thebench.bio #1 Ecosystem for the Bio-Tech Industry
All Quality Control questions

Question 7 of 20 Fundamentals Concept

Which performance characteristics does ICH Q2(R2) ask you to validate?

Short answer

ICH Q2(R2) groups validation into specificity or selectivity, range (the response or calibration model and the lower range limits), accuracy and precision. Robustness is expected but is now normally shown during development under ICH Q14. Which tests apply depends on the procedure's purpose.

Why

The 2023 revision of Q2 reorganized the old list of characteristics. Specificity or selectivity shows the procedure measures the analyte without interference, which can be proven by absence of interference, by comparison with an orthogonal procedure, or by a justification inherent to the technology. Range replaces the old separate headings for linearity and detection or quantitation limits.

It covers the response, which can be linear, non-linear or multivariate, and validation of the lower range limits. The reportable range is usually derived from the specification. For an assay that is typically 80 to 120 percent of declared content or of the specification limits.

Accuracy is shown against a reference material, by spiking, or by an orthogonal procedure. Precision covers repeatability, with at least 9 determinations across the range or 6 at 100 percent, intermediate precision across days, analysts and equipment, and reproducibility between laboratories when needed. A new option is to assess accuracy and precision together against a combined criterion, using a prediction or tolerance interval.

The purpose decides the set. An identity test needs specificity but not a reportable range, while an impurity test needs a lower range limit. Data from development can count toward validation, so the old habit of repeating everything in a separate study is not required.

A strong answer covers

  • Specificity or selectivity, range, accuracy, precision
  • Range includes response model and lower range limits
  • Repeatability, intermediate precision, reproducibility
  • Robustness shown in development under Q14
  • Tests chosen by the procedure's intended purpose

Common mistakes

  • Listing linearity, LOD and LOQ as separate headings as if Q2(R1) still applied unchanged.
  • Validating every characteristic for every procedure regardless of purpose.
  • Ignoring that development data can be used as validation evidence.