# How to Handle a Request for Secondary Use of Stored Biospecimens Primary use means analyzing a biospecimen for the study it was collected for. Secondary use means any later analysis of that same stored material for a different question, often posed by a different team years afterward. The rule of thumb for handling a request: the wider the gap between what donors originally agreed to and what the new request wants, the heavier the process the request deserves, ranging from a documented review under existing consent all the way to re-contacting participants or declining outright. A biospecimen is any biological material kept for research or clinical reference, such as blood, tissue, saliva, urine, hair, or archived dental casts. Requests to reuse these materials land on biobank managers, principal investigators, core facility directors, and anyone who inherits a freezer. Students and early-career researchers meet the problem sooner than they expect, because the person who says yes or no is often the person holding the keys to the freezer, not a dedicated ethics officer. This article lays out a secondary use of stored biospecimens decision framework built around six concrete options, a side-by-side comparison, worked scenarios, and the mistakes that most often cause trouble. The framework draws on the responsible conduct of research tradition that treats donor trust as infrastructure, not paperwork. ## Where Secondary Use Requests Come From Requests rarely announce themselves as ethical questions. They arrive as scientific opportunities, and that is what makes them easy to mishandle. Analytical chemists recently developed a method to detect titanium dioxide nanoparticles in human hair, a specimen type not previously explored for emerging pollutant monitoring, using enzymatic hydrolysis with pancreatin to release the nanoparticles from the keratin matrix before single particle ICP-MS analysis [1]. Someone with archived hair samples and an interest in environmental exposure could reasonably read that paper and ask to run the method on stored material. Whether that is appropriate depends entirely on why the hair was collected and what donors were told. Forensic researchers reanalyzed maxillary dental casts from a longitudinal twin cohort, material originally collected to track dentition development, to test whether three-dimensional palatal surface scans could match antemortem and postmortem records for human identification [2]. The casts were made for pediatric dental research. Identification science is a different enterprise with different privacy stakes. Clinical genetics laboratories now compare genome and [exome sequencing](/blog/guides/exome-sequencing) workflows across sample types such as saliva, buccal swabs, and dried blood spots, and their reports can carry laboratory-reported secondary or additional findings unrelated to the clinical question that prompted testing [3]. Each of these situations sits at a different point on the consent spectrum, which is exactly why an improvised answer fails. ## The Five Questions That Drive the Decision Before choosing among options, answer five questions in writing. The written answers become your record if the decision is ever questioned, which is a core habit of responsible conduct of research. **1. What did the original consent actually say?** Find the consent form and the protocol, not just your memory of them. A narrow disease-specific consent covers far less than a broad consent, which the 2018 revision of the federal Common Rule (45 CFR 46) introduced precisely to allow storage and future research use. Some materials were never consented for research at all, such as residual clinical tissue collected during routine care. If no consent exists, treat the request as if consent were absent, because it is. **2. What is the identifiability status of the material?** De-identified specimens carry no codes and cannot be traced back to a person by the research team. Coded (also called pseudonymized) specimens carry a key that links them back to identities, and that key exists somewhere even if you do not hold it. Identified specimens are directly labeled. Under the Common Rule, research on materials that are not individually identifiable may not qualify as human subjects research at all, which changes everything about the required process. Do not make this determination alone. The IRB makes it or confirms it. **3. Who controls the material, and under what agreement?** Custodianship is not the same as ownership. A biobank may hold specimens under a material transfer agreement (MTA) that restricts redistribution, commercial use, or onward sharing. A clinician who collected samples may not have authority to release them. The department, the institution, or a funding agency may hold claims. Check the MTA before promising anything to a requester. **4. How sensitive is the new use?** Genomic sequence data is effectively identifiable no matter how it is labeled, because genome sequences can be matched to other data. Specimens tied to addiction history, reproductive care, psychiatric diagnoses, or stigmatized conditions carry elevated risk even when coded. A new assay for titanium dioxide exposure in hair [1] is low sensitivity. Re-sequencing the same donors' genomes is not. **5. What does the requester plan to do downstream?** Will the material be shared further, sequenced, deposited in a public database, or used toward a commercial product? Does the requester's institution have its own IRB? A request that ends in your freezer is managed differently from one that ripples outward. ## Secondary Use of Stored Biospecimens: The Decision Framework Options Six options cover nearly every situation. Each is explained on its own terms with pros and cons, then the comparison table pulls them together. ### Option 1: Proceed Under the Original Consent If the consent form and protocol clearly cover the new use, and the custodian agrees, the request can proceed on the existing approval with a documented note in the study file explaining the fit. **Pros.** This is the fastest path. Participants bear no new burden, no new approval cycle delays the science, and the decision is defensible when the consent language is genuinely clear. **Cons.