Ethical issues in the use of leftover samples and associated personal data obtained from diagnostic laboratories

Abstract

Diagnostic laboratories are an integral part of the research ecosystem in biomedical sciences. Among other roles, laboratories are a source of clinically-characterized samples for research or diagnostic validation studies. Particularly during the COVID-19 pandemic, this process was entered by laboratories with different experience in the ethical management of human samples.

The objective of this document is to present the current ethical framework regarding the use of leftover samples in clinical laboratories. Leftover samples are defined as the residue of a sample that has been obtained and used for clinical purposes, and would otherwise be discarded.

Secondary use of samples typically demands institutional ethical oversight and informed consent by the participants, although the latter requirement could be exempted when the harm risks are sufficiently small. However, ongoing discussions have proposed that minimal risk is an insufficient argument to allow the use of samples without consent. In this article, we discuss both positions, to finally suggest that laboratories anticipating the secondary use of samples should consider the adoption of broad informed consent, or even the implementation of organized biobanking, in order to achieve higher standards of ethical compliance which would enhance their capacity to fulfill their role in the production of knowledge.

1. Introduction

Diagnostic laboratories, in particular those within medical institutions, are an integral part of the research ecosystem in biomedical sciences. Their participation in research projects is specially appreciated for their proven expertise in the use of validated analytical methods and processes, including the critical aspects of sample identification, traceability and storage; all features that reassure confidence in the results.

While this has been true for decades, it gained prominence by the amount and kind of research carried out during the COVID-19 pandemic. The COVID-19 pandemic not only triggered a large number of research projects involving human samples, but it impressed a sense of urgency to produce meaningful results [1], [2]. Diagnostic laboratories naturally became sources of samples for much-needed investigation of the new disease. This process was entered by laboratories with widely heterogeneous backgrounds in research, and thus with diverse expertise in the ethical management of human samples.

The objective of this document is to present the current ethical framework regarding the use of leftover samples for research or method validation in clinical laboratories. While most studies will demand institutional ethical oversight and informed consent (IC) by the participants, there are a number of situations in which these could be waived off. This article will provide overview on various ethical concepts to understand these situations, discussing in particular the distinctive features of research carried out in diagnostic laboratories.

This article was prepared taking into account recent reviews and presentations on the subject, which capture the current points of concern around the secondary use of leftover samples. Importantly, recommendations outlined in the article have been endorsed by the Task Force Ethics of the International Federation of Clinical Chemistry. Given the global scope intended for the recommendation, the article avoids discussing country-specific regulatory affairs.

2. Leftover samples and associated data

Leftover sample can be defined as the residue of a sample that has been obtained and used for clinical purposes, and would otherwise be discarded [3], [4]. Fractions of samples that must be stored due to legal or technical reasons are not considered leftover, at least until the required storage period is over. Leftover samples can also be found in literature as remnant samples or residual biological material. In the context of this document, we will not discuss remnant fractions of samples from individuals enrolled in prospective studies, even if it involved clinically relevant tests, or samples taken from subjects participating in interventional trials.

One important point to consider is that the biological fluid or tissue is only one part of what patients entrust to laboratory professionals. In medical laboratories, biological samples are always associated with various kinds of personal data. Associated data involves personal identifiers (name, address, for example), health-related information (chronic diseases, previous hospital stays) or other kinds of potentially sensitive data (travel history, family history). Both the nature and the accessibility of the associated data will be crucial to define the ethical requirements for the use of the samples.

In the clinical laboratory, the use of leftover samples could be considered quality control, to generate sample pools or to assemble a panel of control samples [5]. Alternatively, the samples might be useful for research projects, either carried out by external investigators or by investigators affiliated to the lab.

There are multiple advantages in the use of leftover samples and their associated data. First, their availability facilitates the advancement of research, particularly the fields of translational and personalized medicine. These samples represent more accurately the study population, since they are not subject to selection bias during recruitment, thus strengthening the validity of the results. Occasionally, leftover samples may represent experimental conditions which would be impossible to produce in prospective studies, such as old samples from untreated patients when a treatment is now available. Finally, from the ethical point of view, research designed to use leftover samples obviates the need to recruit new participants, and therefore avoids exposing them to the concomitant burden and risks.

While these are compelling arguments in favor of secondary research on leftover samples, their use must be contemplated with caution. The individual that entered a medical laboratory did so with the specific objective of seeking clinically relevant results. Profound ethical implications derive from the use of their sample for other purposes.

