Molecular basis of 5-fluorouracil-related toxicity: lessons from clinical practice.

We aim to present a comprehensive review of the molecular basis of 5-fluorouracil (5-FU) toxicity, of which dihydropyrimidine dehydrogenase (DYPD) deficiency is a well-known mechanism. The prevalence of partial DYPD deficiency is fairly common, ranging between 3-5% in the general population, whereas it can be as high as 12% in African-American females. More than 50 genetic polymorphisms have been described as being associated with decreased enzymatic activity, whereas the c.1905+1G>A point mutation is the most commonly found (52% of cases), with a prevalence of heterozygosity in the general population ranging between 1-2%. Several methods have been utilized to identify reduced DYPD activity; functional tests are expensive and only available in specialized centers. Genotyping alone is not reliable enough, as some of the polymorphisms may not result in significantly reduced DYPD activity. The rate of cardiotoxicity associated with 5-FU or capecitabine does not seem to be related to DYPD deficiency, and has been estimated to range between 1.2-8%. Several pathophysiological mechanisms seem to contribute to 5-FU cardiotoxicity, including coronary spasm, increased endothelial thrombogenicity and myocardial inflammation. Tegafur/uracil and raltitrexed may be alternative options for patients with partial DYPD deficiency and previous manifested 5-FU cardiotoxicity, respectively. Pharmacogenetics is expected to further identify and clarify the mechanisms associated with 5-FU-related toxicity, thus aiding the oncology societies to formulate specific guidance on pre-treatment testing.