Approach to maintaining comparability of biochemical data during long-term clinical trials

Area of Science:

• Biochemistry
• Clinical Trials
• Data Management

Background:

• Long-term clinical studies require consistent biochemical data.
• Changes in laboratory methods, specimen handling, and quality control can affect data comparability.
• Secular trends can also introduce variability over time.

Purpose of the Study:

• To design a structured approach for maintaining biochemical data comparability in long clinical trials.
• To address challenges posed by evolving laboratory practices and analytical methods.
• To ensure the reliability of data collected over extended study periods.

Main Methods:

• Implementing robust quality control and external quality assurance programs.
• Conducting direct comparisons of laboratory methods during assay updates.
• Utilizing statistical techniques to detect assay bias and realigning data to current methods.
• Establishing laboratory comparisons of new versus old assay methodologies.
• Employing a reference population for long-term monitoring and defining decision rules for clinical significance.

Main Results:

• The United Kingdom Prospective Diabetes Study (UKPDS) provided a real-world setting for validating these methods.
• Data from 5102 subjects randomized since 1977 were analyzed.
• The implemented techniques successfully managed data comparability despite numerous changes.

Conclusions:

• A structured approach is essential for ensuring biochemical data comparability in long-term clinical trials.
• Detailed procedures and comparisons with reference methods should be specified for inter-laboratory data comparison.
• These strategies are vital for the integrity and interpretability of findings from extensive research studies.