3. Ethical concerns regarding the use of leftover samples

Ethical concerns involved in the use of leftover samples can be analyzed in terms of the known principles of respect for autonomy, beneficence and justice. The first concern relates to the commitment to respect the autonomy of the individual providing the sample. This individual approached a health care institution for a clinically relevant test, and the implied consent is valid only for that test. The standard way to deal with the secondary use of this sample would be to ask for permission, usually in the form of an informed consent. There are instances in which it is difficult or impossible to obtain consent for the use of the leftover sample, and sometimes it might not even be considered ethically necessary. Nevertheless, efforts should be done to avoid this situation, given that the informed consent is the most important tool to assert the autonomy of the individual [5].

Another concern relevant to consider the use of leftover samples is the possibility of causing harm to the subject. In other words, the balance between benefits and risks must be satisfactory for the individual involved. In terms of risk assessment, breach of privacy is the most evident and therefore it is crucial to provide for the safety of the associated data. A person whose personal and medical data results exposed becomes at risk of stigmatization, discrimination and overall harm to their integrity. Risks could be related to the exposure of the initially recorded personal information, medical history, and also the newly obtained results. Anonymous samples (those with no recorded personal data) are hardly the case in medical laboratories, but the breach of privacy risk can be minimized when leftover samples are irreversible separated from identity data. Mitigation of the risk of harming an individual is a decisive issue regarding the possibility of using a leftover sample without asking for consent, as discussed in the next section.

There is also the possibility of finding a clinically relevant result, other than the ones that those the patient came looking for; also known as incidental results. Risk of causing distress and anxiety to subjects derive from the possibility of notifying them unsolicited information regarding their identity, their health or even the health of their relatives. Management of this sensitive information obtained from leftover samples should be planned. Verification and notification of incidental results should be analyzed in terms of their clinical importance and the existence of a beneficial intervention [6], [7]. When the use of the leftover sample is planned, a question can be added to the consent process so that the researcher is aware whether the individual would like to know any unexpected result. This particular risk is also avoided when using irreversibly de-identified samples, so that it becomes impossible to assign the incidental result to any one subject. Importantly, another risk to avoid is the loss or deterioration of the sample itself as a consequence of its secondary use. For example, the performance of the clinically relevant test must not be jeopardized to save enough material for future research studies. In this regard, it is not Another related risk to be avoided is the inclination to extract more sample (for example, draw more blood volume) than what would be normally required for the diagnostic process. Hence, it is good standard practice in the laboratory to define the amount of sample before the extraction, as opposed to during the extraction.

A word of caution must be said regarding arguments of beneficence to justify the use of the samples for research. It is incorrect to argue for the use of a leftover sample, either with or without consent, only by citing a “greater good” derived from the results, which will -eventually- benefit the human kind. This line of reasoning is dangerous, as resonant cases have proven before [8], [9]. Still, public interest in the area of health care is frequently cited as an argument for informed consent exemptions, a view that has been reinforced by sample usage and data sharing for the developing diagnostic tools, therapies and vaccines during the COVID-19 pandemic [10].

Finally, there are some considerations regarding the principle of justice. The selection of the samples must be based on technical criteria. Picking samples from vulnerable populations should be avoided. Vulnerable populations are defined as those including persons who have their autonomy already restricted by any other condition (imprisonment for example), or otherwise that it may be regarded that their eventual consent would not be valid (for example, because they are not able to correctly understand what is being asked for). If the objective of a research project is to study a vulnerable population, then the wiser option is to prepare a prospective study and obtain the specific informed consent.

3.1. Rationale supporting the possible exemptions to the requirement of informed consent (IC)

When use of leftover samples is proposed for a project not planned at the time of sample obtention, consent can often be requested by contacting each subject, since laboratories usually retain personal contact information. Nonetheless, sometimes it could be impractical or impossible to look for consent. For instance, a research project may require to use a historical collection of samples stored from long ago. Samples may have lost, or been stored without, identification labels. Some projects may involve the re-analysis of thousands of records. In these and many other occasions, there are arguments to carry out the project considering exemptions to the IC requirement.

First, it is important to establish that IC exemption does not mean lack of ethical oversight. Approval by an independent committee will still be needed, or at least advised, for the use of the samples and their associated data without individual consent.

The rationale behind exemption of the IC requirement is that harm risks are sufficiently small as to overcome the loss of autonomy. Minimal risk is a reasonable assumption given that there is no interaction between researcher and study subject, and there are no physical or psychological interventions on the subject, who incurs in no extra burdens or activities by participating in the study. The main risk is related to disclosure of personal information leading to unintended negative consequences for the individual.

