Innovative Framework for Blinded Evaluation of Methods: Application to Cannabinoid Quantification in Hemp

Regulatory changes and growth of hemp-derived products have increased the need for robust methods to quantify cannabinoids.

To systematically evaluate a liquid chromatography photo-diode array (LC-PDA) method for 11 cannabinoids in hemp using a blinded collaborative framework, with emphasis on both fortified and endogenous analytes.

A previously validated method was set up at a new‑to‑method laboratory (Veterinary Medical Diagnostic Laboratory, VMDL) and verified in-house using fortified samples. Discrepancies between certified and measured values for a hemp reference material raised questions about the method's performance for endogenous cannabinoids. To address this, an independent laboratory conducted a two-round collaborative exercise (Blinded Method Test, BMT) by preparing blind-coded samples using hemp, nettle, and hemp-nettle mixtures with various fortification levels. VMDL quantified cannabinoids by external calibration (EC) in acetonitrile (surrogate solution). Parallelism evaluation and standard-addition calibration were complementarily applied to address matrix effect and accuracy for endogenous analytes.

The method showed good linearity and acceptable accuracy and precision for blinded fortified hemp, nettle, and hemp-nettle samples and quality controls using EC. Parallelism results supported the use of solvent-based EC for endogenous cannabinoids in hemp. Standard-addition estimates were directionally consistent with EC but differed by about 15-45% for some analytes. These findings underscore the need for more detailed characterization of extraction efficiency for endogenous cannabinoids to strengthen accuracy assessment for incurred cannabinoids in complex botanical matrices.

The two-round BMT demonstrates how collaborative exercises can extend conventional validation by integrating multiple matrices, parallelism testing, and complementary quantification approaches. The evaluated method is suitable for routine hemp testing at VMDL, while the findings emphasize the need to evaluate extraction efficiency for incurred versus fortified analytes in complex biomatrices.

A collaborative framework with parallelism and standard addition provides enhanced evaluation of method performance for complex botanical matrices.