Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Association between UGT1A1*28 Polymorphisms and Clinical Outcomes of Irinotecan-Based Chemotherapies in Colorectal Cancer: A Meta-Analysis in Caucasians

  • Xiang Liu,

    Affiliations Ontario Cancer Institute, Toronto, Ontario, Canada, School of Laboratory Medicine, Hubei University of Chinese Medicine, Wuhan, China

    ⨯
  • Dangxiao Cheng,

    Affiliation Ontario Cancer Institute, Toronto, Ontario, Canada

    ⨯
  • Qin Kuang,

    Affiliation Ontario Cancer Institute, Toronto, Ontario, Canada

    ⨯
  • Geoffrey Liu ,

    Geoffrey.Liu@uhn.on.ca (GL); wxu@uhnres.utoronto.ca (WX)

    Affiliations Ontario Cancer Institute, Toronto, Ontario, Canada, Division of Medical Oncology and Hematology, Princess Margaret Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada

    ⨯
  • Wei Xu

    Geoffrey.Liu@uhn.on.ca (GL); wxu@uhnres.utoronto.ca (WX)

    Affiliations Ontario Cancer Institute, Toronto, Ontario, Canada, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada, Department of Biostatistics, Princess Margaret Hospital, University Health Network, University of Toronto, Toronto, Ontario, Canada

    ⨯

Abstract

Background

Whether UGT1A1*28 genotype is associated with clinical outcomes of irinotecan (IRI)-based chemotherapy in Colorectal cancer (CRC) is an important gap in existing knowledge to inform clinical utility. Published data on the association between UGT1A1*28 gene polymorphisms and clinical outcomes of IRI-based chemotherapy in CRC were inconsistent.

Methodology/Principal Findings

Literature retrieval, trials selection and assessment, data collection, and statistical analysis were performed according to the PRISMA guidelines. Primary outcomes included therapeutic response (TR), progression-free survival (PFS) and overall survival (OS). We calculated odds ratios (OR) and hazard ratios (HR) with 95% confidence intervals (CI). Twelve clinical trials were included. No statistical heterogeneity was detected in analyses of all studies and for each subgroup. Differences in TR, PFS and OS for any genotype comparison, UGT1A1*28/*28 versus (vs) UGT1A1*1/*1 (homozygous model), UGT1A1*1/*28 vs UGT1A1*1/*1 (heterozygous model), and UGT1A1*28/*28 vs all others (recessive model, only for TR) were not statistically significant. IRI dose also did not impact upon TR and PFS differences between UGT1A1 genotype groups. A statistically significant increase in the hazard of death was found in Low IRI subgroup of the homozygous model (HR = 1.48, 95% CI = 1.06–2.07; P = 0.02). The UGT1A1*28 allele was associated with a trend of increase in the hazard of death in two models (homozygous model: HR = 1.22, 95% CI = 0.99–1.51; heterozygous model: HR = 1.13, 95% CI = 0.96–1.32). These latter findings were driven primarily by one single large study (Shulman et al. 2011).

Conclusions/Significance

UGT1A1*28 polymorphism cannot be considered as a reliable predictor of TR and PFS in CRC patients treated with IRI-based chemotherapy. The OS relationship with UGT1A1*28 in the patients with lower-dose IRI chemotherapy requires further validation.

Citation: Liu X, Cheng D, Kuang Q, Liu G, Xu W (2013) Association between UGT1A1*28 Polymorphisms and Clinical Outcomes of Irinotecan-Based Chemotherapies in Colorectal Cancer: A Meta-Analysis in Caucasians. PLoS ONE 8(3): e58489. https://doi.org/10.1371/journal.pone.0058489

Editor: Robert Lafrenie, Sudbury Regional Hospital, Canada

Received: October 2, 2012; Accepted: February 4, 2013; Published: March 14, 2013

Copyright: © 2013 Liu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was sponsored in part by the Alan B. Brown Chair in Molecular Genomics, CCO Chair in Experimental Therapeutics and Population Studies and Posluns Family Foundation (https://www.cancercare.on.ca/research/fundedresearch/2009recipients/). This work was also supported by the Scholarship Fund Award for Young Teachers Studying Abroad from the Department of Education of Hubei Province and Hubei University of Chinese Medicine (http://rsc.hbtcm.edu.cn/daohang/shizijianshe/2012-03-26/155.html). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Colorectal cancer (CRC) is the second leading cause of cancer-related death, and the most common cancer in the United States with 148,810 new cases and 49,960 deaths during 2008 alone [1]. Irinotecan (IRI) is one of the most effective chemotherapeutic agents in the treatment of CRC [2], [3]. At least 15% of individuals with new CRC are candidates for IRI therapy [1], [4].