Any such risk is certainly avoided when working with anonymous samples, in which identifiers were not recorded, they were erased irreversibly or, more rarely, the identity was never known. For example, this could entail transferring an aliquot of leftover blood to a new tube re-labeled with a non-linked code. Another example of this situation would be the pooling of samples for quality control. Method verification also poses no risks. Furthermore, in this case it can be argued that the implied consent by the patient includes repetition of the test, by the same or other technique. This practice also carries no risk of incidental findings, since the determination has been consented. Use of leftover samples under these conditions of minimal or absent risk can be considered to not require informed consent from the sample donor. Of note, special attention must be taken when the investigation will produce genomic data. Re-identification of subjects has been shown to be possible in genetic studies [11], [12], [13]. Indeed, risk assessment guides usually indicate the obtention of specific informed consent when the expected results of the secondary use involve this kind of data.

On the other hand, clinical laboratories more likely retain personal records associated to samples. These samples can still be used for secondary research without IC, provided they are rendered non-identifiable before handing them to the investigator. Making a sample non-identifiable means dissociating the identifiers (name, address, contact information, personal ID number, and so on) from the biological specimen and associated databases, so that it would be not possible to ascertain the identity of any study participant. This process does not necessarily involve irreversible removal of information. Sometimes it is possible to maintain a link back to the personal information (coded samples). In such cases, risk must be minimized by setting up procedures making clear that the laboratory cannot share the key with the investigator who will use the leftover samples, and the investigator must not intend to re-identify them.

Still, not all information needs to be concealed from the investigator. At an absolute minimum, for any such leftover sample to be useful, age and sex from the source individual must be known. Other pieces of information might be added as well, such as ethnicity, nationality or any georeference that allows localization of the origin of the sample (although not the personal address). However, more information increases risks. First, the possibility of re-identification of samples should be considered. Re-identification of individual subjects has been achieved using data from public sources, although this has only been done by experts in the field as a proof-of-principle that it be possible[13], [14], [15]. Secondly, by retaining demographic data to samples, there is the risk of singling out a population or group of people. For instance, results could show that a group is at high risk of having a genetic variant associated with a particular disease [16]. This situation underscores the advice to be thoughtful when working with vulnerable populations.

Despite these risks, ethics committees are likely to grant IC exemptions when using leftover samples dissociated from personal information. In addition, clinical laboratories are used to deal with confidential information, and rely on procedures to warrant medical secrecy.

Nonetheless, IC exemption would be difficult to claim when the project is managed by an investigator affiliated to the clinical laboratory, who likely has access to the laboratory database and the personal identification of the people included in the study. This situation does not comply with the requirement that the investigator cannot ascertain the identity of the subjects. In such a case, IC should be asked from the patient at some point. This would be preferably before the sample is obtained, but not necessarily, since it should be clear that the health care service (i.e. the clinically relevant test) will not be altered regardless of the patient’s consent to participation in research.

3.2. A debate regarding the relative weight of participant’s autonomy

Current views on the requirement of informed consent for the secondary use of leftover samples follow the arguments discussed in the previous sections. Accordingly, rules and recommendations at the national or international level emphasize the importance of obtaining informed consent from research participants for any kind of study design, but usually incorporate exemptions based on the practicality of obtaining the IC, always on the condition of minimal harm risk [17], [18].

However, ongoing discussions have put forward the idea that minimal risk is an insufficient argument to allow the use of samples without consent [19], [20]. Besides the risk of breach of privacy, other ways of harming the individual should be considered in the case they find out their samples had been used without consent. For instance, some people may object the use of their bodily fluid on religious or other personal grounds. Some people may not have objections with the use of the sample per se, but instead may not want to contribute with research towards a certain aim. What is more, disclosure of unconsented use of specimens has led in the past to public scandals, such as the resonant cases of the development of HeLa cell line, the studies on Havasupai tribe samples and the organ collection at Alder Hey Hospital [16], [21], [22]. Admittedly, in these cases there were major deviations from ethical procedures, but they nonetheless reflect delicate public feelings regarding the use of remnant specimens. Investigators and laboratory managers should be warned to mind not only regulatory compliance but also expectations from individuals providing samples.

Regarding people’s expectations, extensive empirical research shows that people prefer to decide whether to contribute the remnant of their samples to research [23], [24], [25]. Interestingly, survey participants had opinions on which topics they would rather refuse to contribute, such human cloning, studies involving indigenous populations and research with potential economic profit [26].

All in all, it appears that protocols for the use of leftover samples will incorporate these arguments in the future. Currently, secondary research and method development at clinical laboratories are mostly functioning under protocols based on IC exemption, but this practice should be revised in proposals to update regulations. Ideally, proposed approaches to incorporate informed consent for the use of leftover samples should be planned to avoid imposing restrictive burdens and costs, and also maintain some flexibility on research objectives.

