World Sleep Industry ENCYCLOPEDIA

Sample Testing And Storage

6.5.2

Definition, scope, and adjacent boundaries

Sample testing and storage is an integrated service in which human-derived materials are received, kept under controlled conditions, analyzed, reviewed, reported, and eventually transferred or disposed of according to an intended use. Materials may include blood and its fractions, saliva, urine, stool, hair, swabs, cells, tissues, and extracts. Testing may address hormones, proteins, metabolites, drug concentrations, nucleic acids, microbiomes, or cellular functions. Within WSIC 6.5.2, “Subject and Sample Management,” this entry focuses on the analytical chain from a testable specimen to a defensible result and on how custody before and after testing affects that result.

The service is distinct from the broader life-cycle governance described under biosample management. It is also not synonymous with clinical diagnosis, freezer rental, or statistical analysis of research data. Sample governance determines whether material may be collected, shared, and reused; a testing laboratory demonstrates that a particular method is suitable for a defined matrix and purpose; a storage operator maintains identity, environment, and inventory; and medical interpretation requires applicable clinical evidence and professional authority. One organization may perform several roles, but each responsibility, record, and authorization still needs to be explicit. The label “central laboratory” is not evidence of competence by itself.

The testing plan, laboratory manual, and contract should jointly define the analyte, matrix, method or platform, research or clinical use, expected volume, collection and receipt windows, storage conditions, batching, turnaround time, repeat rules, report fields, raw-data deliverables, residual quantity, retention period, and final disposition. Local law, ethics approval, consent, the laboratory's actual accreditation scope, and transport rules may all apply. Conformity with a standard or possession of an accreditation means only what the documented scope covers; it does not establish that every assay, matrix, site, or use has been assessed.

Intended use, testing plan, and contractual interfaces

Method selection starts with intended use. A result may explore a mechanism, screen a candidate biomarker, characterize pharmacokinetics or pharmacodynamics, form a study endpoint, support eligibility, or be returned to a participant and influence care. The evidentiary threshold for the same analyte differs among exploratory research, regulated trials, and clinical practice. A research-use result does not become a clinical result merely because it appears on a formal report. If individual results may be returned, analytical and clinical validity, applicable regulation, confirmation, communication responsibility, and the handling of urgent or incidental findings should be settled before testing begins.

The testing plan converts the research question into executable specifications: material and additive, planned and actual sampling time, minimum usable quantity, aliquoting, allowable freeze-thaw exposure, stability window, measuring range, units, batch design, masking, calibrators and controls, acceptance and rejection criteria, repeat and dilution rules, and metadata needed for analysis. Sleep and circadian studies particularly need the relation to sleep onset, awakening, the light-dark cycle, time zone, and intervention. Recording only “morning specimen” or the scheduled visit can conceal timing differences that materially affect phase-sensitive measures such as melatonin and cortisol.

The sponsor, collection sites, carrier, storage facility, testing laboratory, and data team need a responsibility matrix and defined handoffs. The agreement should address limits on subcontracting, excess or inadequate volume, priorities when material is depleted, authorization for retesting, deviation notification, amended results, data and intellectual-property rights, audit access, security, recovery, pricing changes, and exit transfer. Work may be outsourced, but final oversight of research quality, participant rights, and fitness of results cannot be outsourced. A laboratory should not expand the panel or retain residual material for development without authorization.

Receipt, pre-analytics, and storage control

At receipt, the laboratory should reconcile specimen identity, labels and manifest, container, material type, quantity or volume, collection and arrival times, transport condition, seal integrity, leakage, and temperature records. Predetermined criteria should lead to acceptance, quarantine, conditional acceptance, or rejection. Identity conflicts, hemolysis or contamination, wrong containers, delay, temperature excursion, and missing critical metadata require a deviation record and an assessment against the intended analysis. Not collected, lost in transit, insufficient, rejected on receipt, and analytically not detected are different states and must not collapse into one blank value.

Written procedures should control clotting, standing, centrifugation, mixing, stabilization, aliquoting, interim holding, extraction, and transfer to long-term storage. Both specified conditions and actual events should be recorded, with critical times derived from controlled clocks or systems. Home collection requires evaluation of the instructions, container, error-resistant design, return transport, and extreme-weather scenarios, not just validation of the in-laboratory assay. Samples from comparison groups should be randomized or balanced across plates, runs, and positions when appropriate so that case status or visit does not become indistinguishable from a technical batch.

Storage temperature, container, light protection, vapor or liquid environment, and retention period should be based on the target, matrix, method, and stability evidence; no single temperature is suitable for all specimens. Every parent tube, aliquot, and derivative needs a unique identity, lineage, location, residual quantity, and freeze-thaw and exposure history. Continuous monitoring, alarms, backup power, reserve equipment, emergency transfer, and inventory reconciliation should be challenge-tested. A compliant temperature trace demonstrates an environmental condition, not by itself the continued fitness of every molecule, cell, or microbial feature.

Method qualification, validation, and run quality control

An analytical method should be shown to be fit for its intended use. The extent of qualification or validation depends on result risk, development stage, method provenance, and regulatory pathway. Common characteristics include selectivity or specificity, the calibration model, measuring range, detection and quantitation capability, accuracy or bias, precision, recovery, matrix effect, dilution integrity, carryover, stability, and robustness. Acceptance criteria should be defined from science and risk before results are inspected and tailored to the target, matrix, and use. One numerical template cannot govern every immunoassay, mass-spectrometric, molecular, or cellular method.