IRI efficacy is dependent on activation by carboxyesterases to form the active metabolite 7-ethyl-10-hydroxycamptothecin (SN-38), which is a potent poison of topoisomerase I that interrupts DNA replication in cancer cells, resulting in cell death [5], [6], [7]. The major route of SN-38 elimination is via glucuronidation by the uridine diphosphate glucuronosyltransferase (UGT) 1A1, an essential enzyme involved in the complex metabolism of IRI [5]. UGT1A1*28 is a common allele with seven TA repeats in the promoter of UGT1A1 compared with the wild-type allele (UGT1A1*1) with six repeats [6], [7], [8]. A seven-repeat allele is associated with decreased gene transcription and expression of UGT1A1 and reduced enzyme activity, which lead to higher or more prolonged exposure of SN-38, the active form of IRI [8], [9].

Given that the UGT1A1 *28 variant influences IRI metabolism through enhanced exposure of its active metabolite SN-38, it is pharmacologically plausible that the UGT1A1*28 allele may be associated with the therapeutic efficacy of IRI in addition to the risk of adverse effects [10], [11]. Researchers have investigated the efficacy of IRI in CRC patients bearing different UGT1A1*28 genotypes [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31]. However, results are both conflicting and difficult to interpret because of small sample sizes and associated poor statistical power. Although a recent meta-analysis was performed to analyze the difference in therapeutic response (TR) between IRI-administered cancer patients with different UGT1A1*28 genotypes [10], it only provided data on TR, a surrogate for the most important outcome: survival, and included studies of different cancers rather than completely focusing on patients with CRC. This meta-analysis will therefore assess effects of UGT1A1*28 polymorphism on the efficacy of IRI-based chemotherapy, not only including TR but also survival. Moreover, it focuses on CRC alone, which will allow an assessment of uniform regimens tied to a single clinical disease site. In addition, two more recent publications on CRC are included [12], [13].

Materials and Methods

Retrieval of Published Studies

A comprehensive search of the PubMed and EMBASE databases was conducted from its inception through to July 2012 with the following search terms ‘irinotecan’, ‘UGT1A1’, ‘UGT1A1 polymorphism’, ‘UGT1A1*28’, ‘colorectal cancer’, ‘chemotherapy’, ‘response’, ‘progression-free survival (PFS)’, and ‘overall survival (OS)’. Furthermore, we screened titles and abstracts to identify relevant studies. Studies in abstract form or meeting reports, without publication of the full paper, were excluded.

The UGT1A1*28 polymorphism is relatively rare in Asian populations and the prevalence of homozygous UGT1A1*28 genotype is significantly greater in Caucasians than in Asian populations [32], [33], [34]. To reduce the heterogeneity among the analyzed studies, only studies involving mainly Caucasians populations were included in this meta-analysis.

Inclusion and Exclusion Criteria

Studies were included in the meta-analysis if (1) they were clinical trials or well characterized observational datasets, (2) they explored the association between UGT1A1*28 and clinical outcomes of IRI-based chemotherapy in patients with CRC, (3) there were sufficient data for TR (defined as a complete or partial response, using the WHO criteria [35] or the Response Evaluation Criteria in Solid Tumors criteria (RECIST) [36]) or PFS/OS data were provided, and (4) they were published in English. Exclusion criteria were as follows: (1) case reports, (2) reviews and opinions, (3) allele frequency studies, (4) studies not involving CRC patients, (5) studies where outcome data were not presented in detail or which had not provided enough information to calculate relevant data, and (6) studies conducted only in non-Caucasian populations. When different publications with overlapping subjects were considered eligible, we only included the one with larger numbers of patients. Figure 1 summarizes the search methods, inclusion and exclusion steps.

thumbnail
Download:
Figure 1. Flow diagram for study selection in meta-analysis.

https://doi.org/10.1371/journal.pone.0058489.g001

Data Extraction

The following information was extracted from each eligible included publication: first author’s name, year of publication, country, primary race of patients, phase of clinical trial, number of patient, gender distribution, age (median or mean), source of population, polymorphism detection method, IRI dose, chemotherapy regimens, study design, response criteria, line of chemotherapy, and genotype data.

Two or three different IRI-containing regimens were administered to patients in some studies [15], [20]. When possible, we analyzed the patients treated with each regimen as separate samples. Patients treated with different regimens were analyzed as a single study only if separate data was not available. Sample sizes abstracted reflect the total number of patients who received IRI, as some trials also included non-IRI treatment arm.

Statistical Analysis

PRISMA guidelines were followed (showed in Checklist S1) [37]. Odds ratios (OR) were used to estimate the association between UGT1A1*28 and TR. The OR was computed from the number of patients with and without TR after IRI-based chemotherapy. We evaluated PFS and OS based on pooled Cox proportional hazard ratios (HR) and 95% confidence intervals (CI) using published methods [38] because a meta-analysis of summary results is statistically as efficient as a joint analysis of individual participant data [39]. Between-study heterogeneity was assessed using the Cochran’s Chi-Squared test and the inconsistency index I2, with a significance level of P<0.05. We performed initial analyses with a fixed-effect model and confirmatory analyses with a random-effects model, if there was potential heterogeneity. We assessed potential publication bias by using a funnel plot and Egger’s test [40]. For meta-analysis that failed the Egger’s test (P<0.05), a trim and fill method was used to adjust for publication bias [41]. All statistical analysis was performed using Review Manager (v5.0; Oxford, England) and Stata software (Stata Corporation, Texas).

