Higher Doses Of Irinotecan Research Articles

Selective TACE with irinotecan-loaded 40 \u03bcm microspheres and FOLFIRI for colorectal liver metastases: phase I dose escalation pharmacokinetic study

BackgroundEfficacy of treatments for colorectal liver metastases after failure of first-line chemotherapy is limited. The aim of this study was to prospectively evaluate the feasibility, tolerability, and pharmacokinetics of selective transarterial chemoembolization (TACE) with irinotecan-loaded 40 μm microspheres combined with systemic FOLFIRI for colorectal liver metastases refractory to oxaliplatin regimen.MethodsThe dose escalation study was conducted in three patient groups with different amounts of irinotecan loaded (50, 75 and 100 mg per mL-microspheres). Selective catheterization was performed to embolize subsegments or segments of located tumors using TACE navigation system. FOLFIRI was administrated 7 days after TACE. Plasma concentration was measured before and time points after administration.ResultsNine patients successfully underwent a total of 22 TACE procedures. Dose-limiting toxicity did not appear at any level. The overall response rate was 55.6%. The median progression free and overall survival were 8.1 and 18.2 months, respectively. The AUC and Cmax of plasma SN-38 per 1 mg injected irinotecan dose were significantly higher in irinotecan-loaded microspheres compared with FOLFIRI (P = 0.009 and P < 0.001, respectively).ConclusionSelective TACE using 40 μm irinotecan-loaded microspheres combined with systemic FOLFIRI was feasible and safe even when a high dose of irinotecan was loaded. Irinotecan-loaded microspheres resulted in a higher plasma concentration and AUC of SN-38 than treatment with FOLFIRI. Further large scale trials to evaluate the efficacy are mandatory.Trial registrationUniversity Hospital Medical Information Network (UMIN) Clinical Trials Registry, Registration number; UMIN000015367; Registered date; 08,10,2014.

PD-032 - Pharmacogenetic clinical randomized phase II trial to evaluate the efficacy and safety of FOLFIRI with high dose of irinotecan (FOLFIRI-HD) in metastatic colorectal cancer patients according to UGT1A 1 genotype

PD-032 - Pharmacogenetic clinical randomized phase II trial to evaluate the efficacy and safety of FOLFIRI with high dose of irinotecan (FOLFIRI-HD) in metastatic colorectal cancer patients according to UGT1A 1 genotype

A Phase I Study of UGT1A1 *28/*6 Genotype-Directed Dosing of Irinotecan (CPT-11) in Korean Patients with Metastatic Colorectal Cancer Receiving FOLFIRI

Purpose: A UGT1A1 genotype-directed dose escalation of irinotecan (CPT-11) was performed in patients with metastatic colorectal cancer receiving first-line FOLFIRI chemotherapy. Methods: Patients were genotyped for UGT1A1 and stratified according to the number of defective alleles (DA; *28 and *6). The irinotecan dose was escalated with a fixed dose of 5-fluorouracil and leucovorin in a standard 3 + 3 design. Results: In 43 enrolled patients, the maximum tolerated dose (MTD) was 300 mg/m² for the 1 DA group, while the MTD was not reached for the 0 DA group with 1 dose-limiting toxicity (DLT) at 330 mg/m² and for the 2 DA group with 0 DLT at 150 mg/m². Because of the risk of being exposed to unsafe doses, the trial was terminated before the MTD was reached in the 0 DA and 2 DA groups. The recommended doses were 300 (0 DA), 270 (1 DA) and 150 (0 DA) mg/m². The 2 DA group displayed 27% lower SN-38 exposure levels relative to the 0 and 1 DA groups (95% CI, 0.47-1.15). Conclusions: The MTD of irinotecan differed according to the UGT1A1 genotype, and higher doses of irinotecan are feasible with sLV5FU2 compared to the present regulatory approved doses. © 2014 S. Karger AG, Basel

Abstract 4627: Fasting reduces the systemic exposure to irinotecan and its active metabolite SN-38

