article uses bone marrow, site of acquisition reported is required. If the article states a vague site, but has more specific anatomical site information supplementary data, then status will be “Reported”	Information indirectly inferred/assumption made/unclearNOTE: If the article states disease type (e.g., melanoma, small cell cancer, sarcoma, osteosarcoma, soft-tissue sarcoma, medulloblastoma, mesothelioma, head and neck, glioblastoma, leukemia) that could be from multiple sites but a reasonably knowledgeable scientist would not know the site from the given term, will be reported as “Unclear, inferred or assumed”.
3.	Disease status of patients	No mention of whether patient has disease or is a healthy control	Disease status of patient clearly statedNOTE: If article states a patient underwent surgery to remove lesions of a certain type of disease, we can assume the patient's status is diseased.
4.	Clinical characteristics of patients	No clinical characteristics of patients mentionedNOTE: If article states clinical information obtained but does not show that information anywhere, status will be “Not reported”	Clinical characteristics of patients are mentioned in methods, results or supplementary data
5.	Vital state of patients	No mention of whether patient was alive or deceased at time of biospecimen acquisition	Vital state of patient clearly stated or can be clearly assumed from collection methodNOTE: If patient underwent surgery to remove a tumour, we can assume the vital state is alive, will be “Reported”.	Information indirectly inferred/assumption made/unclearNOTE: If TMAs were obtained from a biobank or previous research group, no mention of surgery, biopsy or resection is found and no definitive information on vital state is available, status will be “Unclear, inferred or assumed”.
6.	Clinical diagnosis of patients	No mention of clinical diagnosis of patient, and a reasonable assumption cannot be made with the information provided	Clinical diagnosis is clearly stated with supporting information. May be obtained from the paper itself or supplementary dataNOTE: If article states that a tumour of the breast was removed from a patient then it can be assumed the clinical diagnosis is breast cancerIf TMAs used, the clinical diagnosis can be assumed if the tumour type is stated clearly.
7.	Pathology diagnosis	No mention of pathology diagnosis or histological review	Paper clearly states biospecimens were reviewed by a pathologist, (e.g., Dr.____), was histologically analyzed, or reviewed and graded according to WHO guidelines.	Information indirectly inferred/assumption made/unclearNOTE: Articles with supplementary data tables containing grade or pathology information, but no direct statement of pathology review, status will be “Unclear, inferred or assumed”For articles that use blood, status will be “Unclear, inferred or assumed” if a clinical diagnosis has been made.
8.	Collection mechanism	No information provided regarding biospecimen collection mechanism.	Method of biospecimen collection is clearly stated.	Information indirectly inferred/assumption made/unclearNOTE: If article states tissue obtained from biobank, or TMA obtained from some research group but collection mechanism not stated, status will be “Unclear, inferred, or assumed”
9.	Mechanism of stabilization	No information on type of stabilization.NOTE: If article states tissue “immediately snap-frozen” or fixed without reporting the specific location of the fixation, status is “Not reported”. If the location is generally stated as “in pathology” or “biobank”, status is “Not reported” because unknown if stabilization occurred at point of biospecimen collection or harvesting.	The type of stabilization between the time biospecimen taken from patient prior to preservation or use and location in which stabilization occurred at point of harvesting is stated. Must state that stabilization occurs at point of harvesting (e.g., in the OR or frozen section room).
10.	Type of long-term preservation (Format of biospecimen)	No information on format of biospecimen for long-term preservation	Article states whether biospecimen is frozen, FFPE or fresh.	Information indirectly inferred/assumption made/unclear	No long-term preservation was required, biospecimen used immediately in research assay.NOTE: If biospecimen is immediately created into a cell line/cell culture then status will be “Not applicable”.
11.	Constitution of preservative	No information on what the biospecimen is preserved inNOTE: If article states frozen in liquid nitrogen,	Article states what the biospecimen is preserved in.		Biospecimen used immediately in research assay.NOTE: If biospecimen is immediately created into a cell line/cell culture then status will be “Not applicable”.
12.	Storage temperature	No information on temperature provided	Storage temperature of biospecimens is clearly statedNOTE: If states frozen in liquid nitrogen, status will be “Reported” because temperature of liquid nitrogen is known.	Information indirectly inferred/assumption made/unclearNOTE: If known FFPE, a reasonably knowledgeable scientist could infer that storage is at room temp, status will be “Unclear, inferred or assumed”	Biospecimen used immediately in research assay.NOTE: If biospecimen is used to make a cell line/cell culture then storage temperature, then status will be “Not applicable”
13.	Storage duration	No information on when biospecimen was collected, or storage duration	Article provides information on storage duration or period of time when biospecimens were obtained		Biospecimen used immediately in research assay.NOTE: If biospecimen is used to make a cell line/cell culture then storage duration then status will be “Not applicable”
14.	Shipping Temperature
	A Phase 1	No information on shipping temperature	Shipping temperature clearly statedNOTE: If stated shipped on dry ice, status will be “Reported” because dry ice temperature is known.
	B Phase 2	No information on shipping temperature	Shipping temperature clearly statedNOTE: If stated shipped on dry ice, status will be “Reported”		Biospecimen used immediately in research assay.
15.	Composition assessment and Selection	Reasons for choosing specific biospecimens in experiments is not stated	Selection criteria or composition assessment of biospecimens is statedNOTE: If article states randomly selected, status will be “Reported”
16.	Documentation of consent	No statement made regarding patient consent	Patient consent to participate is stated or waiver of consent obtained from REB/IRB
17.	Documentation of REB approval	Article does not state REB/IRB approval	Paper clearly states REB/IRB approvalNOTE: If article gives REB protocol # status will be “Reported”
18.	Name of REB/IRB	No name of REB/IRB institution stated	Paper clearly states name of REB/IRB board	Infer REB/IRB board usedNOTE: If article states ethics approval by Institutional Review Board, and we can infer which institution