For TR, we compared the following: UGT1A1*28/*28 versus (vs) UGT1A1*1/*1 (homozygous model), UGT1A1*1/*28 vs UGT1A1*1/*1 (heterozygous model) and UGT1A1*28/*28 vs all others (recessive model). Two models (homozygous and heterozygous model) were examined in the analysis of PFS and OS. To assess the influence of IRI dose on the association between UGT1A1*28 and clinical outcomes, we carried out stratified analyses based on different IRI doses. In dose intensity analysis, 150 mg/m2 of IRI dose was set as the cutoff value between medium/high (High IRI) and low dose (Low IRI). In some studies [14], [20], [22], [23], the patients received different IRI doses at different time points and only combined data were available. The average dose was calculated to classify these studies.

Results

Characteristics of the Studies

Figure 1 shows the process of study selection. In total, 27 full-text studies were fully reviewed. Of these, five did not provide sufficient individuals’ genotype data [24], [25], [26], [28], [29]. Three only provided a summary description of their results in the text [27], [30], [31]. Four analyzed only Asian populations and were excluded [42], [43], [44], [45]. Three combined CRC patients with those that had other cancers [46], [47], [48]. Thus, only 12 studies were eligible for inclusion in our meta-analysis.

Characteristics of the included studies are summarized in Table 1. Methodologic components of study designs may be critically important to understand the meta-analyses results [49]; thus we utilize a modified set of criteria to report methodological issues and quality of the studies [50]. The criteria assessed study design, polymorphism detection method, combination regimens, Line of therapy, and grading systems for response (Table 1).

thumbnail
Download:
Table 1. Characteristics and methodological quality of studies included in meta analysis.

https://doi.org/10.1371/journal.pone.0058489.t001

Of the 12 studies, three did not clearly report the race of the participants [12], [16], [19], but they were conducted in Europe or America. Because the UGT1A1*28 allele frequencies were similar to Caucasians, these three were classified with Caucasian studies. The study by Shulman et al [13] was included in the Low IRI subgroup based on the authors’ own comments. The results of meta-analysis are summarized in Table 2.

thumbnail
Download:
Table 2. The association between UGT1A1*/28 polymorphisms and therapeutic response, progression-free survival and overall survival.

https://doi.org/10.1371/journal.pone.0058489.t002

Association between UGT1A1*28 and TR

Analysis of pooled data from all samples indicated that UGT1A1*28 allele was not associated with TR in CRC. In the homozygous model, the OR was 1.09 (95% CI = 0.74–1.60; P = 0.66) (Figure 2–2A). For UGT1A1*1/*28 vs UGT1A1*1/*1, the OR was 1.00 (95% CI = 0.80–1.26; P = 0.70) (Figure 2–2B). The recessive comparison had an OR of 1.08 (95% CI = 0.80–1.25; P = 0.69) (Figure 2–2C). Stratified analysis by IRI dose showed that the differences in TR between genotype groups were not statistically significant for any of the IRI dose levels (Table 2, Figure 2). The heterogeneity across all studies was not statistically significant for any model. I2 values were 32 (P = 0.13), 0 (P = 0.65) and 24% (P = 0.20) respectively for homozygous, heterozygous and recessive models (Table 2). No publication bias was detected by either the funnel plot (Figure S1) or Egger’s tests (P>0.05, each comparison).

thumbnail
Download:
Figure 2. Forest plots of three comparisons; outcome: therapeutic response.

2A: *28/*28 versus *1/*1; 2B: *1/*28 versus *1/*1; 2C: *28/*28 versus *1/*28 or *1/*1.

https://doi.org/10.1371/journal.pone.0058489.g002

Association between UGT1A1*28 and PFS

Pooled data from all samples for two genotype comparisons indicated that the UGT1A1*28 allele was not associated with a significant decrease of hazard for PFS in CRC (Figure 3 and Table 2). In homozygous and heterozygous models, the HRs were 0.86 (95% CI = 0.72–1.04; P = 0.18 for heterogeneity, I2 = 35%) and 1.00 (95% CI = 0.86–1.17; P = 0.14 for heterogeneity, I2 = 39%), respectively. Subgroup analyses based on IRI dose did not show any significant difference in terms of the association between UGT1A1*28 genotype and the hazard of PFS. There was no evidence of publication bias given the symmetrical distributions of funnel plots (Figure S2) and Egger’s tests (P = 0.28 and 0.14).

thumbnail
Download:
Figure 3. Forest plots of two comparisons, outcome: progression-free survival.

