Authors’ response: Relationships between the adverse effects of aspirin and polymorphisms in genes encoding aspirin metabolizing enzymes

We thank Miners & Day for their interest in our work and for their critical comments on our paper [1]. We agree with Miners & Day that the enzymes UGT1A6 and CYP2C9 may not be the main enzymes involved in the metabolism of aspirin [2]. However, in answer to the comments of Minors & Day, we want to address following: 1) Our main goal was to study possible associations between genetic polymorphisms in UGT1A6 and CYP2C9, both involved in the metabolism of aspirin, and gastrointestinal complaints as caused by cardiovascular dosed (80–100 mg) aspirin or one of its metabolites [1]. This does not necessarily need to be the main metabolite(s) of aspirin. It is not irrational to hypothesize, that the highly reactive product(s) of a minor oxidative metabolic pathway, often catalyzed by CYPs (here CYP2C9), may cause more damage or complaints as compared with a main metabolic product. 2) The data on metabolism of aspirin in humans, as given by Minors & Day, are based on studies by Hutt et al.[2], giving an oral dose of 900 mg aspirin, whereas our patients received 80–100 mg aspirin daily. This difference in dose may result in a different pattern of metabolites in urine from that given by Minors & Day, and since glucuronidation is often capacity limited, a low dose of aspirin may result in a larger contribution of glucuronide metabolites. (3) Minors & Day state that the UGTs involved in glucuronidation of salicylic acid are not characterized unambiguously, but by referring to the study of Keuhl et al.[3] they suggest that UGT1A6 may only be just one of several UGTs involved. However, apparent K_ms for both phenolic- and acyl-glucuronidation of salicylic acid are lowest for UGT1A6, suggesting that this enzyme may be most relevant in the glucuronidation of low-dose aspirin.

Association studies between single nucleotide polymorphisms, with possible consequences for the activity of the drug metabolizing enzymes they are coding for, and a clinical outcome are interesting, but as Miners & Day also indicated, conclusions may not always be generalized. We agree on this point; our reported study aimed, in the first place, to reveal the aetiology of side-effects of low-dose aspirin. From a scientific point of view such studies are worthwhile, but conclusions are not warranted and must be drawn carefully.

This is also the case in the studies cited by Miners & Day about the association between UGT1A6, CYP2C9 and colorectal adenoma or carcinoma risk. Bigler et al.[4] reported an inverse association between adenoma risk and aspirin use in subjects with a CYP2C9*1/variant UGT1A6, where McGreavey et al.[5] found that polymorphisms in CYP2C9 and UGT1A6 did not influence the protective effect of NSAIDs on the risk of colorectal cancer. Both studies were based on the same hypothesis, that genetic polymorphisms in aspirin metabolizing enzymes may influence the preventive effects of the drug, whereas in our study we were interested in negative side-effects of the drug. However, in addition to effects on the drug metabolizing enzymes of the polymorphisms under study, results may also be highly dependent on the dose of the drug applied, which is not given in both studies cited above [4, 5].