For any of the data elements:

• if article references a previous article by the same group with all appropriate information, status will be “Reported.”

• if article references a previous article by the same group with inaccurate or incomplete information, status will be “Unclear, inferred or assumed.”

• if article references a previous article by a different group, then status will be “Not reported.”

References

Grizzle

, Bell

, Sexton

. Issues in collecting, processing and storing human tissues and associated information to support biomedical research. Cancer Biomarkers, 2011; 9:531–549.

Barnes

, Parisien

, Murphy

et al. Influence of evolution in tumor biobanking on the interpretation of translational research. Cancer Epidemiol Biomarkers Prev, 2008; 17:3344–3350.

Engel

, Moore

. Effects of preanalytical variables on the detection of proteins by immunohistochemistry in formalin-fixed, paraffin-embedded tissue. Arch Pathol Lab Med, 2011; 135:537–543.

McShane

, Altman

, Sauerbrei

et al.

Reporting recommendations for tumor marker prognostic studies (REMARK)

J Natl Cancer Inst, 2005; 97:1180–1184.

Betsou

, Lehmann

, Ashton

et al. Standard preanalytical coding for biospecimens: Defining the sample PREanalytical code. Cancer Epidemiol Biomarkers Prev, 2010; 19:1004–1011.

Moore

, Kelly

, Jewell

et al. Biospecimen reporting for improved study quality. Biopreserv Biobank, 2011; 9:57–70.

Equator Network. 2011[cited 2012 May 30, 2012]http://www.equator-network.org. 2012 November 15.

Hughes

, Watson

. Biospecimen use in cancer research over two decades. Biopreserv Biobank, 2010; 8:89–97.

Riegman

, Morente

, Betsou

et al. Biobanking for better healthcare. Mol Oncol, 2008; 2:213–222.

10.

Betsou

, Barnes

, Burke

et al. Human biospecimen research: Experimental protocol and quality control tools. Cancer Epidemiol Biomarkers Prev, 2009; 18:1017–1025.

11.

Simeon-Dubach

, Perren

. Better provenance for biobank samples. Nature, 2011; 475:454–455.

12.

Lim

, Dickherber

, Compton

. Before you analyze a human specimen, think quality, variability, and bias. Anal Chem, 2011; 83:8–13.

13.

Hemingway

, Philipson

, Chen

et al. Evaluating the quality of research into a single prognostic biomarker: A systematic review and meta-analysis of 83 studies of C-reactive protein in stable coronary artery disease. PLoS Med, 2010; 7,6:e1000286.

14.

Mallett

, Timmer

, Sauerbrei

et al. Reporting of prognostic studies of tumour markers: A review of published articles in relation to REMARK guidelines. Br J Cancer, 2010; 102:173–180.