3A: *28/*28 versus *1/*1; 3B: *1/*28 versus *1/*1.

https://doi.org/10.1371/journal.pone.0058489.g003

Association between UGT1A1*28 and OS

Five studies involving 551 patients were analyzed for the homozygous model and five studies (893 cases) for heterozygous model. The UGT1A1*28 allele was associated with a non-significant increase in the hazard of death in two models (homozygous model: HR = 1.22, 95% CI = 0.99–1.51, P = 0.06; heterozygous model: HR = 1.13, 95% CI = 0.96–1.32, P = 0.14) (Figure 4 and Table 2). However, subgroup analysis found a statistically significant increase in the hazard of death in Low IRI subgroup for the homozygous model (HR = 1.48, 95% CI = 1.06–2.07; P = 0.02). No associations were seen in the heterozygous model. No publication bias was detected in the funnel plots (Figure S3) and Egger’s tests (P>0.05), and there was no heterogeneity in each model (I2<5%, P>0.05).

thumbnail
Download:
Figure 4. Forest plots of two comparisons, outcome: overall survival.

4A: *28/*28 versus *1/*1; 4B: *1/*28 versus *1/*1.

https://doi.org/10.1371/journal.pone.0058489.g004

Discussion

While the association between UGT1A1*28 and IRI-related toxicity has been extensively studied, data are limited regarding the potential impact of the UGT1A1*28 genotype on tumor responsiveness and patient survival following IRI therapy [10]. Published meta-analyses have demonstrated dose-dependent associations between UGT1A1*28 genotype and IRI-induced neutropenia or diarrhea [51], [52], [53]. The U.S. Food and Drug Administration in 2005 recommended that gene-related information be added to the drug product label and approved the diagnostic UGT1A1*28 test to identify homozygous patients advising a lower dose of IRI in these patients [54]. However, a reduction in dosage might also be associated with reduced tumor response and/or increased morbidity [55]. The Evaluation of Genomic Applications in Practice and Prevention (EGAPP) working group and some cost-effectiveness analyses have indicated that UGT1A1*28 genotyping will only be clinically useful if dosing IRI on basis of genotype improves the safety of IRI without compromising the efficacy of the therapy [55], [56], [57], [58], [59], [60]. Thus, whether UGT1A1*28 genotype is associated with clinical outcomes of IRI-based chemotherapy is an important gap in existing knowledge to inform clinical utility [55], [56].

A previous meta-analysis [10] of 12 studies (8 studies [14], [15], [16], [17], [18], [20], [21], [23] were included in our meta-analysis) was performed to analyze the difference in TR between IRI-administered cancer patients with different UGT1A1*28 genotypes. Results indicated that differences in TR for all genotype comparisons were not statistically significant. Subgroup analyses based on IRI dose and tumor type (CRC and lung cancer) did not show any significant difference in terms of the association between UGT1A1 genotype and TR. The present meta-analysis assessed the association of UGT1A1*28 polymorphisms with clinical outcomes of IRI-based chemotherapy in a single cancer site, CRC. In our meta-analyses, the differences in TR between the different UGT1A1*28 genotype patient groups also did not attain statistical significance. Also, in contrast to IRI-induced toxicities, there was no convincing evidence to suggest that the association between UGT1A1*28 genotype and TR is modified by IRI dose. This is consistent with the meta-analysis of Dias et al [10], where no association between UGT1A1*28 genotypes and IRI response was found in an analysis across various tumor types including CRC. Similar results were detected in association between UGT1A1*28 genotypes and PFS. However, the UGT1A1*28 allele showed significant or marginal association with poorer OS, especially in Low IRI subgroup of homozygous model.

Our OS results are in the opposite direction of our original hypothesis. Possible explanations for why OS could be lower in patients carrying the UGT1A1*28 allele include suboptimal treatment due to the severity of adverse effects and the decreased dose intensity resulting from frequent dose reduction or treatment delay [13], [59]. These two parameters are intrinsically correlated but not necessarily consistent with one another [61]. OS is defined as the time from randomization to death caused by any reason and represents the gold standard metric for establishing efficacy. This typically requires phase III trials of large sample size with lengthy follow-up. TR and PFS as the alternative end points for OS occur earlier and can evaluate the effect of an intervention faster, at less cost with fewer trial subjects [62]. However, Prediction of TR and PFS are more complicated because other factors, such as tumor-related factors, environmental factors, and patient’s characteristics, should be considered [59]. Hence, TR and PFS may correlate with a real clinical endpoint (OS) but do not necessarily have a guaranteed relationship. That a reduction in IRI dosage among UGT1A1*28*28 patients with CRC may reduce the long-term survival (OS), but not influence on TR and FPS is intriguing; however, an underlying mechanism needs to be clarified.

The study conducted by Shulman et al was included in this meta-analysis, which may have driven the OS findings due to its large sample size (329 cases). This study suffered in particular from an unspecified IRI dosage and thus was only marginally accepted based on its methodology. An additional meta-analysis excluding the study showed that the results of HRs were 1.09 (homozygous model) and 1.08 (heterozygous model), respectively, which were slightly different from the HRs in the overall estimate of 1.22 and 1.13. However, the analysis failed to reach a statistical significance because of the insufficient power with the small sample size in this meta-analysis (only four studies involving 222 patients were analyzed for the homozygous model and 564 cases for heterozygous model). Thus, while our OS relationship is intriguing, much more validation is needed.