A commercial kit, published procedure, or method validated elsewhere still requires local evidence that personnel, equipment, reagents, software, matrix, and measuring range perform as needed. A platform, reagent lot, calibration system, extraction procedure, laboratory, or algorithm change calls for impact assessment and, as warranted, bridging, parallel testing, or revalidation. High correlation between two methods does not make their results interchangeable. Systematic bias, performance at low concentrations, and matrix-dependent effects can change longitudinal trends or threshold classification, so multicenter and multiplatform work must preserve method version and plan comparability.

Each analytical run should contain suitable calibration, blanks, and independent control materials, with rules for accepting a run, handling a local failure, repeating, diluting, and reporting. Controls should cover meaningful parts of the measuring interval, and trends should be reviewed rather than treating each point as an isolated pass or fail. Proficiency testing or another external quality assessment should be used where applicable. Retesting must retain the initial result, trigger, authorization, and rule for the reported value. Repetition after viewing group identity or the desired direction, followed by selection of a favorable value, is not acceptable. Masking, sample sequence, and plate position should remain traceable.

Data processing, reporting, and limits of interpretation

A defensible deliverable is more than a final spreadsheet. The laboratory should retain enough information to reconstruct a result: raw signals or images, sample-to-well maps, calibration curves, control results, dilution factors, instrument and reagent lots, processing settings, equations, software and algorithm versions, manual review, and audit trail. Transfers require a data dictionary, units, time-zone conventions, codes, integrity checks, and version control. If a released result changes, the laboratory should issue an identifiable amended report and preserve the earlier report rather than silently overwriting it.

Reports should distinguish numeric measurements, qualitative classifications, and reasons why no valid result is available, and should use only the precision supported by the method. Below the quantitation limit, below detection capability, not detected, above the measuring range, quantity not sufficient, unacceptable specimen, technical failure, and genuinely missing do not mean the same thing. The dataset should also carry material, sampling time, receipt status, method version, unit, the source of any reference or decision interval, and material deviations. In circadian analyses, planned time must not replace actual collection time or time relative to sleep during data cleaning.

Interpretation by the laboratory should remain within demonstrated method performance and specimen condition. A difference between research groups is not an individual diagnosis, and a reference interval may not transfer across age, physiological state, collection time, or platform. If a result will be returned or used in care, it should pass through an approved clinical pathway that defines identity confirmation, interpretation, documentation of communication, and follow-up. Research findings may generate hypotheses, but a single biomarker result should not be presented as the cause of a sleep disorder, a treatment choice, or a promise of risk prediction.

Quality, safety, information systems, and continuity

The quality system should cover personnel qualification and continuing competence, document control, equipment calibration and maintenance, reagents and supplies, environmental monitoring, internal controls, external assessment, deviations, nonconformities, corrective and preventive action, complaints, and internal audit. Management review can combine turnaround time, first-pass acceptance, invalid and repeat rates, control trends, external performance, temperature excursions, inventory accuracy, report amendments, and time to close incidents. Metrics should be stratified by specimen type, site, and method where useful so that an overall average does not hide a systematic problem in one run or population.

Biosafety measures should follow risk assessment for the material and procedure, including receipt, opening, centrifugation, aerosols, sharps, culture, spills, waste, and exposure response. Sensitive pathogens or genetic materials may also warrant enhanced biosecurity and access restrictions. Identity, health, genomic, and time-behavior data require tiered protection, least privilege, logs for access and export, and review of subcontracting, cloud services, remote maintenance, and cross-border transfer. De-identification does not eliminate every re-identification risk, and the specimen itself may convey information about an individual and biological relatives.

Laboratory information systems, instrument interfaces, and analytical software should be validated in proportion to use and risk. Accounts should be individual, permissions appropriate, clocks synchronized, audit trails usable, and backups demonstrably recoverable. Continuity plans need to address equipment failure, power loss, shortage of cooling supplies, network outage, cyberattack, fire, flood, severe weather, and loss of key staff, with priorities for specimen transfer, alternative testing, notification, and restoration. Exercises should test real capacity, routes, and data readability. Possessing a backup file or spare freezer does not establish that the service can be recovered.

Procurement acceptance, change control, and exit delivery

Procurement should evaluate scenarios and specimens resembling the project rather than relying on price, advertised sensitivity, or a general certificate. Due diligence may examine the actual scope of accreditation and licensing, validation summaries, key personnel, instruments and reserve capacity, sample tracking, quality events, external assessment, data formats, cybersecurity, subcontracting, and evidence of disaster recovery. A pilot should include routine specimens and difficult cases such as inadequate volume, label conflict, out-of-range results, temperature excursion, retesting, and amended reporting so that interfaces and escalation paths are shown to work.

Changes during the project require control. The laboratory should give advance notice of changes to site, platform, reagents, calibrators, algorithm, software, reference interval, subcontractor, storage equipment, or procedure that could affect results. The parties can then use risk to decide on document review, bridging, parallel analysis, added validation, or a new baseline. Comparability between historical and new batches needs evidence; “newer technology” is not a comparability assessment. Major deviations and safety events also require timely escalation and an assessment of released results, stored material, and subsequent analyses.

At service end, reconciliation and verifiable delivery should cover the inventory of specimens and derivatives, location and quantity, freeze-thaw and deviation history, destruction or transfer certificates, complete results, raw data, control and calibration records, method and software versions, data dictionaries, audit trails, open issues, and retention obligations. The recipient should verify count, identity, documentation, and readability. After authorization, the former provider should delete copies as required and retain evidence that must lawfully remain. An exit clause saying only “return the samples,” without format, cost, media, schedule, data migration, and transit risk, cannot preserve the ability to review the study after a provider changes or closes.