Irinotecan unexpectedly causes severe toxicity of leukopenia or diarrhea. Irinotecan is metabolized to form active SN-38, which is further conjugated and detoxified by UDP-glucuronosyltransferase (UGT) 1A1 enzyme. Genetic polymorphisms of the UGT1A1 would affect an interindividual variation of the toxicity by irinotecan via the alternation of bioavailability of SN-38. In this case-control study,retrospective review of clinical records and determination of UGT1A1 polymorphisms were performed to investigate whether a patient with the variant UGT1A1 genotypes would be at higher risk for severe toxicity by irinotecan. All patients previously received irinotecan against cancer in university hospitals,cancer centers, or large urban hospitals in Japan. We identified 26 patients who experienced severe toxicity and 92 patients who did not. The relationship was studied between the multiple variant genotypes(UGT1A1*28 in the promoter and UGT1A1*6, UGT1A1*27, UGT1A1*29, and UGT1A1*7 in the coding region) and the severe toxicity of grade 4 leukopenia (≤0.9 × 109/liter)and/or grade 3 (watery for 5 days or more) or grade 4 (hemorrhagic or dehydration) diarrhea. Of the 26 patients with the severe toxicity, the genotypes of UGT1A1*28 were homozygous in 4 (15%) and heterozygous in 8 (31%), whereas 3 (3%) homozygous and 10 (11%)heterozygous were found among the 92 patients without the severe toxicity. Multivariate analysis suggested that the genotype either heterozygous or homozygous for UGT1A1*28 would be a significant risk factor for severe toxicity by irinotecan(P < 0.001; odds ratio, 7.23; 95%confidence interval, 2.52–22.3). All 3 patients heterozygous for UGT1A1*27 encountered severe toxicity. No statistical association of UGT1A1*6 with the occurrence of severe toxicity was observed. None had UGT1A1*29 or UGT1A1*7. We suggest that determination of the UGT1A1 genotypes might be clinically useful for predicting severe toxicity by irinotecan in cancer patients. This research warrants a prospective trial to corroborate the usefulness of gene diagnosis of UGT1A1 polymorphisms prior to irinotecan chemotherapy.

Irinotecan3(CPT-11) is a camptothecin analogue with strong antitumor activity through an inhibition of topoisomerase I. Although the drug is now used widely, especially for colorectal and lung cancers (1, 2, 3, 4),patients and oncologists have grave concerns about the dose-limiting toxicity of irinotecan, resulting in leukopenia and/or diarrhea(4, 5, 6). Severe, occasionally fatal, toxicity happens sporadically, even in a better risk patient who participates in well-controlled clinical trials (1, 2, 3, 4). Indeed, during a period of its clinical trials, the deaths of 55 patients of 1245 were attributed to side effects (5, 6). The Ministry of Health and Welfare in Japan has allowed irinotecan to be used at a medical institution that is sufficiently equipped to provide emergency treatment for these adverse reactions and under the supervision of specialists thoroughly experienced in chemotherapy (6). In addition, all patients treated with irinotecan have to be studied and reported during its Post Marketing Surveillance until January 2000, and each patient must be judged appropriate for the administration of the drug using the checklist on registering (6). Now, an innovative way of predicting the toxicity is strongly required.

Irinotecan is metabolized by carboxylesterase to form an active SN-38,which is further conjugated and detoxified by UGT (EC 2.4.1.17) to yield its β-glucuronide (7). The glucuronide is excreted in the small intestine via bile, where bacterial glucuronidase resolves the glucuronide into the former SN-38 and glucuronic acid(8). Interindividual differences in pharmacokinetics of SN-38 are suggested to cause the variation in drug effect (9, 10). On the other hand, there are two UGT enzymes in humans,UGT1 and UGT2, and the UGT1 family consists of one gene along with multiple promoters and the first exons which are spliced to the mutual exon 2 (11). Thus, the substrate specificity of the enzyme depends on the first exon. The UGT1A1 gene is composed of a promoter and the first exon closest to exons 2–5(11, 12). UGT1A1 enzyme, which is primarily responsible for conjugating bilirubin, can glucuronidate drugs (e.g.,ethinylestradiol), xenobiotic compounds (e.g., phenols,anthraquinones, and flavones), and endogenous steroids(13). At present, >30 genetic variations in a promoter region and exons have been known to decrease the enzyme activity,leading to constitutional unconjugated jaundice, Crigler-Najjar or Gilbert’s syndrome (12). Recent in vitroanalyses have revealed that the UGT1A1 isoform would be responsible for the glucuronidation of SN-38 and that the genetic variation would associate with the decreased activity of SN-38 glucuronidation as well as bilirubin (14, 15). Additionally, we have suggested an interindividual difference in the pharmacokinetics of SN-38 and SN-38 glucuronide, depending on the UGT1A1 genotype(16). Thus, we speculated that the variant genotypes would increase the toxicity by irinotecan via excessive accumulation of its active metabolite SN-38.

Genotypes involved in Gilbert’s syndrome rather than Crigler-Najjar syndrome II would be clinically important for explanation of patient-patient variations in the reaction to a drug that is mainly conjugated by UGT1A1. Hyperbilirubinemia in a patient with Gilbert’s syndrome is usually milder than that in Crigler-Najjar syndrome II, and 3–10% of the general population are estimated to have Gilbert’s syndrome (17). Moreover, genotypes found in Gilbert’s syndrome are also noted in seemingly healthy individuals and do not always cause hyperbilirubinemia (18, 19, 20, 21, 22), probably because of nongenetic factors including diet and therapeutic drug use. Thus,cancer patients carrying the genotypes associated with Gilbert’s syndrome may be possible candidates for irinotecan chemotherapy.

This study retrospectively investigated the impact of the genetic polymorphism of UGT1A1 on the likelihood of severe toxicity in patients receiving irinotecan in cancer chemotherapy. The genotype analyses were centered on those associated with Gilbert’s syndrome(Table 1). Two types of variant genotypes have been reported in this syndrome. One is a 2-bp insertion (TA) in the TATA box in the promoter [normal (TA)6TAA],resulting in the sequence (TA)7TAA, UGT1A1*28(12, 18, 19), and the other is a heterozygous (sometimes homozygous) single nucleotide change in the coding region (23), all of which have been reported to reduce UGT1A1 activity (19, 24, 25). Our hypothesis is that a patient with the variant genotypes would be at higher risk for severe hematological toxicity and/or diarrhea because of a relatively increased bioavailability of active unconjugated SN-38, and that some of the unexpected severe toxicity might be explained by the genetic factor. The goal of the present study is to explore a clinical advantage of determining UGT1A1 polymorphisms prior to irinotecan chemotherapy for predicting the toxicity.