In our meta-analysis, three parameters (TR, PFS and OS) were used to assess the influence of UGT1A1*28 polymorphism on clinical outcomes. This presents a more comprehensive assessment than a single parameter as a prior meta-analysis had performed [61]. Moreover, our study only focused on CRC, reducing the potential heterogeneity across the studies. In addition, we paid attention to methodological components of study designs in the literature. The certain items such as study design, polymorphism detection method, combination regimens, Line of therapy, and grading systems for response, are reflective of methodological and reporting quality of the studies. It is beneficial to analyze the heterogeneity in this meta-analysis and improve the chances to replicate initial significant findings in subsequent pharmacogenetic studies [50].

There are limitations of this analysis. Firstly, some studies were excluded from our analysis because of lack of individual genotype data [24], [25], [26], [27], [28], [29], [30], [31]; this could cause some bias in our estimates, but was unlikely to change our major conclusions, as these excluded studies showed no association between UGT1A1*28 polymorphism and either TR or PFS in Caucasians. Secondly, there is inherent heterogeneity to all meta-analyses. In the analyzed studies, there were differences in study design, the source of population, IRI dose, polymorphism detection method, response grade criteria, therapeutic regimens, line of therapy, and performance status of patients. Additionally, although the difference in distribution of stage at diagnosis across studies will contribute to the heterogeneity in our meta-analysis, we did not perform a stratified subgroup analysis on stage at diagnosis because only three studies [13], [18], [21] provided data on stage and none explored the association between UGT1A1*28 and clinical outcomes of different stage at diagnosis. Likewise, other stratified subgroup analyses such as on the localization of primary tumor (six studies reported the localization of primary tumor [12], [14], [17], [18], [19], [20], but they did not provided the separated outcome data) could not be conducted. Thus, we performed initial analyses using a fixed-effects model and confirmatory analyses using a random-effects model. Results were similar between these two methods. Thirdly, articles included in this meta–analysis were restricted to English publishing studies. Articles with potentially high-quality data in other languages were excluded because of anticipated difficulties in obtaining accurate medical translation.

Although meta-analysis can synthesize the results of multiple studies into a summary of results, it is different from a combined analysis which uses the full information of individual patient data and can provide more comprehensive assessment. Our meta-analysis was useful because the collection of detailed information of each clinical trial was impractical and in some cases, not allowable due to local ethics board concerns [63].

In summary, this meta-analysis provided modest evidence for the association between UGT1A1*28 polymorphism and OS of IRI-based chemotherapy in CRC. UGT1A1*28 polymorphism cannot be considered a reliable predictor of TR and PFS to IRI-based chemotherapy in patients with CRC. In contrast, the OS may be affected by UGT1A1*28 status; the UGT1A1*28*/28 patients with CRC have a worse OS after lower-dose IRI therapy. The clinical significance of this last finding requires replication and additional research. In particular, as IRI metabolism is complex, numerous genes in addition to UGT1A1 should be interrogated [27].

Supporting Information

Figure S1.

Funnel plots of three comparisons, outcome: therapeutic response. 1A: *28/*28 versus *1/*1; 1B: *1/*28 versus *1/*1; 1C:*28/*28 versus *1/*28 or *1/*1.

https://doi.org/10.1371/journal.pone.0058489.s001

(TIF)

Figure S2.

Funnel plots of two comparisons, outcome: progression-free survival. 2A: *28/*28 versus *1/*1; 2B: *1/*28 versus *1/*1.

https://doi.org/10.1371/journal.pone.0058489.s002

(TIF)

Figure S3.

Funnel plots of two comparisons, outcome: overall survival. 3A: *28/*28 versus *1*1; 3B: *1/*28 versus *1/*1.

https://doi.org/10.1371/journal.pone.0058489.s003

(TIF)

Author Contributions

Conceived and designed the experiments: WX GL XL. Performed the experiments: XL DC QK. Analyzed the data: WX XL GL. Contributed reagents/materials/analysis tools: DC QK. Wrote the paper: XL WX GL.

