Key takeaway
Better decisions require reliable laboratory data. Sponsors should define a fit-for-purpose laboratory strategy early, considering analytical methods, sample requirements, logistics and data flows before the study becomes operationally complex. Laboratory planning should therefore be part of the evidence-generation strategy from the beginning, not simply a testing decision made later in study development.
Clinical trials generate an enormous amount of data
Patient demographics, medical history, clinical assessments, imaging, questionnaires, electronic patient-reported outcomes, treatment exposure, safety events and, often less visibly, laboratory results.
Laboratory data may not always receive the same attention as the clinical endpoints that ultimately make the headlines. Yet laboratory testing is embedded throughout the clinical development process, from screening and eligibility through safety monitoring, pharmacokinetics and pharmacodynamics to biomarker analysis and the assessment of treatment response.
This raises an important question: How important are laboratory data to the reliability of the evidence generated by a clinical trial?
The proportion of clinical trial data generated through laboratory testing varies considerably depending on the therapeutic area, study phase, protocol and endpoint strategy. However, the Clinical Data Interchange Standards Consortium (CDISC) describes laboratory data as one of the largest components of clinical trial data and has developed a dedicated Laboratory Data Model to support their acquisition and exchange between laboratories, sponsors and CROs.
One of the largest components of clinical trial data is laboratory data. Developing a standard lab data format is critical to achieving CDISC’s mission of creating standard data models that support the end-to-end data flow of clinical trials, from the data sources into an operational database and through to analysis and submission. Use Cases for Clear Data | CDISC
More importantly, the value of laboratory data is not determined simply by volume. It is determined by their relevance, reliability and impact on the decisions made throughout the trial.
From laboratory result to clinical evidence
A laboratory result is not simply a number entered into a database.
It is the final output of a chain of interconnected processes, including patient preparation, sample collection, tube selection, processing, storage and transportation, followed by analytical testing, calibration, quality control, result review, data transfer and reporting. A weakness at any stage of this process can affect the reliability of the final result.
This becomes particularly important in multicentre and multinational clinical trials, where laboratory testing may involve different sites, countries, instruments, analytical methods and local laboratory practices.
In the study Harmonization of laboratory results by data adjustment in multicenter clinical trials, researchers compared results from 10 clinical laboratories using different analytical methods. Even with appropriate traceability, results across laboratories were not necessarily directly comparable, highlighting the importance of standardisation and harmonisation when laboratory data from multiple locations are combined and interpreted within the same study.
This is one reason why central laboratories can play an important role in clinical research. Their value extends beyond performing tests to supporting a more consistent and controlled laboratory data-generation process across sites and countries.
Reliable laboratory data starts before the sample reaches the laboratory
When we talk about laboratory quality, it is easy to focus primarily on the analytical method. Reliable laboratory data, however, starts much earlier. During study planning, a number of practical questions need to be addressed:
- Is the selected sample type appropriate for the intended analysis?
- Is the collection tube compatible with the assay?
- How quickly does the sample need to be processed?
- What storage conditions are required?
- How stable is the analyte under the intended shipping and storage conditions?
- Is the analytical method validated for the intended matrix and purpose?
- Are sample volumes sufficient?
- Are local and central testing methods comparable where both are used?
- How will laboratory results be transferred into the clinical database?
- What turnaround time is required?
- How will deviations or unexpected results be managed?
These may sound like operational questions. They are also data-quality questions.
A decision about sample handling made during study design can affect the quality of a result generated months later. Choosing local rather than central testing may create a need for additional data harmonisation.
Introducing a new biomarker later in the programme may require additional assay development or validation of analytical process. A sample matrix that was not carefully considered at the beginning may later prove unsuitable for the intended analytical method.
None of these situations necessarily prevents a clinical trial from proceeding. They can, however, create additional work, delays, amendments, validation activities or uncertainty that could have been reduced through earlier planning.
Laboratory planning as part of study design
Early involvement of laboratory experts can help identify potential constraints before key study decisions are finalised. Laboratory requirements are sometimes discussed only after key elements of a study have already been established. Yet laboratory decisions are closely connected with the protocol, sample collection procedures, site capabilities, logistics, timelines, data management and statistical analysis.
Bringing laboratory experts into the discussion early does not mean that they need to influence every scientific decision. It means asking the right questions early enough to identify potential constraints, confirm feasibility and design a fit-for-purpose laboratory data-generation process.
This approach is also consistent with ICH E6(R3), which places greater emphasis on proactively designing quality into clinical trials, identifying factors critical to trial quality and applying proportionate, risk-based approaches. Its Principles and Annex 1 have been effective in the EU since July 2025, while Annex 2, addressing additional considerations for non-traditional trial designs, including decentralised clinical trials, will become effective in January 2027.
