Originally published as JCO Early Release 10.1200/JCO.2006.10.4182 on February 25 2008

This Article Abstract Full Text (PDF) Publisher's Note Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schwab, M. Articles by Eichelbaum, M. Search for Related Content PubMed PubMed Citation Articles by Schwab, M. Articles by Eichelbaum, M. Related Articles Related Editorial Related Correspondence Social Bookmarking                            What's this?

Role of Genetic and Nongenetic Factors for Fluorouracil Treatment-Related Severe Toxicity: A Prospective Clinical Trial by the German 5-FU Toxicity Study Group

Matthias Schwab, Ulrich M. Zanger, Claudia Marx, Elke Schaeffeler, Kathrin Klein, Jürgen Dippon, Reinhold Kerb, Julia Blievernicht, Joachim Fischer, Ute Hofmann, Carsten Bokemeyer, Michel Eichelbaum

From the Dr Margarete Fischer-Bosch Institute of Clinical Pharmacology and Department of Mathematics, University of Stuttgart, Stuttgart; University of Tuebingen and Department of Internal Medicine II, University Medical Center, Tuebingen; Epidauros Biotechnology AG, Bernried; and Department of Internal Medicine II (Oncology, Hematology, Bone marrow transplantation, Pneumology), University Medical Center Hamburg-Eppendorf, Hamburg, Germany

Corresponding author: Ulrich M. Zanger, PhD, Dr Margarete Fischer-Bosch Institute of Clinical Pharmacology, Auerbachstraβe 112, D-70376 Stuttgart, Germany; e-mail: uli.zanger{at}ikp-stuttgart.de and Carsten Bokemeyer, MD, Department of Internal Medicine II (Oncology, Hematology, Bone marrow transplantation, Pneumology), University Medical Center Hamburg-Eppendorf, Martinistrasse 52, D-20246 Hamburg, Germany; e-mail: c.bokemeyer{at}uke.uni-hamburg.de

	ABSTRACT

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Purpose To assess the predictive value of polymorphisms in dihydropyrimidine dehydrogenase (DPYD ), thymidylate synthase (TYMS ), and methylene tetrahydrofolate reductase (MTHFR ) and of nongenetic factors for severe leukopenia, diarrhea, and mucositis related to fluorouracil (FU) treatment.

Patients and Methods A multicenter prospective clinical trial included 683 patients with cancer treated with FU monotherapy. Toxicity was documented according to World Health Organization grades. DPYD, TYMS, and MTHFR genotypes were determined, and DPYD was resequenced in patients with severe toxicity.

Results Grade 3 to 4 toxicity occurred in 16.1% of patients. The sensitivity of DPYD*2A genotyping for overall toxicity was 5.5% (95%CI, 0.02 to 0.11), with a positive predictive value of 0.46 (95% CI, 0.19 to 0.75; P = .01). Inclusion of additional DPYD variants improved prediction only marginally. Analysis according to toxicity type revealed significant association of DPYD with mucositis and leukopenia, whereas TYMS was associated with diarrhea. Genotype, female sex, mode of FU administration, and modulation by folinic acid were identified as independent risk factors by multivariable analysis. A previously unrecognized significant interaction was found between sex and DPYD, which resulted in an odds ratio for toxicity of 41.8 for male patients (95% CI, 9.2 to 190; P < .0001) but only 1.33 (95% CI, 0.34 to 5.2) in female patients. Homozygosity for the TYMS enhancer region double repeat allele increased risk for toxicity 1.6-fold (95% CI, 1.08 to 2.22; P = .02).

Conclusion DPYD, TYMS, and MTHFR play a limited role for FU related toxicity but a pronounced DPYD gene/sex-interaction increases prediction rate for male patients. Toxicity risk assessment should include sex, mode of administration, and folinic acid as additional predictive factors.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
The fluorinated pyrimidine analog fluorouracil (FU) has been widely used since the 1960s as first-line therapy for colorectal cancer both in the adjuvant and palliative setting and remains the backbone of many combination chemotherapy regimens.^1-3 Although the benefits of FU-based adjuvant chemotherapy in reducing the risk of relapse and prolonging survival are well established, major side effects are frequent and often lead to treatment discontinuation. Gastrointestinal and hematologic toxicities are most common, and dose, schedule, and mode of application have been implicated as factors that can modulate the type and severity of toxic reactions.⁴ FU undergoes complex anabolic and catabolic biotransformations that play an eminent role for both antitumor activity and toxicity (Appendix Fig A1, available online only). Although effectiveness of FU depends on its bioactivation, resulting in 5-fluoronucleotides that interfere with normal DNA and RNA function preferentially in cancer cells, more than 80% of a given dose are rapidly metabolized by dihydropyrimidine dehydrogenase (DPD), the rate-limiting enzyme of pyrimidine catabolism, to the inactive 5-fluoro-5,6-dihydrouracil.⁵ A potential role of DPD in determining FU toxicity was initially suggested by Diasio et al,⁶ who showed complete deficiency of DPD enzyme activity in a patient with the rare familial metabolic disorder, pyrimidinemia, leading to severe neurotoxicity during FU treatment. Later studies demonstrated correlation between lymphocyte DPD activity and FU clearance.^7-9 Additional evidence came from lethal drug interactions between FU and sorivudine, a DPD suicide inhibitor, in Japanese patients.¹⁰ Numerous genetic variants in the DPYD gene were described, including the most common variant, a G to A change in the 5'-splice recognition site of intron 14 (exon 14-skipping mutation), which leads to a corresponding mRNA lacking exon 14 and an enzymatically deficient enzyme.^11-16 Other patient characteristics including age and sex have also been suggested to influence FU clearance,^17,18 and two recent studies have shown that female patients experience FU toxicity more frequently and more severely than men.^19,20 Other candidate genes considered as potential factors for FU toxicity include the thymidylate synthase (TYMS ), which is strongly inhibited by FU and considered to be the major drug target, as well as methylenetetrahydrofolate reductase (MTHFR ), which forms the reduced folate cofactor essential for TYMS inhibition.²¹ The variable number of tandem repeat polymorphism (VNTR) of a 28–base pair (bp) sequence in the TYMS 5'-untranslated region and a 6-bp insertion/deletion polymorphism in the 3'-untranslated region (3'-UTR) have been associated with altered TYMS expression and clinical response.^22-27 Two tightly linked nonsynonymous MTHFR polymorphisms 677C>T and 1298A>C result in decreased enzyme activity, thus affecting intracellular folate metabolites and possibly FU sensitivity.^21,27-31