References

  1. 1. Jemal A, Siegel R, Ward E, Hao Y, Xu J, et al. (2008) Cancer statistics, 2008. CA Cancer J Clin 58: 71–96.
  2. 2. Garcia-Carbonero R, Supko JG (2002) Current perspectives on the clinical experience, pharmacology, and continued development of the camptothecins. Clin Cancer Res 8: 641–661.
  3. 3. Ulukan H, Swaan PW (2002) Camptothecins: a review of their chemotherapeutic potential. Drugs 62: 2039–2057.
  4. 4. Wu X, Cokkinides V, Chen VW, Nadel M, Ren Y, et al.
  5. 5. Mathijssen RH, van Alphen RJ, Verweij J, Loos WJ, Nooter K, et al. (2001) Clinical pharmacokinetics and metabolism of irinotecan (CPT-11). Clin Cancer Res 7: 2182–2194.
  6. 6. Miners JO, McKinnon RA, Mackenzie PI (2002) Genetic polymorphisms of UDP-glucuronosyltransferases and their functional significance. Toxicology 181–182: 453–456.
  7. 7. Toffoli G, Cecchin E, Corona G, Boiocchi M (2003) Pharmacogenetics of irinotecan. Curr Med Chem Anticancer Agents 3: 225–237.
  8. 8. Beutler E, Gelbart T, Demina A (1998) Racial variability in the UDP-glucuronosyltransferase 1 (UGT1A1) promoter: a balanced polymorphism for regulation of bilirubin metabolism? Proc Natl Acad Sci U S A 95: 8170–8174.
  9. 9. Iyer L, Hall D, Das S, Mortell MA, Ramírez J, et al. (1999) 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: 576–82.
  10. 10. Dias MM, McKinnon RA, Sorich MJ (2012) Impact of the UGT1A1*28 allele on response to irinotecan: a systematic review and meta-analysis. Pharmacogenomics 13(8): 889–899.
  11. 11. Innocenti F, Undevia SD, Iyer L, Chen PX, Das S, et al. (2004) Genetic variants in the UDP-glucuronosyltransferase 1A1 gene predict the risk of severe neutropenia of irinotecan. J Clin Oncol 22: 1382–1388.
  12. 12. Lamas MJ, Duran G, Balboa E, Bernardez B, Candamio S, et al. (2012) The value of genetic polymorphisms to predict toxicity in metastatic colorectal patients with irinotecan-based regimens. Cancer Chemother Pharmacol 69(6): 1591–1599.
  13. 13. Shulman K, Cohen I, Barnett-Griness O, Kuten A, Gruber SB, et al. (2011) Clinical implications of UGT1A1*28 genotype testing in colorectal cancer patients. Cancer 117(14): 3156–3162.
  14. 14. Martinez-Balibrea E, Abad A, Martínez-Cardús A, Ginés A, Valladares M, et al. (2010) UGT1A and TYMS genetic variants predict toxicity and response of colorectal cancer patients treated with first-line irinotecan and fluorouracil combination therapy. Br J Cancer 103(4): 581–589.
  15. 15. McLeod HL, Sargent DJ, Marsh S, Green EM, King CR, et al. (2010) Pharmacogenetic predictors of adverse events and response to chemotherapy in metastatic colorectal cancer: results from North American Gastrointestinal Intergroup Trial N9741. J Clin Oncol 28(20): 3227–3233.
  16. 16. Boige V, Mendiboure J, Pignon JP, Loriot MA, Castaing M, et al. (2010) Pharmacogenetic assessment of toxicity and outcome in patients with metastatic colorectal cancer treated with LV5FU2, FOLFOX, and FOLFIRI: FFCD 2000–05. J Clin Oncol 28(15): 2556–2564.
  17. 17. Glimelius B, Garmo H, Berglund A, Fredriksson LA, Berglund M, et al. (2011) Prediction of irinotecan and 5-fluorouracil toxicity and response in patients with advanced colorectal cancer. Pharmacogenomics J 11(1): 61–71.
  18. 18. Toffoli G, Cecchin E, Corona G, Russo A, Buonadonna A, et al. (2006) The role of UGT1A1*28 polymorphism in the pharmacodynamics and pharmacokinetics of irinotecan in patients with metastatic colorectal cancer. J Clin Oncol 24(19): 3061–3068.
  19. 19. Rouits E, Charasson V, Pétain A, Boisdron-Celle M, Delord JP, et al. (2008) Pharmacokinetic and pharmacogenetic determinants of the activity and toxicity of irinotecan in metastatic colorectal cancer patients. Br J Cancer 99(8): 1239–1245.
  20. 20. Kweekel DM, Gelderblom H, Van der Straaten T, Antonini NF, Punt CJ, et al. (2008) UGT1A1*28 genotype and irinotecan dosage in patients with metastatic colorectal cancer: a Dutch Colorectal Cancer Group study. Br J Cancer 99(2): 275–282.
  21. 21. Ruzzo A, Graziano F, Loupakis F, Santini D, Catalano V, et al. (2008) Pharmacogenetic profiling in patients with advanced colorectal cancer treated with first-line FOLFIRI chemotherapy. Pharmacogenomics J 8(4): 278–288.