Patients and Clinical Information.

The subjects were Japanese cancer patients who had received irinotecan-containing chemotherapy in the participating institutes from July 1994 to June 1999 (median, October 1998). They all gave informed consent in writing between July 1998 and June 1999 (median, January 1999) for their peripheral blood to be used for the research. According to the strict requirement of the Ministry (6), each patient was primarily ensured to have an adequate bone marrow function,as a leukocyte count of 3 × 109/liter or more and a platelet count of 100 × 109/liter or more before the use of irinotecan. In addition, patients who had evidence of active infection, watery diarrhea, paralytic ileus, pulmonary interstitial pneumonia or fibrosis, massive ascites or pleural effusion, apparent jaundice, or anamnesis of hypersensitivity to irinotecan were excluded from the irinotecan use. After the administration of irinotecan, the complete blood count, platelet count, and serum chemistry were assessed at least once a week in accordance with the warnings issued by the Ministry. Other appropriate investigations depended on each assigned physician. Almost all subjects were treated as in-patients, and bilirubin levels were always measured after overnight fasting.

We retrospectively reviewed the clinical records including patient characteristics (age, gender, primary disease and previous treatments,evidence of distant metastasis, Eastern Cooperative Oncology Group performance status, and major complications), dosage, and schedule of irinotecan administration, concurrent use of other drugs or radiotherapy, and observed toxicity after irinotecan infusion. We counted the number of days when patients received granulocyte-colony stimulating factors or loperamide hydrochloride, which is commonly prescribed for irinotecan-induced diarrhea in Japan. Prophylactic uses of granulocyte-colony stimulating factor could not be clearly distinguished from those for neutropenia. Because the dose-limiting toxicity of irinotecan results in leukopenia and diarrhea(4), we defined “severe toxicity” in this research as leukopenia of grade 4 (≤0.9 × 109/liter) and/or diarrhea of grade 3 or worse(grade 3, watery for 5 days or more; grade 4, hemorrhagic or dehydration), classified in accordance with the Japan Society for Cancer Therapy criteria (26). The other toxicity was not included in the analysis because anemia would be influenced by miscellaneous patients’ backgrounds including gastrointestinal lesions or nutritious status, and because simultaneous uses of cisplatin or carboplatin probably result in extremely exacerbated nausea/vomiting or thrombocytopenia, respectively. Serum total bilirubin levels were obtained just prior to irinotecan administration along with the highest of those after initiation of the therapy. The study was approved by the Ethical Committees of Nagoya University School of Medicine and the participating institutes.

Genotyping.

Blood sampling and genetic analyses were performed after irinotecan administration in each patient. Genomic DNA was prepared from whole blood (100–200 μl) using the QIAamp Blood kit (Qiagen, Hilden,Germany). We researched the following variant sequences (Table 1; Ref.12): a two-extra-nucleotide insertion (TA) within the TATA box resulting in the sequence (TA)7TAA (−39 to−53, UGT1A1*28; Refs. 18 and 19);a transition (+211 from the initial site of the transcription, G to A)at codon 71 in exon 1 that changes glycine to arginine (G71R, UGT1A1*6; Refs. 23 and 27); a transversion (+686, C to A) at codon 229 in exon 1 that alters proline to glutamine (P229Q, UGT1A1*27; Ref. 23); a transversion (+1099, C to G) at codon 367 in exon 4 that converts arginine to glycine (R367G, UGT1A1*29; Ref.23); and a transversion (+1456, T to G) at codon 486 in exon 5 that transforms tyrosine into aspartic acid (Y486D, UGT1A1*7; Ref. 27).

UGT1A1*28 was distinguished from the most common allele(UGT1A1*1) by direct sequencing (−147 to +106) of 253–255 bp produced by PCR using the method described previously (18, 20). Cycle sequencing was performed with a dye terminator sequence reaction (ABI Prism DNA Sequencing kit; Perkin-Elmer, Foster City, CA) using an ABI PRISM 310 Genetic Analyzer. The remaining variant sequences were distinguished from UGT1A1*1 by PCR-RFLP assay. For the analysis of exon 1, the first-step PCR amplification of a 923-bp fragment containing the exon 1 was performed in accordance with the reported method (21). Subsequently,for the analysis of UGT1A1*6, the second set of PCR amplifications was carried out using nested primers designed to amplify a 235-bp segment. The mismatched forward and the reverse primer was 5′-CTAGCACCTGACGCCTCGTTGTACATCAGAGCC-3′ (+178 to +210; underlining indicates mismatched site) and 5′-CCATGAGCTCCTTGTTGTGC-3′ (+393 to +412), respectively. The forward primer was designed to introduce a MspI (Takara Shuzo Co.,Ltd., Otsu, Japan) restriction site in UGT1A1*1 (+209 to+212), not in UGT1A1*6. The 1000-fold diluted product of the first PCR was subjected to nested PCR in a volume of 50 μl containing 0.2 mm of each deoxynucleoside triphosphate, 50 mm KCl, 10 mm Tris-HCl (pH 8.3), 1.5 mm MgCl2, 0.5μ m of each primer, and 1.3 unit of Taq polymerase (Takara Shuzo Co., Ltd.). PCR conditions were: 95°C for 5 min followed by 25 cycles of 94°C for 30 s, 60°C for 40 s, and 72°C for 40 s (PCR Thermal Cycler MP; Takara Shuzo Co.,Ltd.). A 1-μl PCR product was digested with 4 units of MspI for 1 h at 37°C. DNA from UGT1A1*1was digested into 203- and 32-bp fragments, DNA from UGT1A1*6 gave an undigested 235-bp fragment, and DNA from the heterozygous genotype gave all three fragments. For the sequence of UGT1A1*27, another set of the second PCR amplifications was performed using hemi-nested primers 5′-AGTACCTGTCTCTGCCCAC-3′ (+485 to+503) and 5′-GTCCCACTCCAATACACAC-3′ (+865 to +867 and intron 1),designed to amplify a 399-bp segment. Two BsrI (New England Biolabs, Inc., Beverly, MA) restriction sites exist in UGT1A1*27 (+552 to +556 and +684 to +688), but only one site(+552 to +556) exists in UGT1A1*1. The set of PCR amplifications was identical with that for MspI RFLP described above. Digestion of PCR products with 2.5 units of BsrI for 1 h at 65°C gave 199-, 132- and 68-bp fragments from UGT1A1*27 or 331- and 68-bp from UGT1A1*1. The heterozygous genotype gave all four fragments.