Available data thus suggest a role of both genetic and nongenetic factors for FU toxicity, but current evidence is mostly based on case reports, retrospective studies, or analyses of single predictive factors. In this multicenter clinical trial, we investigated prospectively the predictive value of genetic and nongenetic factors for severe toxicity in patients with cancer receiving FU monotherapy.

	PATIENTS AND METHODS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Study Population and Design
The study was approved by the ethics committee of the University Hospital Tuebingen, Germany, and written informed consent was obtained from all patients. The study protocol specified the inclusion criteria as follows: (1) only patients prospectively receiving FU monotherapy or in combination with folinic acid or levamisole were eligible (Appendix Table A1, online only); (2) other chemotherapeutic agents were not allowed; (3) selection of chemotherapeutic treatment regimen was independent of this study and was exclusively left to the discretion of the attending physician; (4) the same chemotherapy regimen was maintained during study observation; and (5) type of tumor included colon cancer, other gastrointestinal tumors, cancer of unknown primary, and breast cancer. Before FU treatment, patients were medically assessed and baseline demographic data and medical history, including the diagnosis of tumor and Dukes' classification, were recorded by the responsible physician. Patients were consecutively recruited at each study center. Throughout treatment, patients were monitored for onset, severity, and type of adverse events assessed using WHO toxicity classification criteria (http://www.fda.gov/cder/cancer/toxicityframe.htm). The primary end point of the study was the onset of toxicity independent of WHO grade. Analysis was restricted to the three major toxicity types (leukopenia, diarrhea, and mucositis), because only these are clearly related to FU treatment.^4,19

Genotyping for DPYD, TYMS, and MTHFR
Genomic DNA was prepared from peripheral leukocytes by standard methods. The DPYD splice site mutation IVS14 + 1G>A (DPYD*2A ) was analyzed as previously published.³² Assays for further DPYD variants tested are available in the Appendix. The TYMS VNTR polymorphism (rs34743033) was analyzed as described.²² The homozygous double repeat variant genotype was designated as 2/2, the heterozygous genotype as 2/3, and the homozygous triple repeat genotype as 3/3 . Variants with higher numbers of repeats were grouped into 2/3 (n = 1) or 3/3 (n = 2) genotypes as appropriate. Genotyping of MTHFR 677C>T (rs1801133), 1298A>C (rs1801131), and TYMS 3'-UTR 6-bp deletion (rs16430; designation: wt allele, 6; deletion allele, 0) was performed by newly established assays (Appendix). All genotyping assays were performed using appropriate positive and negative controls and 10% of randomly selected samples were routinely reanalyzed for confirmation. Laboratory personnel were blinded to case status of the study participants during the entire genotyping process. All clinicians were unaware of the genotyping results.

Resequencing of the DPYD Gene and Other Methods
All 28 patients with grade 4 and 28 patients with grade 3 toxicity as well as 28 control patients (11, grade 0; two, grade 1; 12, grade 2) were resequenced throughout all exons and exon/intron boundaries. Nucleotide changes apparently associated with grade 3 to 4 toxicity were subsequently genotyped in larger panels. Methods for quantitating DPD protein, activity, and DPYD promoter methylation are described in the Appendix.

Statistical and In Silico Analyses
Univariate analyses were carried out to identify factors predictive for grade 3 to 4 toxicity. For multivariable analysis, several ordinal logistic regression models were considered with toxicity grade as an ordinal response variable increasing from grade 0 to IV (Appendix). The packages SPSS (version 14.0; SPSS Inc, Chicago, IL), and R (version 2.4.1; www.r-project.org) together with the Design library³³ were used for statistical analyses including sample size calculation. Values are presented as median and range. Unpaired t test, Mann-Whitney U test, ² analysis, or Fisher's exact test were applied as appropriate to assess differences or proportions of clinical and genetic data. Odds ratios (OR) are given with 95% CIs and two-sided P values. Statistical significance was defined as P < .05. Observed and expected allele and genotype frequencies within populations were compared by means of Hardy-Weinberg equilibrium calculations (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl). Additional details including sample size calculation are available in the Appendix.

	RESULTS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Overall Toxicity and Nongenetic Factors
Demographic and clinical characteristics of the 683 patients fulfilling the inclusion criteria are listed in Appendix Table A2 (online only). The frequency of grade 3 to 4 toxicity (diarrhea, mucositis, or leucopenia) was 16.1% (Table 1; see Appendix Table A3 [online only] for other toxicities). None of the patients was reported to have died as a consequence of FU-related toxicity. In agreement with previous studies, patients in the severe toxicity group tended to be older as compared with those in the low toxicity group, although this was not significant (Appendix Table A2). The average creatinine level was slightly higher in male patients of the severe toxicity group as compared with those in the low toxicity group, but creatinine levels above the clinically relevant threshold (1.2 mg/dL) were not predictive for severe toxicity. Of the patients treated with bolus Mayo regimen, 24.4% developed severe toxicity compared with only 11.7% of those who received high-dose infusion (OR, 2.44; 95% CI, 1.52 to 3.91; P < .001; Table 1). Overall, 63 female patients (21.0%) but only 47 male patients (12.3%) developed grade 3 to 4 toxicity (OR, 1.9; 95% CI, 1.26 to 2.87; P = .002; Table 1; Fig 1A). Analysis of individual toxicity types revealed that this was mainly due to an increased risk of female patients to develop mucositis (OR, 2.37; 95% CI, 1.32 to 4.26; Table 2). Folinic acid, obligatory in all treatment regimens except in no. 4 and no. 6 (Appendix Table A1), was of borderline significance by univariate analysis (OR, 2.55; 95% CI, 1.0 to 6.5; P = .052).

