The Coriell personalized medicine collaborative pharmacogenomics appraisal, evidence scoring and interpretation system

Table 5 Genotype-phenotype drug response interpretations of the Pharmacogenomics Advisory Group-approved drug-gene pair celecoxib and CYP2C9

Common diplotypes^b	Expected population frequency, %^c			PK/PD phenotype	Clinical phenotype	FDA guidelines
	Caucasian ancestry	African ancestry	East Asian ancestry
CYP2C91/1	67	84.5	92	EM: normal enzymatic function and drug elimination	Expected to have a normal analgesic response at standard dose of celecoxib. Colorectal adenoma treatment: no additional efficacy with 400 mg celecoxib twice daily compared with 200 mg twice daily	–
CY2C91/2, CYP2C1/3*, and other rare diplotypes (see Additional file 1: Table S10)	30	15	8	Likely IM: reduced enzymatic function and drug elimination, leading to increased drug exposure	Insufficient data; predicted risk of side effects is unknown	–
CYP2C93/3, CYP2C2/3, CYP2C2/2, and other rare diplotypes (see Additional file 1: Table S10)	3	0.6	0.2	PM: greatly reduced enzymatic function and drug elimination, leading to greater drug exposure	Greater risk of adverse cardiovascular events with 400 mg celecoxib twice daily. Colorectal adenoma treatment: decreased recurrence with 400 mg celecoxib twice daily	Consider 50% of the standard starting dose in PMs; consider alternative treatment in PMs with juvenile rheumatoid arthritis

Abbreviations: EM Extensive metabolizer, FDA Food and Drugs Administration, IM intermediate metabolizer, PD pharmacodynamic, PK pharmacokinetic, PM poor metabolizer.
^aSupporting evidence may be found in Additional file 2: Section S3.0 to 3.7 and Additional file 1: Tables S8 to S10. These include summaries of the PhAESIS evaluation and referenced publications supporting the drug-gene clinical phenotypes.
^bDiplotypes with frequencies of less than 0.4% in Caucasians are included above. Other rare diplotypes that fall under the same phenotype category can be found in the genotype-phenotype Punnett table (see Additional file 1: Table S10). Diplotypes above and in the genotype-phenotype Punnett table include both clinically validated genetic results (those that include variants with evidence code 1) and results that include variants with evidence scores 2 to 7 (potentially clinically relevant). The latter require further validation to support their inclusion for clinical reporting.
^cPopulation frequencies are estimated based on reported gene variant allele frequencies (see Additional file 1: Table S9) and Hardy-Weinberg principles.

ISSN: 1756-994X