The sequence of UGT1A1*29 was also identified using a nested PCR-RFLP assay. The first-step PCR amplification encompassing exons 2,3, and 4 was performed according to the reported method with minor modifications (21). The mismatched forward and the reverse primers for the second PCR amplification designed to amplify a 285-bp segment was 5′-TCCTCCCTATTTTGCATCTCAGGTCACCCGATGGCC-3′(intron 3 and +1085 to +1098; underlining indicates mismatched site)and 5′-TGAATGCCATGACCAAA-3′ (intron 4), respectively. The forward primer was designed to introduce a Cfr13I (Takara Shuzo Co.,Ltd.) restriction site in UGT1A1*1 (+1095 to +1099) but not in UGT1A1*29. The PCR reaction mixture was the same as that used in the second PCR examination for UGT1A1*6. A PCR product was digested with Cfr13I enzyme. DNA from UGT1A1*1 was digested into 252- and 33-bp fragments, and DNA from UGT1A1*29 gave an undigested 285-bp fragment. For detection of UGT1A1*7, the PCR amplification for a 579-bp fragment of exon 5 was carried out using the primer described previously (21). The reaction mixture was the same as that used in the second PCR assay for UGT1A1*6. There is a BsrI restriction site in the sequence of UGT1A1*1(+1452 to +1456) but not in UGT1A1*7. After incubation with BsrI enzyme, DNA from UGT1A1*1 was digested into 365- and 214-bp fragments, and DNA from UGT1A1*7 gave an undigested 579-bp fragment.

Restriction fragments were analyzed by 4% agarose gel electrophoresis and ethidium bromide staining. The representative genotyping results of every variant genotype were confirmed by direct sequencing analyses.

Statistical Analysis.

Possible factors analyzed to assess associations with the severe toxicity or the polymorphisms of UGT1A1 were as follows;gender, age, performance status, primary disease, presence of distant metastasis, previous treatments, complications of diabetes or liver diseases, chemotherapy regimens, concurrent radiotherapy, and the intended schedule and dosage for each infusion of irinotecan. The chemotherapy regimens were categorized into three groups: irinotecan alone, irinotecan plus platinum (cisplatin or carboplatin), and irinotecan plus other agents (paclitaxel, docetaxel, etoposide,mitomycin C, or 5-fluorouracil). The correlation or association between potential variables was assessed using χ2 test or Fisher’s exact test for categorical variables, or with Mann-Whitney U test for continuous ones. Possible variables that seemed to be associated with severe toxicity (P < 0.1) were considered for inclusion in an unconditional multiple logistic regression analysis. We did not include the following factors in the multivariate analysis because they were highly dependent on the outcome of chemotherapy: total actual dosage and uses of granulocyte colony-stimulating factor and loperamide hydrochloride. The variables in the final models were chosen using forward and backward stepwise procedures at the significance level of 0.25 and 0.1, respectively. The importance of the genetic polymorphism for occurrence of severe toxicity was verified when controlling for the other variables. We performed these analyses using JMP version 3.0.2 software (SAS Institute, Inc., Cary, NC). A difference was considered statistically significant when the two-tailed P was <0.05.

Toxicity of Irinotecan.

We collected blood samples from the 118 patients with their clinical information (Tables 2 and 3). Nine (8%) and 38 (32%) patients experienced leukopenia of grade 4 (≤0.9 × 109/liter) and grade 3 (1.9–1.0 × 109/liter), respectively. Diarrhea was reported in 3 patients (3%) with grade 4 (hemorrhagic or dehydration)and 19 patients (16%) with grade 3 (watery for 5 days or more). Five of the 9 patients with grade 4 leukopenia also had grade 3/4 diarrhea,and 16 of the 22 patients with grade 3/4 diarrhea encountered grade 3/4 leukopenia. Then, we identified 26 patients who experienced severe toxicity and 92 patients who did not (Tables 2 and 3). During the first course, 19 of the 26 patients (73%) experienced severe toxicity.

Lower total amounts of actual irinotecan (<300 mg/m2) and more frequent use of granulocyte-colony stimulating factor or loperamide hydrochloride were observed in patients suffering from severe toxicity (Table 3). Complete or partial responses were observed more often in patients who experienced severe toxicity, although the difference was not significant (Table 3).

Distribution of Genotypes.

The genotypes were determined in the all 118 patients including 9 patients whose UGT1A1*28 genotyping results have been reported elsewhere (16). The allele frequencies of UGT1A1*28 were 0.308 (95% CI, 0.182–0.433) and 0.087 (95%CI, 0.046–0.128), and those of UGT1A1*6 were 0.077 (95%CI, 0.004–0.149) and 0.136 (95% CI, 0.086–0.185) among the patients with and without severe toxicity, respectively (Table 4). The difference in allelic distribution between the patients with and without severe toxicity was significant for UGT1A1*28 (P < 0.001) but not significant for UGT1A1*6(P > 0.2; GENEPOP version 3.1d software, the Laboratoire de Génétique et Environment, Montpellier,France). The co-occurrence of the polymorphisms was found in 5 patients: 2 patients were heterozygous for both UGT1A1*28and UGT1A1*6, and 3 patients were heterozygous for UGT1A1*27 and homozygous (2 patients) or heterozygous (1 patient) for UGT1A1*28. We did not examine the cis or trans arrangement of the variant sequences in these 5 patients. None of the patients had UGT1A1*29 or UGT1A1*7.

The total bilirubin levels prior to and the highest during the chemotherapy were obtained in 117 patients; the one patient missing the measurement had no apparent jaundice and was homozygous for UGT1A1*1. The 2 patients heterozygous for both UGT1A1*28 and UGT1A1*6 had bilirubin levels within the normal range: 13.9 (μmol/l) and 15.4 prior to therapy and 10.3 and 15.4 after initiation of therapy, respectively. Except for these 2 patients, the differences in the bilirubin levels among the genotypes were statistically significant prior to the therapy(P = 0.031, Kruskal-Wallis test) and after the initiation of therapy (P < 0.001; Table 5). There was no significant association between the genotypes and objective responses (data not shown).

Genotypes and Toxicity.