View larger version (21K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 1. Frequency distribution of sex and DPYD*2A allele among all 683 patients receiving fluorouracil monotherapy according to their toxicity grade (TOX). (A) Distribution of sex in percentage; (B) distribution of DPYD*2A allele.

By fitting a sequence of ordinal logistic regression models, it was shown that sex; DPYD, TYMS VNTR, and MTHFR genotypes; mode of application; and presence of folinic acid are independent and statistically significant prognostic factors for toxicity (Table 4 and Appendix Table A5, online only). For instance, TYMS 2/2 genotype increased the risk for toxicity 1.56-fold (95% CI, 1.08 to 2.27; P = .018). The strong DPYD genotype/sex interaction (P = .0009) was confirmed in the multivariable analysis, resulting in an OR for toxicity of 41.8 (95% CI, 9.2 to 190; P < .0001; Table 4) in men but only 1.33 (95% CI, 0.34 to 5.16; P = .68) in women. No interaction between sex and TYMS VNTR or MTHFR 677C>T genotypes was found by multivariable analysis. Furthermore, multivariable analysis of individual toxicity types revealed a sex-dependent influence of DPYD genotype on all toxicity types, whereas TYMS was confirmed to affect diarrhea only (data not shown). The final proportional odds model was used to develop a nomogram that predicts the probability for toxicity in a given patient (Fig 2).

View larger version (17K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig 2. Nomogram for estimating individual fluorouracil (FU) toxicity risk based on the multifactorial model. The nomogram was developed on the basis of the final logistic regression model (Table 4 and Appendix, Statistical Analysis). It allows one to calculate the probability of experiencing toxicity under FU monotherapy for a given patient. For each factor on the left of (A), the appropriate number of points is determined on the upper scale. The total sum of points is transformed into the estimated probability (B) for toxicity grade 2, 3, or 4. mt, mutant; wt, wild type.

Influence of Sex and Promoter Methylation on DPD Expression in Human Liver
Because previous work has indicated sex-dependent differences of FU clearance^17,18,34 and of DPD in lymphocytes,^7,8 we analyzed DPD protein expression and enzymatic activity in 82 human liver samples in relation to their DPYD*2A genotype (Appendix Fig A2). Three heterozygous carriers of DPYD*2A showed 1.66-fold reduced DPD-protein content (P = .04) and 1.89-fold reduced enzyme activity (P = .014) as compared with noncarriers, consistent with a gene dose effect of DPYD genotype on phenotype. However, there was no difference between the livers of male and female patients with respect to enzyme activity and DPD protein content.

To investigate the possible role of promoter methylation as an epigenetic factor influencing DPYD, we prepared genomic DNA from the same human liver samples. We did not find evidence for methylation of the DPYD promoter in any of these liver DNAs. We also investigated DPYD promoter methylation in genomic DNA isolated from blood of the patients with grade 4 toxicity. In agreement with the liver analysis, no evidence for methylation was found (Appendix Fig A3).

	DISCUSSION

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
This prospective large-scale clinical study on candidate gene polymorphisms in DPYD, TYMS, and MTHFR for FU toxicity risk revealed a number of important results. First, heterozygosity for the defective DPYD*2A allele was found to be a risk factor for severe mucositis and leucopenia but was not associated with diarrhea. Interestingly, the TYMS VNTR polymorphism was associated with a protective effect against diarrhea, showing for the first time a differentiating potential of these polymorphisms with respect to the major toxicity types. Second, the predictive value of all polymorphisms significantly associated with toxicity was rather limited. Thus the sensitivity of the DPYD*2A test was only 7.7% and 13% for mucositis and leucopenia, respectively, whereas TYMS VNTR genotyping had higher sensitivity to predict diarrhea (57%) but lacked specificity (Table 3 and Appendix Table A4). Third, additional unexpected factors strongly modulate FU toxicity risk, in particular, an interaction between DPYD genetics and patients' sex, as well as folinic acid coadministration.

With only six of 110 patients with grade 3 to 4 toxicity being heterozygous for DPYD*2A, the overall sensitivity of this genetic test was approximately 5%. On the basis of previous retrospective studies, we initially expected to find at least 20% of patients with severe FU-related toxicity being predictable by this single genetic defect.^12,14 These discrepancies may be explained by our strict inclusion criteria requiring treatment with FU as monotherapy, which excluded confounding by other anticancer drugs. Moreover, many retrospective studies lacked power because of limited sample size. The positive predictive value of DPYD*2A genotyping was also limited, because the risk for DPYD*2A carriers to develop severe toxicity (grade 3) was only approximately 50%, confirming recent findings by others.¹⁶ The search for additional mutations by completely sequencing the DPYD gene revealed the novel amino acid variant C953S found in a single patient with grade 4 toxicity. In addition, the known variant D949V was found in three patients with grade 3 to 4 toxicity but also in two patients with low or absent toxicity. Interestingly, the C953S amino acid change is predicted by in silico analysis to disrupt an annotated functional site involved in NADPH binding, and both SNPs are predicted to be functionally damaging. Taken together, the inclusion of additional DPYD variants did not substantially improve prediction rates.

Our data on TYMS polymorphism are in agreement with previously published smaller studies^23,25 confirming a significant inverse association of the number of 28-bp tandem repeats in the TYMS promoter region and the severity of toxicity. However, whereas in the study by Lecomte et al,²⁵ patients with the 2/2 genotype were 20 times more likely to have severe toxicity compared with 3/3 carriers, this effect was much less pronounced in our study. MTHFR 677C>T had previously been correlated with better clinical response to FU-based chemotherapy,^30,31 corroborating the hypothesis that an increase in the MTHFR substrate, 5,10-MTHF, might enhance the formation of a ternary complex between 5,10-MTHF, fluorodeoxyuridylate, and thymidylate synthase.^21,29 The impact of MTHFR polymorphisms on severe FU-related toxicity thus seems to be negligible based on our prospective data.

