Capecitabine: Have We Got the Dose Right?

Rachel Midgley; David J Kerr


Nat Clin Pract Oncol. 2009;6(1):17-24. 

In This Article

Interindividual Variation in Capecitabine Metabolism

Gender and Age

A range of factors can contribute to the differential absorption, distribution, metabolism and pharmacodynamics of capecitabine. The literature on the influence of gender on the pharmacokinetics of fluoropyrimidines is conflicting. The cumulative body of evidence, however, indicates that the clearance of capecitabine in women is less than that in men, with the area under the plasma drug concentration–time curve (AUC) and the maximum concentration of the drug in the body after dosing (Cmax) of FBAL being approximately 10% and 20% higher, respectively, in women than in men. The 2005 report from the manufacturers of capecitabine to the European Medicines Evaluation Agency describes an age-related increase in concentration of FBAL (20% increase in age leads to a 15% increase in AUC).[5] This increase in FBAL concentration was thought to be due to an age-related decrease in renal function. In addition, the manufacturers state that because there is also increased sensitivity to the toxicity of 5-FU with increasing age, special care must be taken when using fluoropyrimidine-based chemotherapy in elderly patients.

Body Weight

Pharmacokinetic studies indicate that a high body surface area is associated with a high volume of distribution and a decrease in the clearance of FBAL, the main end metabolite of capecitabine. Low body weight can influence the accuracy of calculations of creatinine clearance (usually causing a relative over-estimation), which has an effect on the calculated safe dose of capecitabine and could lead to relative overdosing of the drug. These observations fit in with clinical data from early breast cancer trials, which indicate that there is an inverse 'bell-shaped' curve relationship between body weight and drug toxicity, with worse safety in low-weight and high-weight patients.[6] The poor safety profiles in these two groups might be effectively compensated for by using alternative methods to assess creatinine clearance at the extremes of the weight range and perhaps by considering capping body surface area to 2 m2 at the upper end of the weight range.

Hepatorenal Dysfunction

Given the propensity of colorectal cancer to metastasize to the liver, hepatic dysfunction is relatively common in patients with this malignancy. A doubling of alkaline phosphatase activity in patients receiving capecitabine is associated with an 11% decrease in 5-FU clearance and a 12% increase in the AUC.[7] Twelves et al. correlated the degree of hepatic dysfunction in patients who had metastatic colorectal cancer with calculated pharmacokinetic parameters for capecitabine.[8] The Cmax and the AUC for the main metabolites of capecitabine were elevated to some extent in patients with mild to moderately impaired liver function, but not sufficiently so to warrant a recommendation of dose reduction.

Capecitabine can induce sufficient hemolysis to cause an isolated rise in bilirubin levels (reported in 20–70% of patients), which can often lead to a degree of confusion over appropriate dose because 5-FU can be contraindicated in patients with hepatic dysfunction. If baseline bilirubin levels were normal before starting capecitabine therapy, however, and there is no other reason to suspect deterioration in hepatic function, it is reasonable to proceed with the regular drug dose. Interestingly, Gieschke and coauthors found that capecitabine-related grade 3 and/or 4 bilirubinemia correlated positively with the AUC of 5-FU (P = 0.025), but negatively with the Cmax of FBAL (P = 0.014).[7] These results indicate that slow conversion of 5-FU to its inactive catabolite leads to prolonged exposure of red blood cells to the active drug and increased hemolysis.

More importantly, a 50% decrease in creatinine clearance is associated with a 35% decrease in FBAL clearance and, subsequently, a 41% and 53% increase in FBAL, Cmax and AUC, respectively.[7] Although FBAL does not have antiproliferative activity, concentration–effect analyses have shown a positive relationship between the AUC of FBAL and treatment-related grade 3 and/or 4 diarrhea (P = 0.035) and between the Cmax of FBAL and treatment-related grade 3 and/or 4 adverse effects. This relationship between exposure to FBAL and safety does not necessarily mean that FBAL is causing these adverse effects, but rather that FBAL, the main catabolite of 5-FU, might be a marker of the amount of 5-FU formed in the tissues. The pharmacokinetic data translate into a clear recommendation that physicians should tailor capecitabine dose on the basis of renal function (20–25% reduction in standard dose for a creatinine clearance of 30–50 ml/min; therapy withheld if creatinine clearance is less than 30 ml/min).

Drug–drug Interactions

During the first cycle of capecitabine development, three concomitant drugs were found to affect 5-FU clearance.[5] Paracetamol and morphine increased 5-FU clearance by 26% and 41%, respectively, and loperamide decreased 5-FU clearance by 31%; however, in later treatment cycles, the AUC of 5-FU was similar in patients with and those without these concomitant medications. Possible explanations for this subsequent lack of effect of capecitabine on AUC might include reduction in doses due to initial toxic events, such as hand–foot syndrome requiring analgesics or diarrhea requiring loperamide. It is difficult to draw firm conclusions from these limited data (9–12 patients) but further exploration of these interactions is warranted.


The three metabolic enzymes responsible for the biotransformation of capecitabine to 5-FU show fairly specific expression in liver and tumor tissue. Following conversion to 5-FU, 60–90% of capecitabine is catabolized by dihydropyrimidine dehydrogenase (DPD) ultimately to FBAL and 10–20% is excreted in urine unchanged.

There is a growing body of evidence to indicate that germ-line single nucleotide polymorphisms (SNPs) in candidate genes associated with either the molecular target(s) of a drug or the enzymes and/or transporters involved in its metabolism and excretion can determine the pharmacokinetic profile and hence spectrum of toxic effects and degree of efficacy of the drug.[9] The majority of SNP analyses for capecitabine have focused on its target, thymidylate synthase (TS), and the dominant enzyme involved in catabolism of 5-FU, DPD, although there are some preliminary data linking SNPs in genes encoding carboxylesterase and cytidine deaminase to clinical outcome.[9]

There is no doubt that DPD deficiency is a source of life-threatening toxic effects for patients treated with capecitabine and, although large confirmatory studies are required, the DPD genotype might be highly informative for prediction of adverse events. The most common mutation in DPD-deficient patients, a mutation in the splice donor consensus sequence of intron 14 that results in a truncated protein, has been observed in the Caucasian population at frequencies of 0.91–0.94%.[10] Conversely, there are some patients in whom DPD will be overexpressed, decreasing tumor cell exposure to the active drug and perhaps attenuating efficacy. An inhibitor of DPD, RO00948889, is being developed and might be useful in this population of patients.[11]

It has also been suggested that SNPs in the promoter region of TS, which lead to increased intracellular enzyme levels, are associated with a poor response to capecitabine therapy.[12] It might be possible to develop a panel of SNPs that could be used to prospectively select patients who would benefit most from capecitabine therapy and those who might require dose reductions to avert toxic effects, irrespective of hepatorenal function. This idea forms the basis of an interesting and potentially productive area of translational research, but this field has so far been hindered by a focus on rather small hypothesis-generating studies rather than definitive trials. Further systematic research on large patient populations is required before validated SNP tests can enter routine clinical practice.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.