  22. 22. Rhodes KE, Zhang W, Yang D, Press OA, Gordon M, et al. (2007) ABCB1, SLCO1B1 and UGT1A1 gene polymorphisms are associated with toxicity in metastatic colorectal cancer patients treated with first-line irinotecan. Drug Metab Lett 1(1): 23–30.
  23. 23. Carlini LE, Meropol NJ, Bever J, Andria ML, Hill T, et al. (2005) UGT1A7 and UGT1A9 polymorphisms predict response and toxicity in colorectal cancer patients treated with capecitabine/irinotecan. Clin Cancer Res 11(3): 1226–1236.
  24. 24. Choi YH, Kim TW, Kim KP, Lee SS, Hong YS, et al. (2012) A Phase II Study of Clinical Outcomes of 3-Week Cycles of Irinotecan and S-1 in Patients with Previously Untreated Metastatic Colorectal Cancer: Influence of the UGT1A1 and CYP2A6 Polymorphisms on Clinical Activity. Oncology 82(5): 290–297.
  25. 25. Freyer G, Duret A, Milano G, Chatelut E, Rebischung C, et al. (2011) Pharmacogenetic tailoring of irinotecan-based first-line chemotherapy in metastatic colorectal cancer: results of a pilot study. Anticancer Res 31(1): 359–366.
  26. 26. Hazama S, Nagashima A, Kondo H, Yoshida S, Shimizu R, et al. (2010) Phase I study of irinotecan and doxifluridine for metastatic colorectal cancer focusing on the UGT1A1*28 polymorphism. Cancer Sci 101(3): 722–727.
  27. 27. Schulz C, Heinemann V, Schalhorn A, Moosmann N, Zwingers T, et al. (2009) UGT1A1 gene polymorphism: impact on toxicity and efficacy of irinotecan-based regimens in metastatic colorectal cancer. World J Gastroenterol 15(40): 5058–5066.
  28. 28. Côté JF, Kirzin S, Kramar A, Mosnier JF, Diebold MD, et al. (2007) UGT1A1 polymorphism can predict hematologic toxicity in patients treated with irinotecan. Clin Cancer Res 13(11): 3269–3275.
  29. 29. Martinez-Balibrea E, Manzano JL, Martinez-Cardus A, Moran T, Cirauqui B, et al. (2007) Combined analysis of genetic polymorphisms in thymidylate synthase, uridine diphosphate glucoronosyltransferase and X-ray cross complementing factor 1 genes as a prognostic factor in advanced colorectal cancer patients treated with 5-fluorouracil plus oxaliplatin or irinotecan. Oncol Rep 17(3): 637–645.
  30. 30. Massacesi C, Terrazzino S, Marcucci F, Rocchi MB, Lippe P, et al. (2006) Uridine diphosphate glucuronosyl transferase 1A1 promoter polymorphism predicts the risk of gastrointestinal toxicity and fatigue induced by irinotecan-based chemotherapy. Cancer 106(5): 1007–1016.
  31. 31. Marcuello E, Altés A, Menoyo A, Del Rio E, Gómez-Pardo M, et al. (2004) UGT1A1 gene variations and irinotecan treatment in patients with metastatic colorectal cancer. Br J Cancer 91(4): 678–682.
  32. 32. Yong WP, Innocenti F, Ratain MJ (2006) The role of pharmacogenetics in cancer therapeutics. Br J Clin Pharmacol 62: 35–46.
  33. 33. Hall D, Ybazeta G, Destro-Bisol G, Petzl-Erler ML, Di Rienzo A (1999) Variability at the uridine diphosphate glucuronosyltransferase 1A1 promoter in human populations and primates. Pharmacogenetics 9: 591–599.
  34. 34. Premawardhena A, Fisher CA, Liu YT, Verma IC, de Silva S, et al. (2003) The global distribution of length polymorphisms of the promoters of the glucuronosyltransferase 1 gene (UGT1A1): hematologic and evolutionary implications. Blood Cells Mol Dis 31: 98–101.
  35. 35. Miller AB, Hoogstraten B, Staquet M, Winkler A (1981) Reporting results of cancer treatment. Cancer 47: 207–214.
  36. 36. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, et al. (2000) New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 92: 205–216.
  37. 37. Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the prisma statement. J Clin Epidemiol 62(10): 1006–1012.
  38. 38. Tierney JF, Stewart LA, Ghersi D, Burdett S, Sydes MR (2007) Practical methods for incorporating summary time-to-event data into meta-analysis. Trials 8: 16.
  39. 39. Lin DY, Sullivan PF (2009) Meta-analysis of genome-wide association studies with overlapping subjects. Am J Hum Genet 85: 862–872.
  40. 40. Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315: 629–634.
  41. 41. Duval S, Tweedie R (2000) A nonparametric “Trim and Fill” method of accounting for publication bias in meta-analysis. Biometrics 56(2): 455–463.