3.3. The process of informed consent for secondary use of leftover samples

The process of informed consent is the main tool to materialize the respect for people’s autonomy. In the case of leftover samples, the person should be asked whether, after the clinically relevant determinations are completed, the remnant material could be used for additional purposes, instead of being discarded. Since the secondary uses are not known at the time of obtaining the sample, the informed consent does not include project-specific information. Instead, the secondary use of leftover samples can be covered by informing the patient only about the possibility of a future use. Although still debated, broad consents should not be considered inferior to project-specific consents. In fact, severe criticism has been raised against the traditional project-specific written informed consent process. Evidence has accumulated indicating that traditional consent documents, which struggle to incorporate a series of required statements, are not understood, or even not read at all [17], [27]. In this sense, studies have found that readability of consent documents to be unacceptably low [28], [29]. For instance, recent reports showed that consent for COVID-19 vaccine trials were long and difficult [28], [30]. In addition, this situation will likely worsen, since legal requirements demand more disclaimer phrases to be incorporated into consents [28].

The experience of biobanks and other sample collections has promoted the development of alternative models in the search for IC [31], [32]. In biobanks, volunteer donors provide a broad consent open for a scope of secondary uses, they are asked what to do with potential incidental findings and they are informed that they can ask for withdrawal of samples and data at any point in the future. While it has been argued that broad consent fails to provide information to the participant, it actually represents a consent to sample governance, a consent to a set of rules on how the samples will be managed [33]. Broad consents ask the individual to decide if they would like to collaborate with research, provided that certain rules are met. Furthermore, evidence showed that broad consents have been perceived positively by study participants [34], [35]. To be sure, it should be noted that current or future projects making use of those samples would still have to be approved by an independent ethics committee.

Clinical laboratories could use the experience of biobanks and find that asking broad consent at the time of sample collection could be a reasonable way to comply with ethical requirements [36], [37]. Whereas consent for clinical care tests is often implied (i.e., it is not standard procedure to request signed consent before diagnostic sample extraction), consent for secondary use of samples should be written and explicit. Admittedly, this implies a modification in the procedures that are familiar to anyone visiting a diagnostic laboratory. While it is uncertain how the public would react, surveys and studies indicate that people are usually prone to collaborate [26], [35], [37]. It is also debatable whether applying a broad consent system would require large changes in infrastructure and procedures, since clinical laboratories already have the capacity to store samples, safeguard specimens and data, and implement new protocols. Eventually, it could be expected that laboratories and the public become used to a new norm.

4. Final words

Our most important mandate at the clinical lab is to provide a result of assured quality. But clinical laboratories also play a larger role in the medical and scientific community, a role in the production of knowledge. To this end, one of the assets of clinical laboratories is the expert handling of human samples, a practice that entails not only technical proficiency but also ethical responsibility.

Regarding the latter, it is clear that interventional research requires the most stringent and specific consent from participants. On the other hand, observational studies which carry minimal risks might make use of a different set of procedures that still respect and protect the rights of the patients. While the use of leftover samples without consent can be ethically acceptable on occasions, there is an ongoing discussion regarding the relative weight of the participant’s right to know and consent. It is possible that participant’s involvement will gain more consideration, even if it means that the ability to carry out some research projects might be affected. In this scenario, participant’s involvement in the use of leftover samples and their associated data could take the form of broad consents obtained during clinical care. We believe it would be convenient that clinical laboratory management be aware of, and ideally transmit to their institutions, the existence of novel alternatives of ethical compliance which would enhance their capacity to fulfill their role in the production of new knowledge.

Recommendations for investigators and management of medical laboratories (box text)

• ●If the laboratory participates in research (including method validation and development), formalize the activity by having written protocols and training the personnel accordingly. Request a copy of the protocols that were approved by an ethics committee.
• ●Clearly define the scope of any project initiated by the lab or accorded with a third party. Anticipate if it will be possible to ask for an exemption of informed consent.
• ●Define the people involved and who has access to personal data of the participants. Establish a base protocol for the preparation and coding of samples to be given to a third party.
• ●Be aware of potential ethical issues to anticipate problems, especially when participating in projects involving genetic research, reproductive tissue samples, or vulnerable populations.
• ●It is a sound policy to always look for ethical review by an independent committee, even if it may seem unnecessary.
• ●Consider novel alternatives such as broad consent to comply with autonomy rights of research subjects, and establish the proper protocols to that end.
• ●If the scale of the sample gathering is too big, if it is sustained for too long, or across multiple projects, laboratory should consider the association with a biobank, or even the creation of one.

The authors did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors for the preparation of this manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

No data was used for the research described in the article.