Clinical Study

British Journal of Cancer (2008) 99, 1239–1245. doi:10.1038/sj.bjc.6604673 www.bjcancer.com
Published online 16 September 2008

Pharmacokinetic and pharmacogenetic determinants of the activity and toxicity of irinotecan in metastatic colorectal cancer patients

E Rouits1,4, V Charasson2,4, A Pétain3, M Boisdron-Celle1, J-P Delord3, M Fonck2, A Laurand2, A-L Poirier1, A Morel1, E Chatelut3, J Robert2 and E Gamelin1

  1. 1Laboratoire d'Oncopharmacologie, Centre Paul-Papin, 2 rue Moll, Angers 49000, France
  2. 2Département de Pharmacologie INSERM U916, Institut Bergonié, 229 cours de l'Argonne, Bordeaux 33076, France
  3. 3Équipe d'accueil 3035, Université de Toulouse, Institut Claudius-Regaud, 20 rue du Pont Saint-Pierre, Toulouse 31052, France

Correspondence: Dr J Robert, E-mail: robert@bergonie.org

4These authors contributed equally to this work.

Received 10 July 2008; Revised 14 August 2008; Accepted 22 August 2008; Published online 16 September 2008.

Top

Abstract

This study aims at establishing relationships between genetic and non-genetic factors of variation of the pharmacokinetics of irinotecan and its metabolites; and also at establishing relationships between the pharmacokinetic or metabolic parameters and the efficacy and toxicity of irinotecan. We included 49 patients treated for metastatic colorectal cancer with a combination of 5-fluorouracil and irinotecan; a polymorphism in the UGT1A1 gene (TA repeat in the TATA box) and one in the CES2 gene promoter (830C>G) were studied as potential markers for SN-38 glucuronidation and irinotecan activation, respectively; and the potential activity of CYP3A4 was estimated from cortisol biotransformation into 6beta-hydroxycortisol. No pharmacokinetic parameter was directly predictive of clinical outcome or toxicity. The AUCs of three important metabolites of irinotecan, SN-38, SN-38 glucuronide and APC, were tentatively correlated with patients' pretreatment biological parameters related to drug metabolism (plasma creatinine, bilirubin and liver enzymes, and blood leukocytes). SN-38 AUC was significantly correlated with blood leukocytes number and SN-38G AUC was significantly correlated with plasma creatinine, whereas APC AUC was significantly correlated with plasma liver enzymes. The relative extent of irinotecan activation was inversely correlated with SN-38 glucuronidation. The TATA box polymorphism of UGT1A1 was significantly associated with plasma bilirubin levels and behaved as a significant predictor for neutropoenia. The level of cortisol 6beta-hydroxylation predicted for the occurrence of diarrhoea. All these observations may improve the routine use of irinotecan in colorectal cancer patients. UGT1A1 genotyping plus cortisol 6beta-hydroxylation determination could help to determine the optimal dose of irinotecan.

Keywords:

colorectal cancer, irinotecan, pharmacokinetics, UGT1A1 gene polymorphism, CYP3A4