Simple logistic regression analysis showed that the genotype either heterozygous or homozygous for UGT1A1*28 proved to be a significant predictor of severe toxicity (odds ratio, 5.21; 95% CI,1.98–13.96; P < 0.001; Table 4). Conversely, no statistical association of UGT1A1*6 with the occurrence of severe toxicity was observed (odds ratio, 0.55; 95% CI,0.15–1.61; P > 0.2).

Besides the variant genotypes, the factors that seemed to affect severe toxicity adversely (P < 0.1) were gender,chemotherapy regimen, and intended schedule of irinotecan infusion(Tables 2 and 3). These factors were assessed for correlation or association. Significant association was found between chemotherapy regimen and intended schedule (P < 0.001,χ2 test); in other words, 12 of 19 patients(63%) treated with irinotecan of 3- or 4-week cycle had received additional anticancer drugs. Because the chemotherapy regimen was the variable with stronger relationship with severe toxicity, we considered the factor of chemotherapy regimen for inclusion in the model. The other correlation or association among chemotherapy regimen, gender,and UGT1A1*28 genotype was not significant. The stepwise procedures identified female gender and use of other anticancer drugs(apart from platinum) as important variables for the occurrence of severe toxicity besides the UGT1A1*28 genotype (Table 6). After adjustment with these two variables, the importance of the UGT1A1*28 genotype was verified (Table 6).

Among the 5 patients who had both grade 4 leukopenia and grade 3 or worse diarrhea concurrently, 2 had both UGT1A1*28 and UGT1A1*27, 2 were heterozygous for UGT1A1*6, and one had none of the variant genotypes analyzed (homozygous for UGT1A1*1). On the other hand, it is noteworthy that 4 of 5 patients (80%) who had the variant sequences both in the promoter(UGT1A1*28) and in exon 1 (UGT1A1*6 or UGT1A1*27) suffered from life-threatening toxicity. There were 3 patients who did not encounter severe toxicity among the 7 patients homozygous for UGT1A1*28. One of the 3 patients received chronically ursodesoxycholic acid and trepibutone for the treatment of cholelithiasis and, additionally, rifampin and isoniazid for 2 weeks, 1 month before the irinotecan infusion. Another ceased chemotherapy because of vomiting of blood caused by gastric ulcer after one-time infusion of irinotecan. The 2 patients homozygous for UGT1A1*6 could be treated without severe toxicity, and all 3 patients heterozygous for UGT1A1*27 experienced severe toxicity (Table 4).

There was a significant increase in the bilirubin levels after irinotecan infusion in both the patients who did(P < 0.001, Wilcoxon signed-rank test) and did not (P < 0.001; Table 4) encounter the severe toxicity. The increase in bilirubin levels after the initiation of therapy tended to be worse in the patients who experienced severe toxicity than in those who did not (P = 0.071, Mann-Whitney U test; Table 4).

The variant genotype in the promoter region, UGT1A1*28,was significantly related to the severe toxicity induced by irinotecan,whereas with UGT1A1*6 in exon 1, it was not. The multivariate analysis suggested that the patients who have UGT1A1*28 would be seven times as likely to encounter severe toxicity from irinotecan than those who do not have it (Table 6). Although the use of other anticancer drugs also significantly affected severe toxicity, it was not beyond the UGT1A1*28 genotype(Tables 2 and 6). The effects of female gender on severe toxicity did not reach significant levels in the current analysis (Tables 2 and 6). These findings clarify the clinical importance of UGT1A1*28for UGT1A1 conjugation activity, especially in acute exposure to irinotecan.

We should mention that several biases might modify the distributions of the UGT1A1 polymorphisms in this study:(a) patients with high bilirubin levels would usually be precluded from irinotecan treatment because of a suspicion of liver dysfunction that may cause severe toxicity; (b) if our hypothesis is correct, potential patients who have the variant genotypes might die of fatal toxicity by irinotecan and inevitably be excluded from the analysis. This speculation is compatible with the fact that no patients had either UGT1A1*29 or UGT1A1*7, which occurs in common exons of UGT1A isoforms,resulting in a substantial reduction of their functional activities. Conversely, the patients who experienced severe toxicity from irinotecan would be more inclined to participate in this research than those who did not. Nevertheless, we consider that the patients analyzed here could approximate a population of Japanese cancer patients,because the incidence of severe toxicity in the patients of this research was comparable with those in the previous Phase II or III trials of irinotecan chemotherapy in Japan (2, 3). Thus,this retrospective research warrants a prospective trial to corroborate the usefulness of gene diagnosis of UGT1A1 polymorphisms prior to irinotecan chemotherapy.

We should be careful to understand the exact clinical importance of UGT1A1*6 for toxicity by irinotecan. According to in vitro expression studies, UGT1A1*6 in the homozygous and heterozygous genetic states decreases the enzyme activity to 32 and 60% of control, respectively (25). In addition, UGT1A1*6 has been reported to be a significant risk factor for nonphysiological hyperbilirubinemia among Japanese neonates(21, 22). The genetic effect of UGT1A1*6 might be somehow masked in the current retrospective research; otherwise, the reduced enzyme activity might be compensated in adults by acquired factors. Particularly, UGT1A1*6 as well as UGT1A1*27 might considerably affect the susceptibility to irinotecan when they coexist with a variant sequence in the promoter UGT1A1*28.

The patients with variant UGT1A1 genotypes were not always those who encountered the severe toxicity by irinotecan, and vice versa. Generally, pharmacogenetic variations definitely alter the relevant drug effects, as observed in a poor metabolizer of thiopurine S-methyltransferase or dihydropyrimidine dehydrogenase(28, 29). In this research, the genetic effect of the genotypes on toxicity by irinotecan would be relatively weak, because they are originally responsible for Gilbert’s syndrome that shows mild hyperbilirubinemia compared with Crigler-Najjar syndrome. Moreover, the bioavailability of SN-38 depends on the capacity of not only UGT1A1 but also carboxylesterase that metabolically transforms irinotecan to SN-38. However, determining the UGT1A1 genotype would be clinically important for Japanese patients because >20% of them have the variant genotypes and possibly have an increased risk of severe toxicity. Besides UGT1A1 polymorphisms, a recent report suggested that the UGT1A7 isoform would glucuronidate SN-38 more than UGT1A1 (30). However, because UGT1A7 is absent in human liver (31), a primary organ for detoxifying i.v. irinotecan, the evidence of SN-38 glucuronidation by UGT1A7 does not deny the role of UGT1A1. In fact, the significant increase in bilirubin levels after irinotecan infusion (Tables 4 and 5) clearly supported that the glucuronidation of SN-38 in the liver should competitively inhibit that of bilirubin, which is a major substrate of UGT1A1. Because UGT1A7 is expressed in gastrointestinal tissue, it might be important especially for the efficacy against colon cancer and the impact on diarrhea by irinotecan. Although genetic polymorphism of UGT1A7 and their relationships with the phenotypic activity have not yet been identified, more precise estimation of the clinical effects by irinotecan might be possible by investigating the genetic variations, if any.