Abstract Limitations to the use of chemotherapeutic agents are the often severe adverse side-effects which may lead to early discontinuation of treatment. Most anticancer agents have a narrow therapeutic index. In previous work we have shown that 72 hours of fasting prior to treatment with a high dose of irinotecan prevents the occurrence of adverse side effects in C26 colon carcinoma bearing mice, while the antitumor activity is not abrogated. To elucidate the mechanism of fasting induced resistance against adverse side effects, we have examined the pharmacokinetics of irinotecan in 72 hours fasted mice in plasma and liver. Male BALB/c mice were divided into four groups (n=18/group). Two groups were fasted for 72 hours and two groups were fed ad libitum. After the fasting period, mice were fed ad libitum again. One group of ad libitum and fasted animals was treated with 50 mg/kg and the other with 100 mg/kg irinotecan intraperitoneally. Plasma and liver were collected at 1,4,8,12,24 and 48 hours after irinotecan injection. Tissues were homogenized in lithium heparinized plasma and concentrations of irinotecan (CPT-11) and the active metabolite SN-38 were determined using a validated reversed-phase high-performance liquid chromatography (HPLC) system with fluorescence detection. Pharmacokinetic parameters, including peak concentration, area under the plasma concentration-time curve (AUC), clearance (CL), and the half-life of the terminal disposition phase were calculated. In the fasted group, plasma levels of SN-38 in the 50 mg/kg and 100 mg/kg group showed a 57% and 53% reduction, respectively, compared with ad libitum fed controls (AUC0-inf 5,149 vs. 2,242 ng*h/mL and 7,507 vs. 3,562 ng*h/mL). For CPT-11 the effects of fasting were smaller. An increase of 26% in the 50 mg/kg group and a reduction of 15% in the 100 mg/kg group were observed (AUC0-inf 9,715 vs. 12,235 and 37,360 vs. 31,924 ng*h/mL). In the liver, SN-38 levels in the 50 mg/kg and 100 mg/kg group showed a reduction of 51% and 30%, respectively, in the fasted animals (AUC0-inf 85.4 vs. 41.5 µg*h/g and 126 vs. 87.6 µg*h/g), while CPT-11 levels were reduced with 19% in the 50 mg/kg group and 28% in the 100 mg/kg group (AUC0-inf 218 vs. 177 µg*h/g and 712 vs. 512 µg*h/mg). Our data demonstrate that 72 hours of fasting prior to irinotecan administration induces an important change in its metabolism. Plasma and liver levels of the pro-drug CPT-11 were only slightly reduced in fasted animals. Levels of the active metabolite SN-38 were considerably lower in fasted animals compared to ad libitum fed animals. These data suggest that the reduction of side effects by fasting is due to the lower systemic exposure to SN-38, and may have important clinical implications if also found in humans. Citation Format: Sander A. Huisman, P de Bruijn, I.M. Ghobadi Moghaddam-Helmantel, J.N.M. IJzermans, E Wiemer, A.H.J. Mathijssen, R.W.F. de Bruin. Fasting reduces the systemic exposure to irinotecan and its active metabolite SN-38. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4627. doi:10.1158/1538-7445.AM2014-4627

A UGT1A1 genotype-directed phase II toxicity and efficacy-finding study of irinotecan combined with S-1 as first-line treatment in advanced gastric cancer.

TPS4154 Background: The best chemotherapy regimen for advanced gastric cancer (AGC) is uncertain, but promising findings have been reported with Irinotecan (IRI) plus S-1. However, IRI can induce severe neutropenia or diarrhea associated with homozygosity of the UGT1A1*28 or UGT1A1*6 alleles. This trial was designed to compare the toxicity and efficacy on different doses of IRI combined with S-1 according to UGT1A1 genotype as first-line chemotherapy in AGC in Chinese patients. Methods: Previously untreated patients with histologically proven gastric or gastroesophageal junction adenocarcinoma, aged between 18 and 75 years with ECOG performance status 0-2 were classified according to UGT1A1 genotype: wild-type (none of *28 or *6 allele); heterozygous (only one of *28 or *6 allele); or homozygous (*28/*28, *6/*6, or double heterozygous for *28 and *6). Patients were randomized (1:1) to receive either low or high dose of IRI given i.v. (90 min) on day 1 in each genotype group. The low and high dose of IRI were 80mg/m2 and 200 mg/m2 in the wild-type group, 80 mg/m2 and 150 mg/m2 in the heterozygous group, 40 mg/m2 and 80 mg/m2 in the homozygous group. S-1 was administered orally at a dose level set on the basis of the body surface area (BSA): 40 (BSA&lt;1.25 m2), 50 (BSA≥1.25 to&lt;1.5 m2 ) or 60 mg (BSA≥1.5 m2) twice a day on day1-7. Courses were repeated every 2 weeks, unless disease progression, unacceptable toxicity or patient refusal. The primary endpoint was to explore the safety and efficacy on different doses of IRI combined with S-1 according to UGT1A1 polymorphism. Based on the frequency of UGT1A1*28 and UGT1A1*6 gene polymorphism in Asian gastrointestinal cancer patients, the chi-square test and fisher exact test were used to evaluate the planned sample size which is 100 in total: 40 for the wild-type, 40 for the heterozygous and 20 for the homozygous group. Until now, 8 patients have been enrolled. Clinical trial information: ChiCTR-OCH-12002472.

