Center for Human Toxicology, University of Utah, Salt Lake City UT 84112, USA

THC-d₃ and THCA-d₃ are added to each 1-mL plasma specimen as internal standards; protein is precipitated with 1 mL of acetonitrile and the resulting supernatants diluted with 4 mL of 0.1 M sodium acetate (pH 7.0) prior to application to Clean Screen (2ST4C020) SPE columns. THC and THCA were eluated separately and then pooled, dried under air and then derivatized with trifluoroacetic anhydride and hexafluoroisopropanol. Under negative chemical ionization conditions the derivatized THC-d₀ gives abundant molecular anions (m/z 410) and the derivatized THCA-d₀ gives abundant fragment ions (m/z 422) formed by loss of (CF₃)₂CHO from its molecular anion.

The recoveries of THC and THCA were 74 and 17%, respectively. Intra- and inter-run accuracy and precision at 1 ng/mL (% target ± % CV) for THC were 98.0 ± 4.1% and 95.7 ± 2.8%, respectively. Even with low recovery of THCA, the method had acceptable intra- and inter-run accuracy and precision of 102.0 ± 2.9% and 104.3 ± 13.8%, respectively, at 1 ng/mL. Similar results were achieved at 10 and 75 ng/mL. A lower limit of quantitation of 0.5 and 1.0 ng/mL was established for THC and THCA, respectively. The upper limit of quantitation was 100 ng/mL. This method is currently being used to establish stability parameters for THC and THCA in plasma and will be employed in a pharmacokinetic study of human subjects exposed to cannabis.