The bilirubin level seems an inadequate parameter to predict severe toxicity by irinotecan. At the suggestion of other investigators(32), we also observed that the bilirubin levels were increased more in the patients who experienced severe toxicity than in those who did not, although the difference was not significant (Table 4). Indeed, the differences in bilirubin levels among the genotypes were statistically significant but seemed clinically negligible as a tool to predict toxicity (Table 5). Furthermore, UGT1A1*28appeared to be important for Gilbert’s syndrome but not sufficient for the complete manifestation of the syndrome (19). The clinical usefulness of the bilirubin level might be improved if patients abstain from drug and alcohol use and are strictly fasted, but it does not seem to be practical.

Inter-ethnic differences can be easily predicted in metabolic profiles and clinical effects of irinotecan, although racial differences in tolerability and clinical outcomes of irinotecan treatment have not been investigated directly. Great differences in the distributions of the UGT1A1 polymorphisms between Caucasians and Japanese populations have been reported; the frequency of UGT1A1*28in Caucasians is higher than among Japanese (20, 24). This implies that Caucasians might be more susceptible to the drug than the Japanese. On the contrary, UGT1A1*6 and UGT1A1*27, the variant sequences in exon 1, have been identified only in the Japanese (19, 21, 22). Although the clinical significance for irinotecan chemotherapy of these genotypes in exon 1 remains uncertain, they might cause Japanese to be more sensitive to the drug than Caucasians. These findings suggest racial differences in the importance of UGT1A1 genotypes in irinotecan toxicity.

Individualization of drug dosage is critical for cancer chemotherapy to reduce unnecessary toxicity and to improve its therapeutic efficacy because the therapeutic index is often narrow. Oncologists traditionally used to predict toxicity by drugs and to optimize the dosage based on the patient’s physiological factors (e.g.,body surface area, age), pathological conditions (e.g.,performance status, organ functions), and clinical history(e.g., previous treatments). Recently, pharmacokinetic and pharmacodynamic analyses in chemotherapy provide more objective information for predicting the clinical effects of drugs. Furthermore,we believe that pharmacogenetic analyses would be an another clue for individualized chemotherapy. If there is the recognized difference in drug disposition and sensitivity caused by the polymorphic drug-metabolizing enzyme, the optimal dosage required for response with the least toxicity would be different in patients with the different genotypes. In the present study, the determination of the UGT1A1 genotypes for irinotecan treatment was suggested to be clinically useful. We are planning a dose escalation study of irinotecan in cancer patients who have been determined to have the variant genotypes.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by a Research Fellowship of the Japan Society for the Promotion of Science for Young Scientists(to Y. A.) and in part by a Grant-in-Aid for Cancer Research from the Ministry of Health and Welfare of Japan.

3

The abbreviations used are: irinotecan,7-ethyl-10-[4-(1-piperidino)-1-piperidino]carbonyloxycamptothecin;SN-38, 7-ethyl-10-hydroxycamptothecin; UGT,UDP-glucuronosyltransferase; CI, confidence interval; RFLP, restriction fragment length polymorphism.

Table 1

Variant UGT1A1 alleles analyzed in this studya

AlleleNucleotide changeEffect on proteinExon
UGT1A1*28 (TA)7TAA Reduced expression Promoter 
UGT1A1*211G→A G71R 
UGT1A1*27 686C→A P229Q 
UGT1A1*29 1099C→G R367G 
UGT1A1*1456T→G Y486D 
AlleleNucleotide changeEffect on proteinExon
UGT1A1*28 (TA)7TAA Reduced expression Promoter 
UGT1A1*211G→A G71R 
UGT1A1*27 686C→A P229Q 
UGT1A1*29 1099C→G R367G 
UGT1A1*1456T→G Y486D 

a The most common (wild-type) UGT1A1 allele is regarded as UGT1A1

*

1.

Table 2

Baseline characteristics of patients

Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
Gender (men/women) 14/12 66/26 0.085b 
Median age (range, yr) 60 (38–76) 61 (41–75) >0.2c 
Performance status   >0.2b 
8 (31%) 31 (34%)  
15 (58%) 51 (55%)  
≥2 3 (12%) 10 (11%)  
Primary disease   >0.2b 
Small cell lung 4 (15%) 17 (18%)  
Non-small cell lung 16 (62%) 49 (53%)  
Colorectal 3 (12%) 18 (20%)  
Other 3 (12%) 8 (9%)  
Distant metastases 21 (81%) 68 (74%) >0.2b 
Previous treatment   >0.2b 
None 12 (46%) 36 (39%)  
Systemic chemotherapy 12 (46%) 47 (51%)  
Surgery 8 (31%) 34 (37%)  
Radiotherapy 3 (12%) 17 (18%)  
Complications   >0.2b 
Diabetes 2 (8%) 8 (9%)  
Liver diseases 3 (12%) 6 (7%)  
Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
Gender (men/women) 14/12 66/26 0.085b 
Median age (range, yr) 60 (38–76) 61 (41–75) >0.2c 
Performance status   >0.2b 
8 (31%) 31 (34%)  
15 (58%) 51 (55%)  
≥2 3 (12%) 10 (11%)  
Primary disease   >0.2b 
Small cell lung 4 (15%) 17 (18%)  
Non-small cell lung 16 (62%) 49 (53%)  
Colorectal 3 (12%) 18 (20%)  
Other 3 (12%) 8 (9%)  
Distant metastases 21 (81%) 68 (74%) >0.2b 
Previous treatment   >0.2b 
None 12 (46%) 36 (39%)  
Systemic chemotherapy 12 (46%) 47 (51%)  
Surgery 8 (31%) 34 (37%)  
Radiotherapy 3 (12%) 17 (18%)  
Complications   >0.2b 
Diabetes 2 (8%) 8 (9%)  
Liver diseases 3 (12%) 6 (7%)  
a

Japan Society for Cancer Therapy criteria.

b

χ2 test.

c

Mann-Whitney U test.