The novel finding of a strong interaction between DPYD genetics and sex is intriguing. In agreement with others, we observed that female patients have approximately a two-fold higher risk for severe FU-related toxicity as compared with male patients (P = .002).^20,31 However, toxicity in women was independent of DPYD genotype in our study. To our knowledge, gene/sex interactions have not been considered in previous studies on FU toxicity. Our analysis of DPD expression and activity in human liver did not reveal any sex-related differences. In a recent study, Mattison et al,³⁴ using an in vivo breath test, reported a 14% lower overall DPD activity in women compared with men. It should be noted, however, that this difference was also affected by ethnicity, being largest among African Americans and almost absent among white patients, in agreement with our findings. We also investigated the previously suggested possibility that epigenetic effects influence DPD expression via methylation of CpG islands in the DPYD promoter region. However, using the same method as described in the original article,³⁵ we found no indication for DPYD promoter methylation in genomic DNA prepared from either human liver or from blood of patients with severe FU toxicity.

Interestingly, our multivariable analysis identified folinic acid as risk factor for toxicity, which is in line with the observation that the abundant supplementation of the United States diet with this substance is well known to enhance toxicity of fluoropyrimidines. By stabilizing binding of fluoropyrimidines to thymidylate synthase, folinic acid not only improves response rate and overall survival²⁸ but also contributes substantially to cell death of normal tissue.

As an aid for practical clinical use, we developed a nomogram based on the ordinal logistic regression model shown in Fig 2A, which includes both genetic and nongenetic factors to estimate the probability that a given patient will suffer from severe toxicity. For each factor, the appropriate number of points is determined and the total sum of points is transformed into the estimated probability to develop FU toxicity (Fig 2B). In male patients, the maximal number of points is 168, predicting a greater than 90% risk of severe toxicity. In contrast, the maximal number of points achievable in female patients is 94, predicting a risk of approximately 46%. This comparison underscores one of the novel findings of this study, namely that FU toxicity is less predictable in women.

In conclusion, our results demonstrate a limited overall impact of DPYD, TYMS, and MTHFR polymorphisms on the risk for developing FU related toxicity. The data suggest the existence of a strong DPYD gene/sex interaction, and they provide a thorough basis for estimating individual FU toxicity risk, including the most predictive factors known to date. Future approaches using genome-wide association analyses may help to identify additional candidate genes causally involved in the pathomechanism of FU toxicity.

	AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
The author(s) indicated no potential conflicts of interest.

	AUTHOR CONTRIBUTIONS

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
Conception and design: Matthias Schwab, Ulrich M. Zanger, Carsten Bokemeyer, Michel Eichelbaum

Financial support: Matthias Schwab, Ulrich M. Zanger, Michel Eichelbaum

Provision of study materials or patients: Matthias Schwab, Ulrich M. Zanger, Carsten Bokemeyer

Collection and assembly of data: Matthias Schwab, Ulrich M. Zanger, Claudia Marx, Elke Schaeffeler, Kathrin Klein, Reinhold Kerb, Julia Blievernicht, Joachim Fischer, Ute Hofmann

Data analysis and interpretation: Matthias Schwab, Ulrich M. Zanger, Elke Schaeffeler, Kathrin Klein, Jürgen Dippon, Julia Blievernicht, Joachim Fischer, Ute Hofmann, Carsten Bokemeyer, Michel Eichelbaum

Manuscript writing: Matthias Schwab, Ulrich M. Zanger, Michel Eichelbaum

Final approval of manuscript: Matthias Schwab, Ulrich M. Zanger, Claudia Marx, Elke Schaeffeler, Kathrin Klein, Jürgen Dippon, Reinhold Kerb, Julia Blievernicht, Joachim Fischer, Ute Hofmann, Carsten Bokemeyer, Michel Eichelbaum

