Role for Drug Transporters Beyond Tumor Resistance: Hepatic Functional Imaging and Genotyping of Multidrug Resistance Transporters for the Prediction of Irinotecan Toxicity

Abstract

The multigenic nature of drug response is increasingly recognized and underscores our current difficulties with adequately predicting interindividual variability in both the response to and toxicity of many chemotherapeutic agents. The clinical pharmacology of irinotecan nicely illustrates the complexity of drug response pathways and highlights the pressing need to develop easily measured markers that are predictive of drug disposition, toxicity, and efficacy. Until recently, the main focus has been on understanding the variation in irinotecan metabolism, since it was considered the primary factor controlling systemic exposure to this drug. Irinotecan is a prodrug that must be metabolized to the active compound SN-38, a potent inhibitor of topoisomerase I. Metabolism of irinotecan to SN-38 is catalyzed largely by carboxylesterase-2 (CES2), and CYP3A4/CYP3A5 catalyzes the formation of inactive metabolites. A significant fraction of SN-38 is converted to the glucuronide conjugate (SN-38G)—a conversion catalyzed largely by UGT1A1. More than 10 years of research have nicely illustrated the critical role of UGT1A1 activity in determining the pharmacologic response to irinotecan, as well as in the development of its rate-limiting toxicities of neutropenia and lateonset diarrhea. In vivo and genetic markers have been established for many drug metabolizing enzymes, and in the case of UGT1A1, genotyping alone is highly predictive of function. However, metabolism is only one factor controlling the systemic and cellular exposure to irinotecan and its metabolites. Drug transporters have also been implicated in the disposition of irinotecan, SN-38 and SN-38G, with evidence for transport by P-glycoprotein (Pgp; encoded by ABCB1), multidrug resistance-associated protein 1 (MRP1; ABCC1), MRP2 (ABCC2), and mitoxantrone resistance protein (MXR; ABCG2). More recently, an organic anion transporting polypeptide (OATP1B1; SLCO1B1) was shown to transport SN-38 but not irinotecan or SN-38G. An accurate prediction of those patients most likely to develop life-threatening toxicities during irinotecan therapy needs to consider variation in not only the metabolism of this important drug but also its transport into and out of both tumor cells and those cells associated with its disposition and toxicity. The article in this issue by Michael et al is an important contribution to the development of more comprehensive biomarkers of irinotecan disposition. Hepatic imaging was performed in advanced colorectal cancer patients using well-characterized substrates of the multidrug resistance transporters Pgp, MRP1 and MRP2, and hepatic functional imaging parameters were correlated with irinotecan pharmacokinetics. The area under the curve (AUC) of SN-38 was positively correlated with the hepatic retention of the imaging agents, a significant finding since SN-38 exposure is linked to the neutropenia and late-onset diarrhea associated with irinotecan therapy. Indeed, there was a trend toward an association between hepatic imaging parameters and toxicity. A preliminary analysis of ABCB1 genotypes was also performed and is consistent with the important role of Pgp in the disposition of the imaging agents. Hepatobiliary imaging was developed for the evaluation of hepatic dysfunction and to monitor the response to drugs and other interventions. The two classes of Tc-labeled agents that were used in these studies are rapidly taken up into the hepatocyte and eliminated into the bile by distinct active transport mechanisms. The iminodiacetic acid (IDA) analogs (Tc-DIDA and Tc-DISIDA) are organic anions and likely require uptake transporters for entry into the hepatocyte. Biliary excretion of TcDIDA and Tc-DISIDA is mediated by MRP2 while transport by MRP1 will eliminate drug from the hepatocyte back into the systemic circulation. Consistent with the significant role of MRP2 in the elimination of IDA agents, their biliary excretion is decreased in Dubin-Johnson syndrome patients who often carry polymorphisms in ABCC2 that result in a loss of MRP2 function. Tc-labeled sestamibi (Tc-MIBI) is a Pgp substrate and has been used to demonstrate Pgp inhibition by pharmacologic agents and genetic disruption. These imaging agents have great potential in the evaluation of drug transport function in the liver and fill a current void in this area. The lack of selective substrates that are metabolically stable has made it difficult to assess in vivo transporter function. Most substrates of Pgp, MRP1, and MRP2 undergo significant first pass and systemic metabolism and pharmacokinetic parameters for JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 24 NUMBER 26 SEPTEMBER 1