** Consent language is often ambiguous in exactly the places a new use lands. Phrases like "related research" invite scope creep. One visible overreach can damage a biobank's reputation with a donor community far beyond the single study involved, and donor trust is the resource that keeps the freezer full. ### Option 2: Request an IRB Determination or Exemption Ask your Institutional Review Board (IRB), the committee that reviews human subjects research, whether the planned activity counts as human subjects research requiring approval. If the material is truly de-identified and no identifiable data will be generated, the IRB may issue a written determination that the activity is not human subjects research. If donors gave broad consent, the 2018 Common Rule revision added an exempt category that can cover secondary research with identifiable data. The federal Office for Human Research Protections (OHRP) oversees this framework for institutions subject to the Common Rule. **Pros.** A determination is cheap, fast compared with full review, and produces a written record that protects everyone involved. For many records-based analyses of stored material, this is the correct and sufficient answer. **Cons.** A determination confirms a regulatory category. It does not repair a consent gap, and it offers no protection if the requester later reintroduces identifiers or shares material beyond the agreed scope. It also cannot cover a design that involves contacting participants. ### Option 3: Full IRB Review With a Waiver of Informed Consent When specimens are identifiable or coded, the new question is worthwhile, and contacting donors is impracticable, the requesting team can seek full IRB review with a waiver of the consent requirement. IRBs grant waivers when the research involves no more than minimal risk, the waiver will not adversely affect participants' rights and welfare, the research could not practicably be carried out without the waiver, and where private information is involved, there is an appropriate plan for protecting it. When protected health information is in play, a parallel waiver under the HIPAA Privacy Rule may be needed. **Pros.** This path unlocks genuinely valuable research on cohorts that can no longer be re-consented, including studies of deceased donors or participants whose contact information has gone stale. **Cons.** Waivers are not guaranteed, the justification paperwork takes real effort, and review adds weeks to months. Requesters who treat the waiver as a formality tend to write justifications that IRBs reject, restarting the clock. ### Option 4: Re-Consent or Re-Contact Participants The strongest ethical footing is asking donors again. Re-contact means reaching participants, describing the new use in plain language, and obtaining fresh consent before their specimens or data are used. **Pros.** Consent given for a known, specific use removes the ambiguity that all other options manage around. Participants who are asked often report more trust in research, and the process surfaces objections early, before they become controversies. For sensitive new uses, commercialization, or anything involving genetic sequencing, re-consent is the default ethical choice when participants can be reached. **Cons.** Re-contact is slow and expensive. Cohorts shrink through attrition and lost contact information, which can bias the resulting sample. Re-contact itself uses personal information, so it needs its own ethical justification and IRB review. Some re-contacts cause distress, for example when families affected by rare disease are asked to revisit painful histories. Donors who are deceased or unreachable make this option impossible. ### Option 5: De-Identify and Share Through an Intermediary A custodian can strip identifiers, separate the coding key from the specimens through an honest broker (a person or office authorized to hold the key and who is not on the research team), and distribute coded or de-identified material under a data use agreement specifying permitted uses. **Pros.** This approach maximizes the scientific value of stored collections while lowering privacy risk, and controlled-access sharing is increasingly expected by journals and funders. It scales well, since one governance decision can enable many projects. **Cons.** True de-identification of genomic data is not realistically achievable, so genome sequences should be treated as inherently identifiable regardless of labels. Separating the code key can also destroy linkages the science actually needs, such as clinical follow-up, and that loss is sometimes discovered after the material has been stripped. Coding keys are a liability that must be stored and audited. Custodians who overestimate how de-identified their material is create the most common privacy failure in this space. ### Option 6: Decline or Narrow the Request Sometimes the right answer is no, or a smaller yes. Decline when the original consent excludes the use, when sensitivity is high and re-consent is impossible, when the requester's downstream plans are vague, or when proposed commercial terms conflict with the agreements governing the material. **Pros.** Declining protects donors, honors the consent contract, and preserves the biobank's standing with its donor community and its institution. **Cons.