	Appendix

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
German 5-FU Toxicity Study Group
N. Brüllke, MD, Asklepios Klinik Barmbek Abtl. Onkologie und Palliativmedizin, Hamburg; P. Hesse, MD, Asklepios Klinik Parchim, Klinik für Innere Medizin, Parchim; H.-K. Heck, MD, Caritasklinik St. Theresia, Klinik für Hämatologie und Onkologie, Saarbrücken; R. Mück, MD, and E. Heidemann, MD, Diakonie Klinikum Stuttgart, Innere Medizin II, Stuttgart; H.G. Fuhr, MD, Dr. Horst-Schmidt-Kliniken GmbH, Klinik Innere Medizin III, Wiesbaden; L. Moschner, MD, Franz-Hospital Dülmen GmBH, Innere Medizin, Dülmen; K. Verpoort, MD, and W. Zeller, MD, Gemeinschaftspraxis Hämatologie/Onkologie, Hamburg; S. Hahnfeld, MD, Gemeinschaftspraxis Innere Medizin, Jena; H.-R. Schmitt, MD, Gemeinschaftspraxis, Gerlingen; R. Hecker, MD, Gemeinschaftspraxis, Groβ-Gerau; H. Eschenburg, MD, Gemeinschaftspraxis, Güstrow; G. Hermesdorf, MD, Gemeinschaftspraxis, Koblenz; G. Höring, MD, Gemeinschaftspraxis, Stuttgart; G. Wilhelm, MD, Harz-Klinikum Wernigerode-Blankenburg GmbH, Medizinische Klinik, Abtl. Hämatologie/Onkologie, Wernigerode; A. Rennert, MD, Humaine Vogtland-Klinikum Plauen GmbH, Chirurgische Klinik, Plauen; V. von Paris, MD, Inselberg Klinik, Onkologie, Tabarz; W. Ebert, MD, Klinik Nürtingen, Innere Medizin, Nürtingen; H. Wilke, MD, Kliniken Essen-Mitte, Evang. Huyssens-Stiftung, Klinik für Innere Medizin IV, Essen; W. Brugger, MD, Kliniken Villingen, Klinik für Innere Medizin II, Villingen-Schwenningen; E. Günther, MD, Klinikum am Steinenberg, Medizinische Klinik, Reutlingen; K. Lange, MD, Klinikum Bremen-Mitte, Medizinische Klinik II, Bremen; U. Gerecke, MD, Klinikum der Hansestadt Stralsund GmbH, Medizinische Klinik, Hämatologie und internistische Onkologie, Stralsund; F. Rothmann, MD and R. Pasold, MD, Klinikum Ernst von Bergmann, Medizinische Klinik Hämatologie/Onkologie, Potsdam; J. Gatter, MD, Klinikum Kempten-Oberallgäu gGmbH, Innere Medizin III, Kempten; K.-T. Steurer, MD, Klinikum Nürnberg Nord, Medizinische Klinik 6, Nürnberg; A. Jakob, MD and F. Hirsch, MD, Klinikum Offenburg, Medizin II Hämatologie und Onkologie, Offenburg; U. Ohl, MD, Klinikum Pforzheim, Medizinische Klinik, Hämotologie/Onkologie, Pforzheim; J. Schleicher, MD, and S. Hiller, MD, Klinikum Stuttgart Katharinenhospital, Klinik für Onkologie mit Tagesklinik, Stuttgart; C. Balzer, MD, Klinikum Weiden, Medizinische Klinik I, Weiden; G.A. Dietrich, MD, Krankenhaus Bietigheim, Innere Medizin I, Bietigheim-Bissingen; M. Schenk, MD, Krankenhaus der Barmherzigen Brüder, Onkologie und Hämatologie, Regensburg; A. Knuth, MD, Krankenhaus Nordwest, II. Medizinische Klinik, Frankfurt/Main; B. Swarorsky, MD, Kreisklinik Albstadt, Klinik für Innere Medizin, Albstadt; A. Ohmenhäuser, MD, Kreiskrankenhaus Böblingen, Allgemeine Innere Medizin, Böblingen; M. Buntrock, MD, Kreiskrankenhaus Böblingen, Klinik für Allgemein- und Viszeralchirurgie, Böblingen; A. Wolf, MD, Kreiskrankenhaus Herrenberg, Medizinische Klinik, Abtl. Gastroenterologie, Herrenberg; K. Huntenburg, MD, Kreiskrankenhaus Neustadt, Medizinische Klinik, Neustadt am Rübenberge; G. Käfer, MD, Kreiskrankenhaus Sigmaringen, Innere Medizin, Sigmaringen; L. Moschner, MD, Marienhospital Herne, Medizinische Klinik III, Herne; W. Abenhardt, MD, and F.-J. Tigges, MD, Münchner Onkologische Gemeinschaftspraxis, München; G. Springer, MD, Onkologische Praxis, Stuttgart; U. Abele, MD, Paracelsus Krankenhaus Ruit, Medizinische Klinik, Ostfildern; W. Aulitzky, MD, Robert Bosch Krankenhaus, Hämatologie, Onkologie und Klinische Immunologie, Stuttgart; H. Dietzfelbinger, MD, Schwerpunktspraxis Hämatologie und Onkologie, Herrsching; J. Papke, MD, Schwerpunktspraxis, Neustadt; S. Müller-Hagen, MD, Schwerpunktsspraxis Onkologie, Hamburg; R. Fuchs, MD, St. Antonius-Hospital, Hämatologie und Onkologie, Eschweiler; A. Bremer, MD, St. Franziskus Hospital, Innere Medizin II, Münster; C. Müller-Naendrup, MD, St. Martinus-Hospital, Medizinische Klinik, Olpe; B. Koch, MD and R. Grün, MD, St. Vincenz-Krankenhaus, Innere Medizin-Gastroenterologie, Datteln; A. Pies, MD, Städtisches Klinikum Braunschweig, Medizinische Klinik III, Braunschweig; T. Kubin, MD, Städtisches Klinikum Karlsruhe gGmbH, Medizinische Klinik II, Karlsruhe; B. Bilsing, MD, Städtisches Klinikum Magdeburg, Hämatologie, Onkologie und Tagesklinik für Onkologie, Magdeburg; J Bednarczyk, MD, and D. Berger, MD, Stadtklinik Baden-Baden, Klinik für Viszeral-, Gefäss- und Kinderchirurgie, Baden-Baden; S. Werle, MD, Theresienkrankenhaus und St. Hedwig Klinik GmbHç Gastroenterologie, gastroenterologische Onkologie und Diabetologie, Mannheim; H. Scherübl, MD, Universitätsklinikum Benjamin Franklin, Medizinische Klinik I, Berlin; F. Hartmann, MD, Universitätsklinikum des Saarlandes, Innere Medizin I, Homburg/Saar; A. Wein, MD, Universitätsklinikum Erlangen, Innere Medizin I, Erlangen; S. Seeber, MD, and U. Vanhoefer, MD, Universitätsklinikum Essen, Innere Klinik-Tumorforschung, Essen; P. Meusers, MD, Universitätsklinikum Essen, Klinik und Poliklinik für Strahlentherapie, Essen; M. Lorenz, MD, Universitätsklinikum Frankfurt/Main, Zentrum für Chirurgie, Allgemein- und Gefäβchirurgie; Frankfurt/Main; W. Kassahun, MD, Universitätsklinikum Leipzig, Chirurgische Klinik und Poliklinik II; Leipzig; J. Fahlke, MD, Universitätsklinikum Magdeburg, Klinik für Allgemein-, Viszeral- u. Gefäβchirurgie, Magdeburg; F. Honecker, MD, Universitätsklinikum Tübingen, Innere Medizin II, Tübingen; S. Hendler, MD and M. Lutz, MD, Universitätsklinikum Ulm, Innere Medizin I, Ulm; H. Rückle-Lanz, MD, Universitätsklinikum Würzburg, Medizinische Klinik und Poliklinik II, Würzburg; C. Wollermann, MD, Westpfalz-Klinikum GmbH, Medizinische Klinik I, Kaiserslautern.