A UGT1A1*28 and *6 genotype-directed phase I dose-escalation trial of irinotecan with fixed-dose capecitabine in Korean patients with metastatic colorectal cancer

UGT1A1 genotypes are important when considering treatment with irinotecan-containing regimens. In this study, we determined the dose, efficacy, and tolerability of irinotecan according to UGT1A1 genotypes when combined with capecitabine in patients with metastatic colorectal cancer. Patients with histologically confirmed metastatic adenocarcinoma of the colon or rectum were enrolled into a UGT1A1 genotype-directed dose-escalation trial of irinotecan plus fixed-dose capecitabine (2,000 mg/m(2)/day). The starting dose of irinotecan was different for each genotype group and ranged from 200 to 280 mg/m(2). Pharmacokinetic concentrations of irinotecan and metabolites were determined by LC/MS/MS. Fifty patients were genotyped for UGT1A1 *28 and *6, and grouped according to the numbers of defective alleles (DA): 0, 1, and 2. Plasma concentrations of irinotecan, SN-38, and SN-38G were measured. The maximum tolerated dose of irinotecan was 350 mg/m(2) for the 0 and 1 DA groups, and 200 mg/m(2) for the 2 DA group. For the 0, 1, and 2 DA groups, mean AUClast ratios of SN-38G to SN-38 were 7.72, 5.71, and 2.72 (P = 0.0023) and relative dose intensities at recommended dose were 85, 83, and 97 %. Irinotecan dosing based on UGT1A1*28 and *6 is feasible, and higher doses of irinotecan can be safely administered in patients with 0 or 1 DA, compared to those with 2 DA.

Abstract 4417: Fasting protects against the adverse side effects of chemotherapy but has no effect on antitumor activity.

Abstract The multi-modality approach to colorectal liver metastases has led to an increase in the number of patients eligible for hepatic resection. The tumor response to chemotherapeutic drugs plays a pivotal role in the success of this approach. The limitations to their use are often severe adverse side-effects, resulting in an early discontinuation of the chemotherapy. In previous work we have shown that fasting leads to increased expression of cytoprotective and antioxidant genes. In this study we have examined the effects of fasting prior to administration of a high dose of the chemotherapeutic agent irinotecan on the occurrence of chemotherapy-associated adverse events and anti-tumor effect in C26 coloncarcinoma bearing mice. Male BALB/c mice were subcutaneously implanted with a 15 mm3 cube of C26 coloncarcinoma cells and divided into 4 groups (n=6/group). Ten days after implantation, 2 groups were fasted for 72 hours and 2 groups were fed ad libitum. After 72 hours all groups were fed ad libitum again. One ad libitum group and one fasting group were treated with a cumulative dose of 400 mg/kg irinotecan ip, administered on days 0, 2 and 4 relative to fasting. Tumor growth was measured daily using Vernier calipers. Ocurrence of adverse side effects was recorded daily. Leucocytes were counted on day 8 after the first irinotecan dose. All mice were sacrificed 10 days after the first irinotecan injection or earlier when tumor volume exceeded allowable range. Tumors were resected, measured and weighted. In the ad libitum fed group mice showed weight loss from the first irinotecan injection, while the fasted mice gained weight during the observation period. In the ad libitum fed mice other adverse side effects were observed from day 4 after the first irinotecan gift. They displayed different behaviour, reduced mobility, had ruffled hair, a hunched posture and diarrhea. The fasted mice showed no visible adverse side effects. The number of leukocytes in the fasted group treated with irinotecan was significantly higher, than in the ad libitum fed treated animals (6.5*10ˆ6/mL vs. 3.2*10ˆ6/mL, p&amp;lt;0.001). The experiment had to be terminated before the end of the observation period in 67% of the non-treated mice due to progressive tumor growth. In both irinotecan treated groups tumor growth was similarly suppressed compared with the fasted and ad libitum fed groups without irinotecan (1271 mg vs. 2106 mg, p&amp;lt;0.001). Our data demonstrate that 72 hours of fasting prior to treatment with a high dose of irinotecan prevents the occurrence of adverse side effects, while the antitumor activity is not affected. These data suggest that a preoperative fasting regimen may improve the therapeutic index of chemotherapeutic agents and increases the efficacy of chemotherapeutic treatment. Citation Format: Sander A. Huisman, Sandra van den Engel, Henk P. Roest, Jan N.M. IJzermans, Ron W.F. de Bruin. Fasting protects against the adverse side effects of chemotherapy but has no effect on antitumor activity. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4417. doi:10.1158/1538-7445.AM2013-4417

