Irinotecan Pharmacogenomics: Challenging Historical Precedent

The concept of “personalized medicine” (PM) or “individualized therapy” continues to become a more central treatment paradigm in the management of malignancies.  Multigene assays of tumor tissue are now commercially available and numerous major cancer centers are implementing processes to analyze the results of these assays and help oncologists translate these findings into patient management. 

As a Clinical Pharmacogenomic Scientist at Moffitt Cancer Center, this is central to my daily work.  However, the somatic genetic assessment of a patient’s tumor is only half of the PM story.  Germline genetics are still important and also determine how a patient responds to a particular chemotherapy.  The topoisomerase I inhibitor, irinotecan, is an example of a common antineoplastic agent whose toxicity is partially related to a germline mutation in the detoxifying enzyme uridine 5'-diphospho-glucuronosyltransferase (UDP-glucuronosyltransferase or UGT).

Irinotecan is a camptothecin analog approved by the FDA in 1996 and is currently used most commonly in the standard of care treatment of metastatic colorectal cancer, in combination with 5-fluorouracil and leucovorin as the FOLFIRI regimen.  Irinotecan is metabolized to its active metabolite, SN-38.  SN-38 is inactivated by UGT1A1, which is a polymorphic hepatic enzyme.  Patients with a dinucleotide repeat sequence in the promoter of the UGT1A1 gene (seven TA repeats instead of the wild-type 6 TA repeats), are classified as carrying the *28 allele.  A direct relationship between the number of TA repeats and risk of toxicity--specifically severe neutropenia and diarrhea--has been reported in patients with the UGT1A1*28 allele.¹  This relationship is most clinically relevant when the drug is given as a single agent every 3 weeks.  These findings led to the addition of the following statement in the irinotecan package insert: “Individuals who are homozygous for the UGT1A1*28 allele are at increased risk for neutropenia following initiation of Camptosar treatment” and the recommendation to reduce the starting dose “at least one level” if using the drug as a single agent.  This lack of specific dose recommendation makes it difficult to translate into clinical practice. 

In the August 2014 issue of the Journal of Clinical Oncology, the results of the first prospective, genotype-guided, dose escalation phase I clinical trial with single agent irinotecan were reported.²  Irinotecan was given as a flat dose, rather than the standard body surface area (BSA) normalized dosing, since it has been shown that exposure does not correlate with BSA.  Sixty-eight patients with histologically confirmed solid tumors or lymphoma refractory to standard therapy were enrolled on one of three arms, based on UGT1A1 genotype.  Thirty-one patients with UGT1A1 *1/*1 (homozygous wild-type) were enrolled at dose levels escalated from 700 mg to 1000 mg, with the maximum tolerated dose (MTD) determined to be 850 mg.  Twenty-eight patients with UGT1A1 *1/*28 (heterozygous) were enrolled at dose levels escalated from 700 mg to 850 mg, with the MTD determined to be the starting dose of 700 mg.  Nine patients with UGT1A1 *28/*28 (homozygous mutant) were enrolled at the initial dose of 500 mg, but based on 3 dose-limiting toxicities (DLT), the dose was de-escalated to 400 mg, which was tolerated with one DLT out of six patients and determined to be the MTD. The most common DLTs were neutropenia and diarrhea, as expected.  The exposure to the active metabolite SN-38 was similar across all of the genotypes when irinotecan was dosed at the MTD. 

Given the phase I nature of the trial, response was not a primary endpoint and patients were heterogeneous and heavily pretreated, so this will need to need to be assessed in future trials, which are underway.  Genotype-guided irinotecan is being assessed in the more standard setting of FOLFIRI with bevacizumab in the metastatic colorectal cancer population, and as part of mFOLFIRINOX in previously untreated advanced gastrointestinal malignancies. 

This trial represents the application of many years of retrospective clinical data and serves as an example of how individualized dosing could ideally be determined; however, several limitations prevent this from becoming the new standard.  As discussed by Phelps and Sparreboom in their accompanying editorial, one notable limitation to this design becoming a standard model in drug development is the amount of information available on the metabolism of irinotecan prior to design of the trial.³  The metabolism pathways are generally known to some degree for novel therapies undergoing early phase clinical trials, but the interaction between variants in these metabolic pathways and outcomes-- like toxicity or efficacy-- are mere speculation making a prospectively designed study time consuming and challenging.  Novel clinical trial design such as this should be encouraged to help optimize the delivery of individualized therapy across somatic and germline genetic variations. 

The results of the ongoing genotype-guided irinotecan trials with FOLFIRI and FOLFIRINOX regimens will provide data-supported dosing recommendations based on genotype for irinotecan in the most frequent clinical settings, allowing for individualized irinotecan dosing to be translated into standard clinical practice.

References:

• Lyer L, et al. (1998). Genetic predisposition to the metabolism of irinotecan (CPT-11). Role of uridine diphosphate glucuronosyltransferase isoform 1A1 in the glucuronidation of its active metabolite (SN-38) in human liver microsomes. J Clin Invest. 101(4): 847-854.
• Innocenti F, et al. (2014). Dose-Finding and Pharmacokinetic Study to Optimize the Dosing of Irinotecan According to the UGT1A1 Genotype of Patients With Cancer.  J Clin Oncol. 32:2328-2334.
• Phelps MA, Sparreboom A.  (2014). Irinotecan Pharmacogenomics: A Finished Puzzle?  .  J Clin Oncol. 32:2287-2292.