Wide interindividual variation in responses to cyclooxygenase (COX) inhibitory drugs limits their clinical utility and safety.Methods
To better understand the molecular responses to COX inhibition, we analyzed the gene expression level of the genes encoding enzymes related to prostaglandin production including the COX-1 gene (PTGS1) and the COX-2 gene (PTGS2), as well as their genetic polymorphisms, and the analgesic response to COX inhibitory drugs such as ibuprofen or rofecoxib or to placebo after minor surgery.Results
Notable heterogeneity in global gene expression was evident between subjects. At 2 to 4 hours after surgery, PTGS1 expression was slightly decreased (36%, P < .001) and PTGS2 expression was markedly increased (300%, P < .001) with wide interindividual variation; at 48 hours after surgery, little detectable change in PTGS1 and PTGS2 expression was found in the control group. However, ibuprofen and rofecoxib treatment significantly increased PTGS2 expression at 48 hours (P = .001 and P = .049, respectively). At 2 to 4 hours after surgery, patients with the G/G allele at the nucleotide position of −765G>C in PTGS2 showed a significantly higher increase in PTGS2 expression (P = .012) compared with G/C and C/C patients, although all of them showed an increase in PTGS2 expression (P < .001 and P = .043, respectively). Among G/G patients, rofecoxib administration resulted in significantly lower pain intensity on a visual analog scale (7.2 ± 2.5 mm) (P = .008) at 48 hours after surgery, as compared with ibuprofen administration (31.3 ± 6.7 mm). The finding regarding pain intensity at 48 hours in G/C and C/C patients was opposite (P = .002), being greater in the rofecoxib group (37 ± 6.8 mm) compared with the ibuprofen group (7 ± 1.9 mm).Conclusions
These results suggest that wide variability in gene expression and functional polymorphisms in PTGS2 may explain part of the interindividual variations in acute pain and the analgesic efficacy of nonsteroidal anti-inflammatory drugs and selective COX-2 inhibitors; this may be useful to define individual responders on the basis of genetic variations to predict patient risk and benefit to drugs.