A UGT1A1 *28 and *6 genotype-directed phase I study of irinotecan plus infusional leucovorin and 5-fluorouracil (sLV5FU2) in patients with previously untreated metastatic colorectal cancer (mCRC).

3592 Background: We designed a phase I study to determine maximum tolerated dose (MTD) of irinotecan when combined with sLV5FU2 in mCRC patients (pts). Methods: Pts were genotyped for UGT1A1 *28 and *6, and stratified into 3 groups according to the number of defective allele (DA), designated 0 (*1/*1), 1 (*1/*28, *1/*6), and 2 (*28/*28, *6/*6, *6/*28). Within each group, the dose of irinotecan was escalated (table) in combination with fixed dose of sLV5FU2. Plasma drug levels and dose-limiting toxicity (DLT) were evaluated at cycle 1. Results: A total of 43 pts were accrued: 19 for 0 DA, 20 for 1 DA and 4 for 2 DA group. The MTD was estimated as 300 mg/m2/2-week for the 1 DA group with 2 DLTs in the level 3, and the MTD was not reached for the 0 DA group with 1 DLT in the level 4 (table). The mean relative extents of glucuronidation, AUClast ratio of SN-38G to SN-38, were 9.36, 6.81, and 5.09 for the 0, 1, and 2 DA groups, respectively (P=0.017). Of the 43 pts, five pts showed AUClast, SN38 that exceeded 400 ng·h/mL (1.02 umol·h/L) and DLT was observed in 40% (2/5). The overall response rate was 67.4% (95% CI, 51.5-80.9) with 6 complete responses and 23 partial responses. Median progression-free and overall survival was 8.0 months (95% CI, 7.1-8.9) and 25.6 months (95% CI, 23.4-27.7), respectively. Grade 3 or 4 toxicity during all treatment cycles included neutropenia (79% [0 DA]; 90% [1 DA]; 75% [2 DA]), leucopenia (21%; 30%; 0%), febrile neutropenia (0%; 10%; 0%) and diarrhea (0; 5%; 0) per patient. Conclusions: Dose-normalized exposure of SN38 was significantly higher in the 2 DA UGT1A1 group. Higher doses of irinotecan based on UGT1A1 genotyping are feasible when combined with sLV5FU2 in mCRC pts. The recommended dose of irinotecan was 330, 270, 150 mg/m2/2-week for pts with 0, 1, 2 DA based on pharmacokinetic analysis. [Table: see text]

UGT1A1 genotype-based dose modification of irinotecan regimens: Impact on hematologic toxicities