PATIENTS AND METHODS
Study Population and Design A total of 796 patients were initially recruited between March 2000 and July 2003. At this time, fluorouracil (FU) monotherapy was commonly used in clinical practice as standard treatment regimen. Monitoring of the patients' data files revealed that 80 of these patients had received additional chemotherapy (eg, irinotecan, oxaliplatin). An additional 29 patients were excluded because of incomplete documentation, and four patients were excluded because they were retrospectively reported toxicity cases. Demographic and clinical characteristics of the 683 patients fulfilling the inclusion criteria, and the different FU treatment regimens, are listed in Appendix Tables A1 and A2. The toxicity grading scale can be briefly described as follows: grade 0 indicates no adverse effects, grade 1 indicates minor effects, grade 2 indicates moderate effects, grade 3 indicates severe effects, and grade 4 indicates life-threatening effects. Once toxicity was observed, the subsequent decision of whether FU treatment had to be discontinued or whether the dosage should be reduced was independent of the study. Classification as toxicity (TOX) grade 0 required an observation time of at least one completed course of chemotherapy (median duration, two cycles; range, 1 to 26 cycles). For each patient, the cumulative dose of FU was also recorded.

Genotyping for MTHFR 677C>T, 1298A>C, and TYMS 3'UTR 6-bp Deletion Genotyping of MTHFR 677C>T (rs 1801133) was performed by a newly established 5' nuclease assay by TaqMan technology with primers and probes designed using the Applied Biosystems (Foster City, CA) primer express program (version 1.5). TaqMan MGB probes were custom synthesized by Applied Biosystems: VIC-TGAAATCGACTCCCG, FAM-AAATCGGCTCCCGC. Primers were synthesized by MWG (MWG-Biotech AG, Ebersberg, Germany): 5'-AAGCACTTGAAGGAGAAGGTGTCT-3; 5'-CACAAAGCGGAAGAATGTGTCA-3'. Polymerase chain reaction (PCR) was performed in 25 µL with 20 ng of genomic DNA, 200 nmol/L of each probe, and 900 nmol/L of forward and reverse primers in 1x Universal PCR Master Mix (Applied Biosystems). Amplification conditions were as follows: one cycle of 50°C for 2 minutes, one cycle of 95°C for 10 minutes, and 40 cycles each of 92°C for 15 seconds and 60°C for 1 minute. Detection of fluorescence signals was performed using the ABI PRISM 7700 detection system (Applied Biosystems), and the results were analyzed by use of the Sequence Detection System Software Version 1.7 (Applied Biosystems). Each TaqMan run comprised four DNA samples homozygous for allele 1 (AL1), four samples homozygous for allele 2 (AL2),and four reactions in which no DNA template or allelic reference was included.

Genotyping of MTHFR 1298A>C (rs1801131) and TYMS-3'UTR 6-bp deletion (rs16430) was performed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry using the MassARRAY Compact system (Sequenom, San Diego, CA). Briefly, 5 ng of genomic DNA was amplified by PCR in a final volume of 6 µL containing specific primers (5'-ACGTTGGATGAGGAGCTGCTGAAGATGTGG-3', 5'-ACGTTGGATGTCTCCCGAGAGGTAAAGAAC-3'; 5'-ACGTTGGATGCAAATCTGAGGGAGCTGAGT-3', 5'-ACGTTGGATGCAGATAAGTGGCAGTACAGA-3') at 167 nmol/L final concentration and 0.1 unit HotStarTaq DNA polymerase (Qiagen, Hilden, Germany). PCR conditions were 95°C for 15 minutes for hot start followed by 44 cycles of denaturation at 95°C for 30 seconds, annealing at 56°C for 30 seconds, and extension for 1 minute at 72°C, followed by final extension at 72°C for 10 minutes. PCR products were treated with shrimp alkaline phosphatase (SAP; Amersham, Freiburg, Germany) for 20 minutes at 37°C to remove excess deoxynucleotide triphosphates, followed by 10 minutes at 85°C to inactivate SAP. Base extension (homogenous MassEXTEND Assay, Sequenom) reactions contained in a final volume of 10 µL the two extension primers 5'- GAGCTGACCAGTGAAG-3', and 5'-AGAGTGTGGTTATGAACTTTA-3'. Base extension reaction conditions were 94°C for 2 minutes followed by 40 cycles of 94°C for 5 seconds, 52°C for 5 seconds, and 72°C for 5 seconds. All reactions including PCR amplification, SAP treatment, and base extension were performed in a PTC 220Dyad PCR thermal cycler (MJ Research, Waltham, MA). The final base extension products were treated with SpectroCLEAN resin (Sequenom) to remove salts from the reaction mixture. For a final volume of 26 µL, 16 µL of resin-water suspension was added into each base extension reaction. After a quick centrifugation (2,000 rpm for 3 minutes), samples were dispensed onto a 384-format SpectroCHIP microarray (Sequenom) by using a SpectroPoint nanodispenser (Sequenom). The MassARRAY Compact System was used for data acquisition, and genotyping calls were made in real time with Mass Array RT software version 3.3.1.3 (Sequenom).