** Forgone requests are forgone science, and requesters may perceive custodians as obstacles. A decline without a recorded rationale invites the same request from someone else next quarter. ## Side-by-Side Comparison of the Main Options | Option | What it requires | Best fit | IRB involvement | Main advantage | Main risk | |--|--|--|--|--|--| | Proceed under original consent | Consent scope clearly covers the use, custodian sign-off, written documentation | New assay or question plainly within the stated scope | Documentation only | Fastest path, no participant burden | Ambiguous consent language, scope creep | | IRB determination or exemption | De-identified material, or broad consent on file | Records-based analyses with no participant contact | Written determination or exempt review | Cheap, quick, creates a paper record | Does not cure a consent gap | | Full review with consent waiver | Identifiable or coded material, documented impracticability of consent | Aging or lost cohorts, deceased donors | Board review and formal waiver | Enables important otherwise-impossible research | Not guaranteed, adds weeks to months | | Re-consent participants | Contact information, staff time, re-consent materials | Sensitive, commercial, or genomic uses in reachable cohorts | Reviewed re-contact protocol | Strongest ethical footing, builds trust | Slow, attrition, re-contact distress | | De-identify and share | Honest broker or key separation, data use agreement | External collaborators needing material or data | Often determination-level | Broad utility, lower privacy risk, funder-friendly | Genomic data resists true de-identification | | Decline or narrow | Governance decision with recorded rationale | Requests outside consent or with unclear downstream plans | None | Protects donors and institutional trust | Forgone research, requester friction | ## The Criteria That Decide Real Choices ![Decision path asking whether material is de-identified, coded or identified, leading to a determination, waiver or re-consent.](/article-images/illustrations/secondary-use-of-stored-biospecimens-decision-framework-identifiability-path.webp) The pros and cons above matter only in relation to the specifics of a request. Six criteria settle most real decisions. **Identifiability.** The more identifiable the material, the heavier the option needed. De-identified material can often move on a determination. Coded material usually needs a waiver or re-consent. Identified material and genomic sequences need the full weight of review, agreements, and often re-consent. **Consent scope.** Read the consent form against the new protocol line by line. A consent covering "childhood development research" does not stretch to forensic identification work on the same dental casts [2], and a consent for clinical diagnosis does not stretch to nanoparticle exposure assays [1]. The gap between the two is the whole decision. **Sensitivity of the new data.** Genomics, reproductive health, mental health, and stigmatized conditions raise the stakes a full level. Where the new analysis would generate genomic sequence from the specimens, plan for secondary findings governance as well, meaning a policy on whether clinically significant incidental results would be returned or reported. **Downstream intentions.** Onward sharing, deposition in public repositories, and commercial interest all push the decision toward the heavier options and toward tighter contract terms, because you are approving not just one use but the trajectory the material will travel. **Participant burden.** Re-contact is the only option that costs participants time and attention. Waivers and determinations cost them nothing directly, which is their ethical appeal, but also the reason IRBs scrutinize them. **Timeline.** Determinations take days to weeks. Waivers take weeks to months. Re-consent campaigns take months to years and shrink the usable sample. Requesters with deadlines should know these realities before they choose what to ask for. ## Worked Scenarios ### Scenario 1: Residual Clinical Tissue for Assay Development A pathology collaborator asks to use leftover diagnostic tissue blocks to validate a new staining assay. No research consent exists because the tissue was collected during care. Your five answers: no research consent, identified material, custodianship sits with the hospital, low sensitivity assay, no downstream sharing planned. The decision framework points to two viable paths. Either de-identify the tissue and seek an IRB determination that the work is not human subjects research, or seek full review with a consent waiver if identifiers must remain linked for clinical correlation. Check the hospital's policies on remnant tissue first, since many institutions have standing language in their clinical consent forms addressing this exact situation. ### Scenario 2: Stored Dried Blood Spots for a New Exposure Question A longitudinal cohort has archived dried blood spots collected under a consent for childhood development research. An environmental health team wants to run a new contaminant assay, analogous in spirit to the hair nanoparticle method [1] but on a matrix with clinical follow-up attached. The consent gap is real but modest, the material is coded, sensitivity is moderate, and the team wants to deposit results in a controlled-access repository. Full IRB review with a consent waiver fits well here, paired with a data use agreement governing the deposition. If community consultation during the original study revealed sensitivities about secondary use, the calculus shifts toward re-consent for the subset of families still reachable, with the waiver covering the rest. ### Scenario 3: External Collaborator Wants Coded Samples Plus Genome Sequencing ![Flowchart for an external request for coded plasma and genome sequencing, branching on consent scope into a bundled agreement or re-consent.](/article-images/illustrations/secondary-use-of-stored-biospecimens-decision-framework-scenario-three-bundle.webp) A laboratory at another university requests coded plasma samples and wants to perform [whole genome sequencing](/blog/guides/whole-genome-sequencing), with publication and possible sharing of sequence data. This request stacks the hardest criteria together: coded material, inherently identifiable genomic output, an external recipient, and public data sharing. The framework points to a bundle rather than a single option. Negotiate an MTA that forbids redistribution and re-identification attempts, require the recipient's IRB approval and a data use agreement with controlled-access terms, decide in advance how secondary findings will be governed, and confirm with your own IRB that the transfer as structured is covered. If the original consent predates genomic sequencing or excludes it, the correct first step is re-consent or a decline, not an MTA. ## Common Mix-Ups - **Confusing secondary use with secondary findings.