13500 Background: Patients homozygous for the UGTA1A*28 allele are at risk for severe neutropenia (ANC) with irinotecan (IRI) chemotherapy. Among these pts, grade 4 ANC rates were 50% with IRI [Innocenti, JCO 2004] and 18.2% -54.5% with IFL & IROX in the N9741 study [Mcleod, ASCO 2006]. Higher doses of IRI are associated with a higher toxicity[Hoskins, ASCO 2007]. The FDA mandated a change in the IRI package insert regarding genotype-based dose modifications but the impact of these recommendations has not been prospectively validated. Methods: We conducted a prospective registry of pts treated with IRI. After informed consent, blood was collected to determine UGT1A1 genotype. For pts with UGT1A1 *28/*28, IRI doses were reduced to minimize toxicity. Toxicities were monitored during the first 2 months of therapy. Results: The genotype frequencies of UGT1A1 polymorphism among 78 pts (M: F = 38:40) were 44.9% (35/78)*1/*1, 43.5% (34/78)*1/*28 and 10.2% (8/78) *28/*28. 88% had colorectal cancer. All pts had prior chemotherapy, (47% had ≥2). 60 pts treated were dosed according to guidelines. Dosing schedules were as follows: weekly (10), bi-weekly (48), q 3 weeks (2). For the 54 patients with complete records, the rate of all grade, gr 3 and gr 4–5 ANC are indicated in the table below. Only 5 of 8 patients with *28/*28 received IRI. All were treated with reduced starting doses (Dose - 1) and none developed grade 4–5 ANC. Three heavily pretreated pts (1 *1/*1; 2 *1/*28) developed Gr 4 ANC after only 1 dose of IRI 180 mg/m2, 2 died of neutropenic toxicities. Conclusions: Prospective genotyping to guide dosing decisions has a positive effect in reducing the incidence and severity of toxicities. This study is ongoing and will be expanded to other institutions to further validate these results. UGT1A1 n % all Gr ANC % Gr 3 ANC %Gr 4–5 ANC *1/*1 27 40.7% (11) 7.4% (2) 3% (1) *1/*28 22 59.1% (13) 13.6% (3) 9% (2) *28/*28 5 60% (3) 40% (2) 0 Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Third Wave Technologies Pfizer, Third Wave Technologies Third Wave Technologies Pfizer

A genotype-directed phase I study of irinotecan in advanced cancer patients

2502 Background: The risk of severe toxicity of irinotecan is in part related to the presence of common polymorphisms in UGT1A1, which codes for the enzyme responsible for the inactivation of SN-38, the active metabolite of irinotecan. The most common polymorphism in non-Asian populations is an extra TA repeat in the promoter (7 vs. 6), which reduces the enzyme expression. The three most common genotypes are: 6/6 (40–50%), 6/7 (40–50%), and 7/7 (10–15%). The 7/7 genotype confers the highest risk of toxicity, and it has been suggested that dose reduction is appropriate for patients with this genotype. As one published study has suggested that doses up to 500 mg/m2 q3w can be tolerated by some patients, we hypothesized that higher doses can be safely administered based on genotype by excluding patients with the 7/7 genotype. This study seeks to find the maximum tolerated dose (MTD) in patients with 6/6 and 6/7 genotypes. Methods: Advanced cancer patients (5 SCLC, 4 esophageal, 5 other tumors) have been genotyped for the UGT1A1 polymorphism. 6/6 and 6/7 patients have been treated with 700–850 mg (flat dose, corresponding to about 400–500 mg/m2) of q3w irinotecan. Dose-limiting toxicity (DLT) at cycle 1 is defined as grade =3 nonhematological toxicity, grade 4 neutropenia lasting more than 4 days, and febrile neutropenia. Results: Three 6/6 patients tolerated 700 mg (no grade &gt;2 toxicity). Three additional 6/6 patients have been treated at 850 mg, and one DLT (grade 4 neutropenia lasting more than 4 days) was observed; 3 additional 6/6 patients will be enrolled at 850. Six 6/7 patients have been treated at 700 mg, with only one DLT (febrile neutropenia). Two additional 6/7 patients have been treated at 850, and one DLT (febrile neutropenia) has been observed. Conclusions: Higher doses of irinotecan can be safely administered in patients with selected UGT1A1 genotypes. On the basis of these interim results, a phase II trial in metastatic colorectal cancer patients with 6/6 and 6/7 genotypes has been designed to evaluate the efficacy of higher doses of irinotecan. [Table: see text]