Table 3

Information on irinotecan chemotherapy

Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
Regimens   0.015b 
Irinotecan alone 3 (12%) 32 (35%)  
Irinotecan and platinum 13 (50%) 45 (49%)  
Irinotecan and other anticancer drugs 10 (38%) 15 (16%)  
Concurrent radiotherapy 1 (4%) 8 (9%) >0.2c 
Intended schedule   0.059b 
Weekly (days 1, 8, and 15) 15 (72%) 62 (67%)  
Every 3 or 4 weeks 8 (31%) 11 (12%)  
Twice every 4 weeks 3 (12%) 19 (21%)  
Intended irinotecan dosage for each infusion (mg/m2  >0.2b 
<60 9 (35%) 18 (20%)  
60 8 (31%) 34 (37%)  
>60 9 (35%) 40 (43%)  
Total actual dosage (mg/m2  0.010b 
<300 15 (58%) 24 (26%)  
301–600 7 (27%) 46 (50%)  
>600 4 (15%) 22 (24%)  
Use of granulocyte colony- stimulating factor (days)   <0.001b 
5 (19%) 61 (66%)  
1–14 11 (42%) 13 (14%)  
≥15 10 (38%) 18 (26%)  
Use of loperamide hydrochloride (days)   0.002b 
4 (15%) 51 (55%)  
1–7 15 (58%) 28 (30%)  
≥8 7 (27%) 13 (14%)  
Objective response   >0.2b 
Complete or partial 11 (48%) 31 (39%)  
Stable disease 8 (35%) 36 (45%)  
Progressive disease 4 (17%) 13 (16%)  
Not measurable 12  
Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
Regimens   0.015b 
Irinotecan alone 3 (12%) 32 (35%)  
Irinotecan and platinum 13 (50%) 45 (49%)  
Irinotecan and other anticancer drugs 10 (38%) 15 (16%)  
Concurrent radiotherapy 1 (4%) 8 (9%) >0.2c 
Intended schedule   0.059b 
Weekly (days 1, 8, and 15) 15 (72%) 62 (67%)  
Every 3 or 4 weeks 8 (31%) 11 (12%)  
Twice every 4 weeks 3 (12%) 19 (21%)  
Intended irinotecan dosage for each infusion (mg/m2  >0.2b 
<60 9 (35%) 18 (20%)  
60 8 (31%) 34 (37%)  
>60 9 (35%) 40 (43%)  
Total actual dosage (mg/m2  0.010b 
<300 15 (58%) 24 (26%)  
301–600 7 (27%) 46 (50%)  
>600 4 (15%) 22 (24%)  
Use of granulocyte colony- stimulating factor (days)   <0.001b 
5 (19%) 61 (66%)  
1–14 11 (42%) 13 (14%)  
≥15 10 (38%) 18 (26%)  
Use of loperamide hydrochloride (days)   0.002b 
4 (15%) 51 (55%)  
1–7 15 (58%) 28 (30%)  
≥8 7 (27%) 13 (14%)  
Objective response   >0.2b 
Complete or partial 11 (48%) 31 (39%)  
Stable disease 8 (35%) 36 (45%)  
Progressive disease 4 (17%) 13 (16%)  
Not measurable 12  
a

Japan Society for Cancer Therapy criteria.

b

χ2 test.

c

Fisher’s exact test.

Table 4

Associations of UGT1A1 genotypes and bilirubin levels with severe toxicity

Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
UGT1A1*28              b   <0.001c 
−/− 14 (54%) 79 (86%)  
+/− 8 (31%) 10 (11%)  
+/+ 4 (15%) 3 (3%)  
UGT1A1*6              b   >0.2c 
−/− 22 (85%) 69 (75%)  
+/− 4 (15%) 21 (23%)  
+/+ 0 (0%) 2 (2%)  
Total bilirubin levels (μmol/l)    
Prior to therapy 8.6 (6.8–13.7)d 8.6 (6.8–12.0)e >0.2f 
Highest after infusion 16.2 (11.8–26.5) 13.7 (10.3–18.8)e 0.071f 
Leukopenia (grade 4) and/or diarrhea (grade 3 or worse)a
Experienced (n = 26)Not experienced (n = 92)P
UGT1A1*28              b   <0.001c 
−/− 14 (54%) 79 (86%)  
+/− 8 (31%) 10 (11%)  
+/+ 4 (15%) 3 (3%)  
UGT1A1*6              b   >0.2c 
−/− 22 (85%) 69 (75%)  
+/− 4 (15%) 21 (23%)  
+/+ 0 (0%) 2 (2%)  
Total bilirubin levels (μmol/l)    
Prior to therapy 8.6 (6.8–13.7)d 8.6 (6.8–12.0)e >0.2f 
Highest after infusion 16.2 (11.8–26.5) 13.7 (10.3–18.8)e 0.071f 
a

Japan Society for Cancer Therapy criteria.

b

Symbols of (−/−), (+/−), and(+/+) denote homozygous absence of the variant allele, heterozygous,and homozygous for the variant allele, respectively.

c

GENEPOP version 3.1d software, the Laboratoire de Génétique et Environment, Montpellier,France.

d

Median (interquantile range).

e

One patient missed the bilirubin measurement during the treatment.

f

Mann-Whitney U test.

Table 5

Association of UGT1A1 genotypes with total bilirubin levels in 115 patients

GenotypesanTotal bilirubin levels (μmol/l)b
TATA box UGT1A1*28Codon 71 UGT1A1*6Codon 229 UGT1A1*27Prior to therapyHighest after infusion
−/− −/− −/− 67 8.6 (6.8–12.0) 13.7 (10.3–17.1) 
−/− +/−, +/+c −/− 25 10.3 (8.6–13.7) 15.4 (11.1–25.7) 
+/− −/− −/−, +/−d 16 8.6 (6.8–10.3) 18.0 (12.0–23.1) 
+/+ −/− −/−, +/−e 12.0 (6.8–20.5) 34.2 (22.2–42.8) 
GenotypesanTotal bilirubin levels (μmol/l)b
TATA box UGT1A1*28Codon 71 UGT1A1*6Codon 229 UGT1A1*27Prior to therapyHighest after infusion
−/− −/− −/− 67 8.6 (6.8–12.0) 13.7 (10.3–17.1) 
−/− +/−, +/+c −/− 25 10.3 (8.6–13.7) 15.4 (11.1–25.7) 
+/− −/− −/−, +/−d 16 8.6 (6.8–10.3) 18.0 (12.0–23.1) 
+/+ −/− −/−, +/−e 12.0 (6.8–20.5) 34.2 (22.2–42.8) 
a

Symbols of (−/−), (+/−),and (+/+) denote homozygous absence of the variant allele,heterozygous, and homozygous for the variant allele, respectively.

b

Median (interquantile range).

c

Two patients homozygous for UGT1A1

*

6 had bilirubin levels of 23.9 and 11.8 prior to therapy and 18.8 and 23.9 following initiation of therapy,respectively.

d

One patient who had both UGT1A1

*

28 and >UGT1A1

*

27 had bilirubin levels of 13.7 prior to therapy and 17.1 after initiation of therapy,respectively.

e

Two patients homozygous for UGT1A1

*

28 and heterozygous for UGT1A1

*

27 had bilirubin levels of 10.3 and 44.5 prior to therapy and 47.9 and 42.8 after initiation of therapy, respectively.