The underlying principle is highly relevant to laboratory planning: quality should be designed into the trial rather than inspected into it later. The same thinking should apply to the laboratory data on which study decisions may depend.
Increasing complexity of biomarker-driven testing
The role of laboratory data is evolving as clinical development becomes increasingly biomarker-driven. Traditional safety testing remains fundamental, but clinical trials now incorporate a growing range of more specialised analyses, including:
- molecular and pharmacodynamic biomarkers
- genomic testing
- immunological and cellular assays
- circulating biomarkers
- pharmacokinetic measurements
- companion diagnostic-related testing
- increasingly complex biomarker panels
These analyses can differ significantly in their sample, analytical, validation and operational requirements. The existence of an assay alone does not guarantee that it will generate reliable data for a specific clinical trial.
This is reflected in the FDA’s April 2026 final guidance, Bioanalytical Method Validation for Biomarkers, which provides recommendations for validating bioanalytical methods used to evaluate biomarker concentrations in support of Investigational New Drug Applications (INDs), New Drug Applications (NDAs) and Biologics License Applications (BLAs).
The guidance reinforces an important point: the reliability of a biomarker result depends not only on the assay itself, but also on appropriate method validation and its application to study samples. As the biological questions addressed in clinical development become more sophisticated, the laboratory strategy and infrastructure supporting them need to evolve accordingly.
Data alone is not oversight
One of my key takeaways from the discussions on outsourcing management and ICH E6(R3) at the PCMG Annual Assembly 2026 was that having KPIs and metrics is not the same as having effective oversight.
The same principle applies to laboratory data. A clinical trial may generate thousands or even millions of laboratory results, but data volume alone does not make those results reliable or useful.
Effective oversight requires an understanding of how the results were generated, whether the underlying processes were controlled, how deviations were managed, whether relevant trends can be identified and whether the data remain fit for their intended purpose.
Data alone is not oversight, and data volume alone is not data quality. The objective is to generate laboratory data that are reliable, traceable, comparable and fit for purpose.
From laboratory data to regulatory evidence
The importance of laboratory data becomes particularly clear when we consider how clinical trial evidence is structured, analysed and ultimately evaluated by regulators.
The European Medicines Agency (EMA) explicitly identifies clinical laboratory results as individual patient data generated in clinical studies, alongside imaging data and patient medical charts. EMA has also explored whether access to structured clinical study data can support and improve the regulatory evaluation of medicines.
Laboratory results are therefore not simply operational outputs delivered to investigators during a study. They form part of the underlying clinical evidence. Once a laboratory generates a result, the data journey continues: laboratory data need to be transferred, structured, reconciled and integrated into the broader clinical data environment.
The Clinical Data Interchange Standards Consortium (CDISC) Laboratory Data Model (LAB) was specifically developed to support the acquisition and exchange of laboratory data, primarily between laboratories and sponsors or CROs. This reinforces an important point: laboratory quality and data quality cannot be considered separately.
A reliable laboratory result needs more than an appropriate analytical process. It also needs a controlled and traceable data flow that preserves its integrity and meaning from the laboratory through to the clinical dataset.
A fit-for-purpose laboratory model
The laboratory model should be fit for purpose. This does not mean that every clinical trial requires a central laboratory. Depending on the scientific and operational requirements of the study, local laboratories, central laboratories, point-of-care testing or a combination of approaches may be appropriate.
Decentralised and hybrid trial models illustrate this particularly well. Some testing may be performed closer to participants, while other analyses may remain centralised because of their complexity, specialised analytical requirements or the need for greater consistency across sites.
ICH E6(R3) Annex 2 addresses additional considerations for non-traditional trial designs, including decentralised clinical trials, and emphasises that trial approaches and data should remain fit for their intended purpose.
The objective should therefore not be centralisation for its own sake. The key question is whether the chosen laboratory model can reliably generate the evidence the study needs. Laboratory strategy should be designed around that evidence from the beginning.
When laboratory decisions come too late
Laboratory decisions become increasingly difficult to change as a clinical trial progresses. During early study planning, sponsors may still have several technically feasible options. Once the protocol has been finalised, sites activated, collection materials distributed, sample logistics established and databases configured, even a seemingly small change to the laboratory process can have consequences across multiple areas of the study.
Depending on the change, this may require:
- additional assay validation
- revised laboratory manuals
- changes to collection materials
- site retraining
- revised logistics
- additional stability work
- changes to data-transfer specifications
- database updates
- additional reconciliation
- protocol or operational amendments
These situations do not necessarily indicate that something has gone wrong. They are often a consequence of decisions being made at different stages of a complex clinical development programme. This is why early laboratory feasibility discussions are valuable even when no obvious problem exists.