Cigarette Smoking and the Personalization of Irinotecan Therapy

Lifelong cigarette smokers have a 50% chance of dying prematurely due to their addiction. Nearly one third of these deaths are due to cancer. Smokers who develop cancer are often highly addicted and many continue to smoke after the diagnosis of cancer. Smoking-induced cancers that are treated with irinotecan include colon cancer, lung cancer, and others. Irinotecan is a topoisomerase I inhibitor, with dose-limiting toxicity that includes leukopenia and severe diarrhea. Irinotecan is a prodrug, which is metabolized primarily by carboxylesterase enzymes to an active metabolite, 7-ethyl-10-hydroxycamptothecin (SN-38). Irinotecan is also metabolized by the enzyme CYP3A4, and one of the metabolites via the CYPA4 pathway can be metabolized further by carboxylesterase to SN-38. SN-38 is in turn detoxified by glucuronide conjugation via the enzyme uridine diphosphate glucuronosyltransferase (UGT) 1A1. The disposition kinetics and related safety and efficacy of irinotecan are influenced to a considerable degree by genetic and environmental factors. A variant genotype in the promoter region of the UGT enzyme, UGT1A1*28, is associated with a remarkably reduced rate of SN-38 glucuronidation, and a higher risk of irinotecan toxicity with standard dosing. This variant also causes the Gilbert syndrome. Clinical diagnostic tests for the UGT1A1*28 variant have been developed and promoted as a way to individualize irinotecan dosing and to prevent serious toxicity. Environmental factors such as the use of other drugs and cigarette smoking also influence response to irinotecan. Antiepileptic drugs such as phenytoin and carbamazepine accelerate irinotecan and SN-38 metabolism substantially, resulting in decreased levels of SN-38. As a consequence, higher doses of irinotecan are recommended in patients receiving antiepileptic drugs as part of treatment of malignant glioma. Another major environmental modulator of irinotecan response, cigarettes smoking, has been shown to decrease the risk of hematologic toxicity from irinotecan, as described by van der Bol et al in this issue. The risk of irinotecan-induced neutropenia in this study was 6% for smokers compared with 38% for nonsmokers. There was a trend toward a lesser incidence of delayed-onset diarrhea (6% v 15%), but this difference was not significant. Cigarette smoking is well known to interact with a variety of drugs, both by pharmacokinetic and pharmacodynamic mechanisms. Pharmacokinetic interactions are related primarily to the effects of cigarette smoking to induce many drug-metabolizing enzymes and to inhibit some drug-metabolizing enzymes. Polycyclic aromatic hydrocarbons in cigarette smoke are responsible for the induction of lung and liver enzymes, including CYP1A1 and CYP1A2. Smoking-induced induction of CYP1A2 accounts for well-known interactions of smoking with drugs such as caffeine, clozapine, haloperidol, fluvoxamine, olanzapine, tacrine, and theophylline. Recently, smoking has been shown to accelerate the metabolism of erlotinib, another drug metabolized by CYP1A1 and 1A2, which likely explains the poorer response of smokers compared with nonsmokers with non–small-cell lung cancer. Cigarette smoking also induces the enzyme CYP2E1 and inhibits CYP2A6, effects that may be mediated by nicotine. There are conflicting data on the effects of cigarette smoking on the induction of CYP3A4. Cigarette smoking also induces the glucuronidation of some drugs, including mexiletine, propranolol, and codeine, although interestingly in the context of the present study, not morphine or bilirubin. Cigarette smoking has many other physiologic effects that could affect drug response, including effects on the hematopoietic system, including an increase in neutrophil count. Smoking seems to stimulate the bone marrow by releasing proinflammatory factors from macrophages. A direct effect of cigarette smoking on neutrophil count must be considered as a possible mechanism by which smoking reduces neutropenia due to irinotecan. The article by van der Bol et al explores the mechanism of the smoking effect on irinotecan toxicity. Key observations in smokers include lower levels of irinotecan and SN-38 in the plasma, and no significant change in SN-38 glucuronide levels. As an important control, UGT1A1*28 genotype prevalence was not different in smokers versus nonsmokers, eliminating genetic predisposition as a confounder for differential risk of toxicity. The data in this article suggest that cigarette smoking accelerates the metabolism of irinotecan to SN-38, and based on the higher ratios of SN-38 glucuronide to SN-38, also accelerates the glucuronidation of SN-38. Given that the levels of SN-38 were substantially lower in smokers compared with those in nonsmokers, the effects of smoking to induce SN-38 glucuronidation seem to far exceed the effects of smoking to accelerate the conversion of irinotecan to SN-38. As the authors of this informative study mention, although cigarette smoking may reduce the toxicity of irinotecan, it does so by lowering the levels of the active metabolite SN-38. Lower levels of circulating SN-38 might be expected to reduce the therapeutic JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 25 NUMBER 19 JULY 1 2007