Table 6

Multiple logistic regression analysis

TermβaSEχ2POdds ratio (95% CI)
Intercept 0.763 0.591    
UGT1A1*28 1.979 0.550 12.95 0.0003 7.23 (2.52–22.3) 
Regimenb 1.510 0.557 7.36 0.0067 4.52 (1.53–13.9) 
Female 0.849 0.508 3.10 0.0782 2.45 (0.90–6.75) 
TermβaSEχ2POdds ratio (95% CI)
Intercept 0.763 0.591    
UGT1A1*28 1.979 0.550 12.95 0.0003 7.23 (2.52–22.3) 
Regimenb 1.510 0.557 7.36 0.0067 4.52 (1.53–13.9) 
Female 0.849 0.508 3.10 0.0782 2.45 (0.90–6.75) 
a

β, coefficient.

b

Regimen of irinotecan plus other anticancer drugs apart from platinum.

We thank the following for their participation: Dr. Hiroshi Saito (Aichi Hospital), Dr. Kiyoshi Mori (Tochigi Cancer Center), Drs. Gyo Asai and Takefumi Komiya (Kinki University), Dr. Nagio Takigawa(National Shikoku Cancer Center), and Drs. Naohiko Kuno and Masahiko Ando (Japanese Red Cross Nagoya First Hospital). We also thank Drs. Toshimichi Yamamoto, Koji Kakihata, and Hidehiko Saito for statistical advice or critical review of the manuscript.