** Secondary use describes reusing stored material for a new question. Secondary findings in genomics describe clinically significant results, reported by a laboratory, that fall outside the reason testing was ordered [3]. The words overlap but the governance questions differ, and mixing them confuses IRB submissions. - **Treating coded samples as de-identified.** A code key existing somewhere means the material is not de-identified. Ask where the key is and who can access it before making any identifiability claim. - **Assuming your study's IRB approval permits sharing.** An approval covering your protocol does not authorize transferring material to another institution. Sharing requires its own review step and, in most cases, its own agreement. - **Ignoring the material transfer agreement.** MTAs commonly prohibit redistribution and commercial use. Promising a collaborator something the MTA forbids creates a contractual problem that no amount of IRB paperwork can fix. - **Reading broad consent as unlimited consent.** Broad consent covers future research described in its terms, within whatever limits the document and the protocol set. It is broad, not boundless. - **Forgetting the recipient's IRB.** The requester's institution owes its own review of work done on its premises. Confirming this is part of responsible custodianship, not bureaucratic excess. ## Common Mistakes and Limitations Specific, practical failure points recur in real secondary use decisions: - Making the human subjects research determination yourself instead of requesting it in writing from the IRB. An undocumented judgment is worth nothing if the decision is later questioned. - Proceeding under original consent without writing down why the consent covers the new use. The note takes ten minutes and becomes the entire defense if a donor or auditor asks. - Running re-contact campaigns without IRB-reviewed scripts, opt-out mechanisms, and a plan for participants who find the contact upsetting. - Declining requests without recording a rationale, which leaves the next requester, and the next custodian, to relitigate the same question from scratch. - De-identifying material before realizing the science needed the clinical linkages, an irreversible loss in many cases. - Approving a request verbally, then discovering the MTA or the funding terms prohibit it. One limitation applies throughout. This framework is general education for planning and training purposes, and individual cases need review by your institution's IRB, privacy office, and technology transfer staff, because institutional policies and the specific agreements on your materials control the final answer. ## Frequently Asked Questions ### What counts as secondary use of stored biospecimens? Any analysis of previously collected biological material for a question beyond the purpose of the original collection counts. Using archived blood to validate a new assay, reanalyzing stored dental casts for a forensic method, or sequencing clinically collected samples for research all qualify, even when the original and new purposes sound similar. ### Do I always need new IRB approval for secondary use? No, but you always need a documented decision. If the material is genuinely de-identified and no identifiable information will be created, the IRB may determine the activity is not human subjects research at all. When identifiers are involved, some pathway through the IRB, either exemption, waiver, or full review, is required. ### What is the difference between de-identified and coded samples? De-identified samples have no codes and cannot be traced back to donors by the research team. Coded samples carry a key that could reconnect them to identities, and that key exists somewhere even if you never hold it. Under the Common Rule, this distinction often determines which approval pathway applies. ### When should I re-contact participants instead of using a waiver? Re-contact when the new use is sensitive, commercial, or genomic, when the consent gap is large, and when participants are reachable. Choose a waiver when consent is impracticable, risk is minimal, and the science cannot proceed otherwise. When both are feasible for a high-stakes use, asking donors is the stronger ethical choice. ### What should a material transfer agreement cover for shared biospecimens? An MTA for biospecimens should define permitted uses, prohibit redistribution and re-identification attempts, address intellectual property and publication, require the recipient's IRB approval, and specify return or destruction of unused material. Read the MTA governing your own material before promising anything, since its restrictions bind you regardless of what the requester wants. ### Does broad consent cover every future secondary use? No. Broad consent covers storage and future research use as described in the consent document, and IRBs still review whether a specific new use fits. Uses outside that scope, particularly commercial or highly sensitive ones, typically need re-consent, a waiver, or a decline. ## Further Reading - [Application of a virtual simulation platform combined with the presentation-assimilation-discussion class in human morphology practical teaching: a quasi-experimental study.](https://pubmed.ncbi.nlm.nih.gov/43009920/) - [Composite indices for assessing cumulative environmental exposure: A methodological comparison at the municipal level in France.](https://pubmed.ncbi.nlm.nih.gov/43009535/) - [Flood-related WASH disruption, point-of-use drinking water contamination and diarrhoea among under-five children in flood-affected communities in Ebonyi State, Nigeria.](https://pubmed.ncbi.nlm.nih.gov/43009892/) ## Sources 1. [Application of enzymatic hydrolysis for the isolation of titanium dioxide nanoparticles from human hair for sp-ICP-MS analysis.](https://pubmed.ncbi.nlm.nih.gov/43003912/) 2. [Evaluation of 3D palatal surface registration for human identification: The impact of developmental dentition stages.](https://pubmed.ncbi.nlm.nih.gov/43007435/) 3. [Real-world comparison of genome- and exome-based diagnostic workflows in a pediatric Albanian rare disease cohort.](https://pubmed.ncbi.nlm.nih.gov/43004392/)