Introduction

High-quality biobanks are important resources for health research, including basic and genetic research, investigations in personalized or stratified medicine (genetic and other biomarkers), and research in widespread diseases. The development of large-scale, population-based as well as disease-specific biobanks has come with new ethical challenges. New consent procedures are needed because big as well as small research biobanks are increasingly practicing long-term storage of biospecimens and data [1,2]. These stored materials and data could be used for future research projects that are unspecified—and even to some extent unforeseen—at the time of the donor’s consent. Some biomaterial donors, therefore, are asked to give consent to a framework for future research of certain types (“broad consent”) instead of the standard consent to a single research project (“project-specific consent”) [3,4]. Other studies might employ so-called “dynamic consent” models [5]. Single research projects based on project-specific consent still exist in parallel.

The increased use of the broad consent model also affects how competent bodies review and approve biospecimen research. To grasp these implications, it is helpful to distinguish two steps in the review and approval of biospecimen research operating under a broad consent model [6]. First, the biobanking itself can be approved (“bank approval”). This approval concerns the “framework of oversight procedures,” such as the broad consent form, data protection model, and other policies relevant to the biobank’s governance. Second, the same or another competent body can approve specific projects that require the release of particular biomaterials and data (“release approval”). The other common scenario is a single research project that collects, analyzes, and then disposes of all remaining specimens. Such single research projects still work appropriately with a standard ethics approval.

Meaningful ethics reporting in this area needs to go beyond standard sentences such as “consent was obtained from all participants” and “the project was approved by the local ethics committee.” From an ethical viewpoint, it is unclear whether “approval” comprised both the banking and the use of material or only one. Thus, the reporting becomes meaningless or even misleading in cases in which, for example, the initial approval given did not include the reuse of biospecimens.

The “Biospecimen Reporting for Improved Study Quality” (BRISQ) guideline [7], the “Guideline to Standardize the Citation Of BioResources in journal Articles” (COBRA) [8], and the “Minimum Information About BIobank data Sharing” (MIABIS) guideline [9] do not include specific requirements for ethics reporting. Likewise, neither the International Committee of Medical Journal Editors (ICMJE) nor the Committee on Publication Ethics (COPE) currently provide specific guidance on ethics reporting in biospecimen research.

Cheah et al. found that 35%–40% of studies with human biospecimens published in two cancer journals between 1998 and 2010 reported on informed consent and that reporting frequency of ethics approval increased from 25% to 60% [10]. However, this study did not assess or discuss whether ethics reporting was meaningful in light of the above-described complexity of consent and approval.

Status Quo: Ethics Reporting in Biospecimen Research

We systematically screened ethics statements from a total of 120 articles reporting on analyses of human biospecimens published between 2014 and 2015 in PLOS Genetics (n = 30), The New England Journal of Medicine (n = 30), Nature Genetics (n = 30), and Nature Medicine (n = 30). S1 Table presents the extracted ethics reporting of all 120 papers. S1 Text presents further details on the search and selection of studies and on the analysis of ethics statements.

The proportion of articles derived from biobank studies was 47% (n = 56), from clinical drug trials 16% (n = 19), and from other clinical studies 38% (n = 45). Only 3 studies (3%) lacked any reporting of either consent or approval. 20% (n = 24) reported on only one category.

A total of 87% (104/120) of articles reported at least some information on consent. Of these 104 articles, 41% (43/104) reported additional details such as proxy consent, timing of consent, consent obtained from another study or project, or data protection specifications. Only 8% (n = 8) of the 104 consent-reporting articles provided information that helped to understand whether the consent was broad or project-specific (S1 Table).

Ninety-three percent (112/120) of the articles reported at least some information on approval. Of these 112 articles, 79% (89/112) reported additional details, such as the name of the institutional review board (IRB) or the approval protocol number. Only 9% (n = 10) of the 112 articles provided information that helped to understand whether the approval was granted for the banking of specimens for future research, for a specific project, or for both. Of these 10 articles, 2 reported on bank approval, 4 articles reported on study-specific approvals, and 4 articles reported on both bank and study approvals (S1 Table).