  42. 42. Sunakawa Y, Fujita K, Ichikawa W, Ishida H, Yamashita K, et al. (2012) A Phase I Study of Infusional 5-Fluorouracil, Leucovorin, Oxaliplatin and Irinotecan in Japanese Patients with Advanced Colorectal Cancer Who Harbor UGT1A1*1/*1,*1/*6 or *1/*28. Oncology 82(4): 242–248.
  43. 43. Okuyama Y, Hazama S, Nozawa H, Kobayashi M, Takahashi K, et al. (2011) Prospective phase II study of FOLFIRI for mCRC in Japan, including the analysis of UGT1A1 28/6 polymorphisms. Jpn J Clin Oncol 41(4): 477–482.
  44. 44. Sunakawa Y, Ichikawa W, Fujita K, Nagashima F, Ishida H, et al. (2011) UGT1A1*1/*28 and *1/*6 genotypes have no effects on the efficacy and toxicity of FOLFIRI in Japanese patients with advanced colorectal cancer. Cancer Chemother Pharmacol 68(2): 279–284.
  45. 45. Liu CY, Chen PM, Chiou TJ, Liu JH, Lin JK, et al. (2008) UGT1A1*28 polymorphism predicts irinotecan-induced severe toxicities without affecting treatment outcome and survival in patients with metastatic colorectal carcinoma. Cancer 112(9): 1932–1940.
  46. 46. Wright MA, Morrison G, Lin P, Leonard GD, Nguyen D, et al. (2005) A phase I pharmacologic and pharmacogenetic trial of sequential 24-hour infusion of irinotecan followed by leucovorin and a 48-hour infusion of fluorouracil in adult patients with solid tumors. Clin Cancer Res 11(11): 4144–4150.
  47. 47. Lara PN Jr, Natale R, Crowley J, Lenz HJ, Redman MW, et al. (2009) Phase III trial of irinotecan/cisplatin compared with etoposide/cisplatin in extensive-stage small-cell lung cancer: clinical and pharmacogenomic results from SWOG S0124. J Clin Oncol 27(15): 2530–2535.
  48. 48. Han JY, Lim HS, Shin ES, Yoo YK, Park YH, et al. (2006) Comprehensive analysis of UGT1A polymorphisms predictive for pharmacokinetics and treatment outcome in patients with non-small-cell lung cancer treated with irinotecan and cisplatin. J Clin Oncol 24(15): 2237–2244.
  49. 49. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, et al.
  50. 50. Jorgensen AL, Williamson PR (2008) Methodological quality of pharmacogenetic studies: issues of concern. Stat Med 27: 6547–6569.
  51. 51. Hoskins JM, Goldberg RM, Qu P, Ibrahim JG, McLeod HL (2007) UGT1A1*28 genotype and irinotecan-induced neutropenia: dose matters. J Natl Cancer Inst 99: 1290–1295.
  52. 52. Zhe-Yi H, Qi Y, Pei Q, Guo C (2010) Dose-Dependent Association between UGT1A1*28 Genotype and Irinotecan-Induced Neutropenia: Low Doses Also Increase Risk. Clin Cancer Res 16(15): 3832–3842.
  53. 53. Hu ZY, Yu Q, Zhao YS (2010) Dose-dependent association between UGT1A1*28 polymorphism and irinotecan-induced diarrhoea: a meta-analysis. Eur J Cancer 46(10): 1856–1865.
  54. 54. O’Dwyer PJ, Catalano RB (2006) Uridine diphosphate glucuronosyltransferase (UGT) 1A1 and irinotecan: practical pharmacogenomics arrives in cancer therapy. J Clin Oncol 24: 4534–4538.
  55. 55. Palomaki GE, Bradley LA, Douglas MP, Kolor K, Dotson WD (2009) Can UGT1A1 genotyping reduce morbidity and mortality in patients with metastatic colorectal cancer treated with irinotecan? An evidence-based review. Genet Med 11(1): 21–34.
  56. 56. Berg AO, Armstrong K, Botkin J, Calonge N, Haddow J, et al.
  57. 57. Gold HT, Hall MJ, Blinder V, Schackman BR (2009) Cost effectiveness of pharmacogenetic testing for uridine diphosphate glucuronosyltransferase 1A1 before irinotecan administration for metastatic colorectal cancer. Cancer 115(17): 3858–3867.
  58. 58. Obradovic M, Mrhar A, Kos M (2008) Cost–effectiveness of UGT1A1 genotyping in second-line, high-dose, once every 3 weeks irinotecan monotherapy treatment of colorectal cancer. Pharmacogenomics 9(5): 539–549.
  59. 59. Kim TW, Innocenti F (2007) Insights, challenges, and future directions in irinogenetics. Ther Drug Monit 29(3): 265–270.
  60. 60. Hoskins JM, McLeod HL (2009) UGT1A and irinotecan toxicity: keeping it in the family. J Clin Oncol 27(15): 2419–2421.
  61. 61. Oye RK, Shapiro MF (1984) Reporting results from chemotherapy trials. Does response make a difference in patient survival? JAMA 252: 2722–2725.
  62. 62. Temple RJ (1995) A Regulatory Authority’s Opinion About Surrogate Endpoints. In: Nimmo WS, Tucker GT, editors. Clinical Measurement in Drug Evaluation. New York: J. Wiley and Sons. pp.
    • 63. Berman NG, Parker RA (2002) Meta-analysis: neither quick nor easy. BMC Med Res Methodol. Aug 9 2: 10.