The purpose is not to anticipate every possible challenge. It is to identify the constraints that matter while there is still enough time and flexibility to evaluate options and address them before they become operationally complex.
The laboratory as part of the evidence infrastructure
As clinical development becomes increasingly biomarker-driven, decentralised and data-intensive, the role of the laboratory is evolving as well. The laboratory is not simply the place where a sample is tested. It is part of a broader evidence-generation pathway:
Patient → Sample → Pre-analytical process → Analysis → Quality control → Result → Data transfer → Clinical database → Analysis → Regulatory evidence
Every stage matters. A high-quality analytical method cannot compensate for inappropriate sample handling. A perfectly collected sample cannot compensate for an unsuitable analytical method. A reliable laboratory result can still lose value if the associated metadata are incomplete or the data transfer is not controlled. Similarly, large quantities of data cannot compensate for uncertainty about how those data were generated. The quality of the final evidence depends on the quality of the entire chain.
The most useful time to have this conversation is before the study becomes operationally complex. At this stage, sponsors have greater flexibility to identify constraints, compare options and design the laboratory component around the needs of the study.
For sponsors, this changes the question that should be asked during study planning. Rather than simply asking, “Which laboratory will perform our testing?”, a more useful question is: “What laboratory strategy will provide the reliable data needed to support the decisions in this study?”
This question naturally leads to a broader discussion about assay feasibility, sample requirements, validation, logistics, standardisation, data transfer, timelines and quality.
Final thought
Clinical research is becoming increasingly sophisticated. Biomarkers are playing a greater role in clinical development, trial models are becoming more distributed, analytical technologies are advancing and AI is enabling the processing of increasingly large and multidimensional datasets.
Yet one principle remains unchanged: better decisions require reliable data.
Reliable laboratory data do not happen by accident. They are designed, generated, controlled, reviewed and protected throughout the entire laboratory process. For this reason, the laboratory should not be considered simply as a testing function introduced later in clinical trial planning. It should be part of the evidence-generation strategy from the beginning.
Reliable laboratory data starts long before the result appears in the database.
Discuss your laboratory strategy with our experts
1. Why is laboratory data quality important in clinical trials?
Laboratory data support decisions throughout clinical development, including patient eligibility, safety monitoring, pharmacokinetics, pharmacodynamics, biomarker analysis and treatment-response assessment. Their value depends not only on the analytical method, but also on sample handling, data transfer, standardisation and the overall reliability of the laboratory process.
2. When should laboratory planning begin in a clinical trial?
Laboratory planning should begin early in study design, before the trial becomes operationally complex. Early involvement of laboratory experts helps sponsors assess feasibility, identify constraints and define a fit-for-purpose laboratory strategy before protocols, logistics, collection materials and databases are finalised.
3. What does a fit-for-purpose laboratory model mean in clinical trials?
A fit-for-purpose laboratory model is one that matches the scientific and operational requirements of the study. Depending on the protocol, this may involve central laboratories, local laboratories, point-of-care testing or a combination of approaches rather than centralisation by default.
4. How can laboratory data from multicentre clinical trials be made more reliable?
Reliability can be improved through appropriate sample handling, validated analytical methods, standardisation, harmonisation, quality control and controlled data transfer. This is particularly important when results are generated across different sites, laboratories, instruments or analytical methods.
5. How does laboratory data contribute to regulatory evidence?
Laboratory results form part of the underlying clinical evidence generated during a trial. To retain their value, the data need to be transferred, structured, reconciled and integrated into the broader clinical dataset in a controlled and traceable way before analysis and regulatory evaluation.
References:
- ICH E6(R3) Guideline for Good Clinical Practice – International Council for Harmonisation (ICH) / European Medicines Agency (EMA). Accessed: 13 August 2026.
- ICH E6(R3) Annex 2: Additional Considerations for Non-Traditional Interventional Clinical Trials – International Council for Harmonisation (ICH). Accessed: 13 August 2026.
- Laboratory Data Model (LAB) / Data Exchange Standards – Clinical Data Interchange Standards Consortium (CDISC). Accessed: 13 August 2026
- Use Cases for Clear Data – LAB – Clinical Data Interchange Standards Consortium (CDISC). Accessed: 13 August 2026.
- Use of clinical study data in medicine evaluation – European Medicines Agency (EMA). Accessed: 13 August 2026.
- Bioanalytical Method Validation for Biomarkers: Guidance for Industry – U.S. Food and Drug Administration (FDA). April 2026. Accessed: 13 August 2026.
- Harmonization of laboratory results by data adjustment in multicenter clinical trials – Lee SG, Chung HJ, Park JB, Park H, Lee EH. The Korean Journal of Internal Medicine. 2018;33(6):1119–1128. Accessed: 13 August 2026.