In brief, most papers on biospecimen research reported on consent and approval but did it in a nonmeaningful way. We must stress that nonmeaningful ethics reporting does not imply inadequate consent and approval processes. For a recent status quo analysis of consent procedures in biobank research, see [11].

Our study included four articles that were all based on the same biobank, deCODE Genetics. The reporting on consent and approval differed substantially across the four articles. Two articles reported details that helped the reader to understand the broad consent, biobank approval, and project-specific approval involved. The other two studies only used standard phrases on consent and approval that were not helpful for an appropriate understanding of the consent and approval issues.

It is interesting to see that some authors already realized the need to report more specifically on consent and approval, even though there is as yet no official guidance. Some examples and variations of meaningful reporting on consent and approval are presented and commented on in Table 1. In the majority of cases, unfortunately, even the more detailed information did not allow us to determine the type of consent or approval. We illustrate this with typical examples in S2 Table.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Selected ethics statement (see also S1 Table) with specifying reporting on informed consent or ethics approval or both.

https://doi.org/10.1371/journal.pbio.1002521.t001

Besides the need to report more specifically on the type of consent and approval, our analysis also discovered that some researchers further report on additional ethical issues such as transborder use, data protection, and handling of protected health information. See Table 2 for examples of such additional ethics statements.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Examples of additional details on consent and approval extracted from ethics statements.

https://doi.org/10.1371/journal.pbio.1002521.t002

We also found differences in the justifications of why no informed consent was needed for anonymized or de-identified samples. One paper, for example, just mentioned that samples were de-identified and that their study was approved. Others explicitly justified research with anonymized samples via project-specific conditions and safeguards. One study, for example, highlighted that samples “were not and will not be used for any other purpose.” Some mentioned reasons for waivers of consent that would need more explanation. For example, it remains unclear why no consent is needed for research with “discarded clinical material.” The fact that personalized material was discarded or “left-over” after surgery does not directly justify research without consent. Similarly unclear cases for research without consent were research with “cadaveric organ transplant donors.” Recent debates and interview research with IRB leaders in the United States highlighted the uncertainties on whether anonymization in biobank research still suffices for waivers of consent [12].

A Roadmap for Improving Meaningful Ethics Reporting

How can the described challenges be addressed in an ethical and pragmatic manner? First, we recommend that ICMJE and COPE as well as individual journals revise their policies on ethics reporting and ask for more meaningful information on:

1. whether project-specific consent, broad consent, or any other consent type was obtained from biospecimen donors; and
2. whether the biobank that collected, stored, and/or released the biospecimen was approved, by which body it was approved (and if not, why not), and whether the particular project was approved and by which body (and if not, why not). If no separate biobank was involved (that is, the research project collected, stored, used, and then disposed of all remaining biospecimens), then this should be explicitly mentioned to avoid misunderstandings.

Furthermore, it is helpful in most cases to know whether the samples used in the study were disposed of at the end of the study or whether remaining samples were stored for reuse. In Table 1, we demonstrated how more meaningful ethics reporting might look without necessarily increasing the word count. Our examples of clearer ethics reporting in Table 1 were restricted to basic information. Future discussion and meta-research should clarify whether even more reporting is needed in particular cases (for example, on whether consent on transborder use of samples was obtained or whether the study’s data protection model was explicitly approved).

Reporting guidelines on biospecimen research (e.g., BRISQ, COBRA, MIABIS) should be revised to add these requirements. We also recommend that each individual biobank include an ethics reporting template into their material transfer agreements to avoid variation in reporting by studies linked to the same biobank.

We recommend a consensus statement by researchers, journal editors, ethicists, and other relevant stakeholders in this regard. More specific reporting on consent and approval in scientific publications should not only be considered as an intrinsic component of good professional conduct in biobank research but could also support the harmonization of biobank research and will inform the reader about opportunities for cooperation.

Supporting Information