Resequencing of the DPYD Gene Primers for amplification and sequencing of the exons and adjacent intronic regions were used as previously described.^1,2 DNA sequencing was performed using the PE ABI 3700DNA Analyzer (ABI, Weiterstadt, Germany) or the LI-COR model 4000 DNA sequencer (LI-COR, Lincoln, NE) and in part by SeqLab Sequence Laboratories (Göttingen GmbH). Sequences were analyzed using the PHRED/PHRAPPOLYPHRED/consed software package (University of Washington, Seattle, WA) or Vector NTI 9.0 (Invitrogen, Karlsruhe, Germany) and inspected for deviations from the cDNA reference sequence (XM056194). All primers for PCR, sequencing, and genotyping were purchased at MWG Biotech AG (Ebersbach, Germany).

Genotyping for Rare and Newly Identified Single Nucleotide Polymorphisms Previously described denaturing high-performance liquid chromatography (HPLC) methods² were used for genotyping of single nucleotide polymorphisms (SNPs) in exon 13 (*4: 1601G>A, rs 1801158; *5: 1627A>G, rs 1801159) and in exon 18 (*6: 2194G>A, rs1801160). Samples showing variant profiles were sequenced to differentiate *4 and *5 in exon 13 fragment and to confirm the *6-mutation in exon 18.

For genotyping 623G>A, 1236G>A, 2858G>C, and 2846A>T, two MALDI-TOF–based assays were developed. Specific PCR primer pairs were designed for the amplification of exon 6 (5'-ACGTTGGATGCTCCTCATCTACTTGACACA-3'; 5'-ACGTTGGATGGAGCCTGAAGTTCCTATATG-3'), exon 11 (5'-ACGTTGGATGCAACTCCAATTCAAGACTCC-3'; 5'-ACGTTGGATGCCAACTACTCTGCCTCATTT-3') and exon 22 (5'-ACGTTGGATGACTATCCAGTCTCCCAAGTT-3'; 5'-ACGTTGGATGCAATATTTGGCACCACTGGT-3') using NC_000001.8 as reference sequence and Vector NTI Advance 9.0 software (Invitrogen, Carlsbad, CA). PCR reactions were performed with 50 ng of genomic DNA, 0.4 Units of HotStarTaq Master Mix Kit, and 0.1 mmol/L of magnesium chloride (Qiagen) in a 8-µL volume. Primer concentrations were 0.3 µmol/L (exon 6, exon 22) and 0.325 µmol/L (exon 11). PCR conditions were 95°C for 15 minutes; 45 cycles at 95°C for 30 seconds, 61°C for 1 minute, and 72°C for 1 minute; and a final step at 72°C for 10 minutes. After removal of excess dNTP with 0.06 units of SAP at 37°C for 20 minutes, 85°C for 10 minutes, and 20°C for 1 second, the PCR products were used as templates for allele-specific primer extension reactions as follows. One µL of buffer C, 6 mmol/L of magnesium chloride (Solis Biodyne, Tartu, Estonia), and 0.8 µmol/L of the extension primer were mixed with the PCR product in a 16-µL volume. SNPs 623G>A (primer: 5'-GTGATGTCAGAGTACCCCAAT-3'), 1236G>A (primer: 5'-TTCCATTTTCCAGTTTCATCTTG-3'), and 2858G>C (primer: 5'-CATTTACCACAGTTGATA-3') were detected simultaneously in one assay, and SNP 2846A>T (primer: 5'-AGCAAGTTGTGGCTATGATTG-3') was detected separately. A total of 0.1 mmol/L of dNTPs (Amersham) and ddNTPs (BioLog Life Science Institute, Bremen, Germany) were added in the particular assay-specific composition. Extension reactions were performed at 94°C for 4 minutes, followed by 55 cycles with 94°C for 30 seconds, 52°C for 30 seconds and 72°C for 30 seconds, and finally 72°C for 2 minutes using 0.01 U of Termipol DNA Polymerase I (Solis Biodyne). The resulting nucleotide extension products were treated with a cationic exchange resin (AG 50W-X8 Resin, Biorad Laboratories, Inc, CA) for 30 minutes. For the preparation of the MALDI target, 1 µL of a 3-hydroxypicolinic acid matrix (10 mg/mL in 10 µL diammoniumhydrogencitrate) was spotted on a 384-format Anchor Chip TM target plate (size, 400 µm; Bruker Daltonic, Bremen, Germany) by a pipette robot (Cybio, Jena, Germany), and 0.5 µL of reaction solution was dispensed on the matrix, air dried, and introduced in the vacuum of the Ultraflex MALDI-TOF mass spectrometer (Bruker Daltonic). One hundred eighty nitrogen laser shots were summarized for each sample. The mass range of the particular allele-specific peaks was set between 4000 and 9000 Da. Genotyping calls were performed using GENOTOOLS software (Bruker Daltonic).

Quantitation of DPD Protein Content and Metabolic Capacity in Human Liver Liver tissue samples used in this study had been previously collected from patients of white European origin undergoing liver surgery for various reasons at the Department of General, Visceral, and Transplantation Surgery, Campus Virchow, University Medical Center Charité, Humboldt University in Berlin, Germany (provided by Drs Natascha Nüssler and Andreas K. Nüssler). Only normal liver tissue was collected and stored at –80°C, and absence of tumorous material had been confirmed by histologic analysis as previously described.³ The study was approved by the ethics committee of the Medical Faculty of the Charité, Humboldt-University, Berlin, and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from each patient. Hepatic microsomal and cytosolic protein fractions were prepared according to standard procedures as described.⁴ Genomic DNA was isolated from blood samples of the same patients using QIAamp DNA Blood Mini Kit (Qiagen) and genotyped for DPYD*2A using a denaturing HPLC method described.² DPD protein (approximately100 kDa) was quantified by Western blotting, after separation of 50 µg of cytosolic protein by sodium dodecyl sulfate polyacrylamide gel eletrophoresis on a 9% separation gel using a polyclonal rabbit antipig DPD antibody (provided by Frank J. Gonzalez, National Cancer Institute, Bethesda, MD). Appropriate secondary antibody was used in combination with SuperSignal West Dura Extended Duration Substrate (both Pierce Biotechnology, Rockford, IL) for chemiluminescence detection with a CCD camera (Raytest, Straubenhard, Germany) and LAS-1000. Relative amounts of DPD protein were determined using a dilution series of one standard sample applied to each gel.