1
Rougier P., Van Cutsem E., Bajetta E., Niederle N., Possinger K., Labianca R., Navarro M., Morant R., Bleiberg H., Wils J., Awad L., Herait P., Jacques C. Randomised trial of irinotecan versus fluorouracil by continuous infusion after fluorouracil failure in patients with metastatic colorectal cancer.
Lancet
,
352
:
1407
-1412,  
1998
.
2
Kudoh S., Fujiwara Y., Takada Y., Yamamoto H., Kinoshita A., Ariyoshi Y., Furuse K., Fukuoka M. for the West Japan Lung Cancer Grou.
p. Phase II study of irinotecan combined with cisplatin in patients with previously untreated small-cell lung cancer. J. Clin. Oncol.
,
16
:
1068
-1674,  
1998
.
3
Masuda, N., Fukuoka, M., Negoro, S., Takada, Y., Sugiura, T., Ohashi, Y., Ariyoshi, Y., Niitani, H., and The CPT-11 Lung Cancer Study Group West. Randomized trial comparing cisplatin (CDDP) and irinotecan (CPT-11) versus CDDP and vindesine (VDS) versus CPT-11 alone in advanced non-small cell lung cancer (NSCLC), a multicenter phase III study. Proc. Am. Soc. Clin. Oncol., 18: 459a, 1999.
4
Negoro S., Fukuoka M., Masuda N., Takada M., Kusunoki Y., Matsui K., Takifuji N., Kudoh S., Niitani H., Taguchi T. Phase I study of weekly intravenous infusion of CPT-11, a new derivative of camptothecin, in the treatment of advanced non-small-cell lung cancer.
J. Natl. Cancer Inst.
,
83
:
1164
-1168,  
1991
.
5
Akabayashi A. Questions raised over release of side-effects data in Japan.
Lancet
,
350
:
124
1997
.
6
Pharmaceuticals and Cosmetics Division, Pharmaceutical Affairs Bureau, Ministry of Health, and Welfare. Summary Basis of Approval (SBA) No. 1 (revised edition): irinotecan hydrochloride. Tokyo: Yakuji Nippo, Ltd., 1996.
7
Kawato Y., Aonuma M., Hirota Y., Kuga H., Sato K. Intracellular roles of SN-38, a metabolite of the camptothecin derivative CPT-11, in the antitumor effect of CPT-11.
Cancer Res.
,
51
:
4187
-4191,  
1991
.
8
Takasuna K., Hagiwara T., Hirohashi M., Kato M., Nomura M., Nagai E., Yokoi T., Kamataki T. Involvement of β-glucuronidase in intestinal microflora in the intestinal toxicity of the antitumor camptothecin derivative irinotecan hydrochloride (CPT-11) in rats.
Cancer Res.
,
56
:
3752
-3577,  
1996
.
9
Gupta E., Lestingi T. M., Mick R., Ramirez J., Vokes E. E., Ratain M. J. Metabolic fate of irinotecan in humans: correlation of glucuronidation with diarrhea.
Cancer Res.
,
54
:
3723
-3725,  
1994
.
10
Kudoh S., Fukuoka M., Masuda N., Yoshikawa A., Kusunoki Y., Matsui K., Negoro S., Takifuji N., Nakagawa K., Hirashima T., Yana T., Takada M. Relationship between the pharmacokinetics of irinotecan and diarrhea during combination chemotherapy with cisplatin.
Jpn. J. Cancer Res.
,
86
:
406
-413,  
1995
.
11
Ritter J. K., Chen F., Sheen Y. Y., Tran H. M., Kimura S., Yeatman M. T., Owens I. S. A novel complex locus UGT1 encodes human bilirubin, phenol, and other UDP-glucuronosyltransferase isozymes with identical carboxyl termini.
J. Biol. Chem.
,
267
:
3257
-3261,  
1992
.
12
Mackenzie P. I., Owens I. S., Burchell B., Bock K. W., Bairoch A., Bélanger A., Fournel-Gigleux S., Green M., Hum D. W., Iyanagi T., Lancet D., Louisot P., Magdalou J., Roy Chowdhury J., Ritter J. K., Schachter H., Tephly T. R., Tipton K. E., Nebert D. W. The UDP glycosyltransferase gene superfamily: recommended nomenclature update based on evolutionary divergence.
Pharmacogenetics
,
7
:
255
-269,  
1997
.
13
Senafi S. B., Clarke D. J., Burchell B. Investigation of the substrate specificity of a cloned expressed human bilirubin UDP-glucuronosyltransferase: UDP-sugar specificity and involvement in steroid and xenobiotic glucuronidation.
Biochem. J.
,
303
:
233
-240,  
1994
.
14
Iyer L., King C. D., Whitington P. F., Green M. D., Roy S. K., Tephly T. R., Coffman B. L., Ratain M. J. 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. Investig.
,
101
:
847
-854,  
1998
.
15
Iyer L., Hall D., Das S., Mortell M. A., Ramírez J., Kim S., Di Rienzo A., Ratain M. J. Phenotype-genotype correlation of in vitro SN-38 (active metabolite of irinotecan) and bilirubin glucuronidation in human liver tissue with UGT1A1 promoter polymorphism.
Clin. Pharmacol. Ther.
,
65
:
576
-582,  
1999
.
16
Ando Y., Saka H., Asai G., Sugiura S., Shimokata K., Kamataki T. UGT1A1 genotypes and glucuronidation of SN-38, the active metabolite of irinotecan.
Ann. Oncol.
,
9
:
845
-847,  
1998
.
17
Isselbacher K. J. Bilirubin metabolism and hyperbilirubinemia Ed. 14 Fauci A. S. Braunwald E. Isselbacher K. J. Wilson J. D. Martin J. B. Kasper D. L. Hauser S. L. Long D. L. eds. .
Harrision’s Principles of Internal Medicine
,
:
1672
-1677, McGraw-Hill New York  
1997
.
18
Monaghan G., Ryan M., Seddon R., Hume R., Burchell B. Genetic variation in bilirubin UDP-glucuronosyltransferase gene promoter and Gilbert’s syndrome.
Lancet
,
347
:
578
-581,  
1996
.
19
Bosma P. J., Roy Chowdhury J., Bakker C., Gantla S., de Boer A., Oostra B. A., Lindhout D., Tytgat G. N. J., Jansen P. L. M., Oude Elferink R. P. J., Roy Chowdhury N. The genetic basis of the reduced expression of bilirubin UDP-glucuronosyltransferase 1 in Gilbert’s syndrome.
N. Engl. J. Med.
,
333
:
1171
-1175,  
1995
.
20
Ando Y., Chida M., Nakayama K., Saka H., Kamataki T. The UGT1A1*28 allele is relatively rare in a Japanese population.
Pharmacogenetics
,
8
:
357
-360,  
1998
.
21
Akaba K., Kimura T., Sasaki A., Tanabe S., Ikegami T., Hashimoto M., Umeda H., Yoshida H., Umetsu K., Chiba H., Yuasa I., Hayasaka K. Neonatal hyperbilirubinemia and mutation of the bilirubin uridine diphosphate-glucuronosyltransferase gene: a common missense mutation among Japanese, Koreans and Chinese.
Biochem. Mol. Biol. Int.
,
46
:
21
-26,  
1998
.
22
Maruo Y., Nishizawa K., Sato H., Doida Y., Shimada M. Association of neonatal hyperbilirubinemia with bilirubin UDP-glucuronosyltransferase polymorphism.
Pediatrics
,
103
:
1224
-1227,  
1999
.
23
Aono S., Adachi Y., Uyama E., Yamada Y., Keino H., Nanno T., Koiwai O., Sato H. Analysis of genes for bilirubin UDP-glucuronosyltransferase in Gilbert’s syndrome.
Lancet
,
345
:
958
-959,  
1995
.
24
Beutler E., Gelbart T., Demina A. Racial variability in the UDP-glucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for regulation of bilirubin metabolism?.
Proc. Natl. Acad. Sci. USA
,
95
:
8170
-8174,  
1998
.
25
Yamamoto K., Sato H., Fujiyama Y., Doida Y., Bamba T. Contribution of two missense mutations (G71R and Y486D) of the bilirubin UDP-glycosyltransferase (UGT1A1) gene to phenotypes of Gilbert’s syndrome and Crigler-Najjar syndrome type II.
Biochem. Biophys. Acta
,
1406
:
267
-273,  
1998
.
26
Japan Society for Cancer Therapy. Criteria for the evaluation of the clinical effects of solid cancer chemotherapy.
J. Jpn. Soc. Cancer Ther.
,
28
:
101
-130,  
1993
.
27
Aono S., Yamada Y., Keino H., Hanada N., Nakagawa T., Sasaoka Y., Yazawa T., Sato H., Koiwai O. Identification of defect in the genes for bilirubin UDP-glucuronosyltransferase in a patient with Crigler-Najjar syndrome type II.
Biochem. Biophys. Res. Commun.
,
197
:
1239
-1244,  
1993
.
28
Wei X., McLeod H. L., McMurrough J., Gonzalez F. J., Fernandez-Salguero P. Molecular basis of the human dihydropyrimidine dehydrogenase deficiency and 5-fluorouracil toxicity.
J. Clin. Investig.
,
98
:
610
-615,  
1996
.
29
Krynetski E. Y., Tai H. L., Yates C. R., Fessing M. Y., Loennechen T., Schuetz J. D., Relling M. V., Evans W. E. Genetic polymorphism of thiopurine S-methyltransferase: clinical importance and molecular mechanisms.
Pharmacogenetics
,
6
:
279
-290,  
1996
.
30
Ciotti M., Basu N., Brangi M., Owens I. S. Glucuronidation of 7-ethyl-10-hydroxycamptothecin (SN-38) by the human UDP-glucuronosyltransferases encoded at the UGT1 locus.
Biochem. Biophys. Res. Commun.
,
260
:
199
-202,  
1999
.
31
Strassberg C. P., Oldhafer K., Manns M. P., Tukey R. H. Differential expression of the UGT1A locus in human liver, biliary, and gastric tissue: identification of UGT1A7 and UGT1A10 transcripts in extrahepatic tissue.
Mol. Pharmacol.
,
52
:
212
-220,  
1997
.
32
Wasserman E., Myara A., Lokiec F., Goldwasser F., Trivin F., Mahjoubi M., Misset J. L., Cvitkovic E. Severe CPT-11 toxicity in patients with Gilbert’s syndrome: two case reports.
Ann. Oncol.
,
8
:
1049
-1051,  
1997
.