Relationship of Serum Bilirubin to Toxicity in Patients With Metastatic Colorectal Cancer Treated With Single-Agent High-Dose Irinotecan

TO THE EDITOR: In the April 15th issue of the Journal of Clinical Oncology, Meyerhardt et al 1 published a study showing thatmodest elevations of baseline bilirubin are associated with increased grade 3 to 4 neutropenia in patients treatedwithweeklyirinotecanafterhavingexperienceddisease progression or recurrence from fluorouracil-based chemotherapy. The authors base their conclusions on a secondary analysis of a cohort of 291 patients enrolled onto two clinical trials evaluating monotherapy irinotecan: either taking irinotecan once a week at the dose of 125 mg/m 2 for 4 weeks followed by a 2-week rest period (95 patients), or once every 3 weeks at the dose of 350 mg/m 2 (88 patients) or 300 mg/m 2 (108 patients). We investigated these results in our study of 49 firstline patients treated with single-agent irinotecan every 3 weeks at the dose of 350 mg/m 2 at the first cycle, which could be escalated to 500 mg/m 2 for the second and subsequent cycles, depending on toxicity. 2 One patient was not evaluated for initial bilirubin. Three patients had bilirubin values greater than 1.5 mg/dL but less than 1.7 mg/dL and werenotincludedinthisanalysisinordertobecomparable to the selection criteria of the Meyerhardt study. Thus, 45 patients were used for this analysis. ComparedwiththeMeyerhardtstudy,thedistribution of baseline bilirubin in our analysis was similar: 33%, 58%, and9%inthe0to0.4,0.5to0.9,and1to1.5mg/dLgroups, respectively, in our study, compared with 33%, 55%, and 12% in the Meyerhardt study. Concerning toxicity, 60% of patients experienced grade 3 to 4 neutropenia in our study, compared with 30% in both of the clinical trials analyzed by Meyerhardt et al. Although our study concerns first-line patients, this higher level of maximum toxicity was to be expected since higher doses of irinotecan were given for a longer period of time. Grade 3 to 4 neutropenia was observed for53%,58%,and100%patientsinthe0to0.4,0.5to0.9,and 1 to 1.5 mg/dL groups, respectively, in our study, compared with 29%, 31%, and 45% in both of the Meyerhardt studies. Thesignificantdifferenceswerethusobservedmostlyforthe levels of bilirubin greater than 1 mg/dL. Although not statistically significant in our study (P .22), we noticed neverthelessthatallfourpatientswithbilirubinvaluesgreaterthan1 mg/dLexperiencedgrade3to4neutropeniatoxicity.Ifweadd those four patients to the three patients with bilirubin values greater than 1.5 mg/dL, six (86%) of seven patients experienced grade 3 to 4 neutropenia toxicity. Meyerhardt et al did not provide detailed information on the monitoring of bilirubin during chemotherapy, but we could imagine that more recent values of bilirubin may have a greater impact on toxicity than values at pretreatmentlevels.Inourstudy,wewereabletocorrelatebilirubin levels before each cycle with the maximum neutropenia grade observed during the same cycle. Assuming each cycle to be independent from each other, among the 258 cycles, we observed 58 cycles (22%) overall with grade 3 or 4 neutropenia. Grade 3 to 4 neutropenia was thus observed for17%,22%,17%,and50%ofcyclesforpatientswith0to 0.4, 0.5 to 0.9, 1 to 1.5, and greater than 1.5 mg/dL values for bilirubin at the beginning of each cycle, respectively (P .01). The differences were thus mainly observed for the 11 (50%) of 22 grade 3 to 4 neutropenia patients who attained bilirubin levels greater than 1.5 mg/dL. It thus seems that bilirubin should be monitored before each cycle, since patients with bilirubin values greater than 1.5 mg/dL are at a significantly higher risk of grade 3 to 4 neutropenia, and could justify prophylactic granulocyte colony-stimulating factor treatment.