For the quantification of DPD enzyme activity, 50 µmol/L thymine was used as substrate. To 100 µg of cytosolic protein in a final volume of 150 µL of 35 mmol/L of sodium phosphate buffer, pH 7.4, and 2.5 mmol/L of magnesium chloride, 10 mmol/L of β-mercaptoethanol, and 250 µmol/L NADPH were added just before preincubating the sample for 5 minutes at 37°C. The reaction was started with substrate and stopped after 10 minutes at 37°C by adding 300 µL of saturated ammonium sulfate and internal standard⁵ and extracted with 600 µL of ethylacetate. After centrifugation, 500 µL of the supernatant was evaporated to dryness in a stream of nitrogen and the samples were derivatized as described. The dihydrothymine formation was analyzed by a gas chromatographic tandem mass spectrometric method as described.⁵ The specific activity was calculated as pmol dihydrothymine per minute and milligrams cytosolic protein.

Analysis of DPYD Promoter Methylation by Denaturing HPLC Genomic DNAs were bisulfite-modified using the EZ DNA Methylation-Gold Kit (Zymo Research, CA). As a methylation positive control, CpGenome Universal methylated DNA (Chemicon, Hofheim, Germany) was used. PCR amplification was performed as described previously.⁶ The primers (forward, 5'-TTTTTGTTTGTAGGTTGGG-3'; and reverse, 5'-CAACCAAAAAACCAAATAACAACAA-3') amplify the same sequences from methylated and unmethylated bisulfite-modified DNA and generate a 209-bp fragment of the DPYD promoter. PCR was performed in a 50-µL volume, using Hot Star Taq Master Mix (Qiagen), 20 ng of bisulfite-modified genomic DNA, and 10 nmol of forward and reverse primer. The PCR was performed on a PTC200 DNA engine (MJ Research): 95°C for 10 minutes followed by 40 cycles of 94°C for 50 seconds, 52°C for 50 seconds, 72°C for 1 minute, and a final extension at 72°C for 10 minutes. Amplified PCR products were analyzed by denaturing HPLC to detect the methylation status as described by Ezzeldin et al⁶ on a Wave System (Transgenomic, Omaha, NE). In brief, samples were analyzed using a 6.6-minutes gradient with 0.9 mL/min flow rate and a gradient temperature of 57°C (46.8% buffer B, 0.5 minutes; 51.8% buffer B [0.1 mol/L of TEAA, 25% acetonitrile] and 4.5 minutes, 60.8% buffer B). Denaturing HPLC results of selected samples were confirmed by sequencing on an ABI Prism 310 genetic analyzer using Big Dye terminator (Applied Biosystems).

Statistical Analysis and In Silico Methods The sample size calculation was originally designed to investigate the relationship between the DPYD*2A allele and FU toxicity. To compare the number (N) of cases to be included in this observational study, we used the fact that severe FU toxicity occurs in approximately 15% of patients.⁷ Furthermore, it is known that DPYD*2A heterozygosity is present in approximately 1% of white patients.⁸ We supposed that a diagnostic test for severe toxicity based on DPYD*2A cannot be recommended if its positive predictive value is less than 80%. To ensure that severe toxicity in a patient with the DPYD*2A allele will be predicted correctly in 80% of cases, a two-sample binomial test on difference in proportions would require a sample size of at least N = 643 to reach a statistical power of 0.9 by a significance level of .05.

The ordinal logistic regression models were computed under the proportional odds assumption, which is appropriate here. This amounts to fit several binary logistic regression models simultaneously with common regression coefficients but possibly different intercepts that depend on whether the TOX classification is j (j = 0, 1, 2, 3, 4).⁹ A sequence of models was adjusted for the following candidate parameters: genotypes with respect to the DPYD*2A allele (DPYD), the TYMS enhancer region polymorphism (TYMS), and the MTHFR 677C>T polymorphism (MTHFR), sex (SEX) of the patient, the type of tumor (TUMOR; ie, colorectal v other), the mode of FU administration either as bolus or as continuous intravenous infusion (MOD), and finally, whether a combination of FU with folinic acid (FOLIN) was given or not.

The final ordinal logistic regression model is represented by

(1)

where

(2)

and

(3)

with the convention that {SEX = f} has to be replaced by 1 if the patient considered is female, and {SEX = f} equals 0 if the condition within braces is false; the same logic applies to the remaining strings using DPYD = mt (mutant) in presence of at least one DPYD*2A allele; MTHFR = mt, genotype 677C/T or T/T; TYMS = mt, enhancer regions double repeat variant genotype 2/3 or 3/3; and FOLIN = yes, regimen with folinic acid; MOD = inf, continuous infusion.

Prediction of tolerability of amino acid variations was calculated using PolyPhen (http://www.bork.embl-heidelberg.de/PolyPhen). Linkage disequilibrium was analyzed using Haploview (http://www.broad.mit.edu/mpg/haploview/index.php; version 4.0).

REFERENCES 1 Wei X, McLeod HL, McMurrough J, et al: Molecular basis of the human dihydropyrimidine dehydrogenase deficiency and 5-fluorouracil toxicity. J Clin Invest 98:610-615, 1996

2 Fischer J, Schwab M, Eichelbaum M, et al: Mutational analysis of the human dihydropyrimidine dehydrogenase gene by denaturing high-performance liquid chromatography. Genet Test 7:97-105, 2003

3 Wolbold R, Klein K, Burk O, et al: Sex is a major determinant of CYP3A4 expression in human liver. Hepatology 38:978-988, 2003

4 Lang T, Klein K, Fischer J, et al: Extensive genetic polymorphism in the human CYP2B6 gene with impact on expression and function in human liver. Pharmacogenet Genomics 11:399-415, 2001

5 Hofmann U, Schwab M, Seefried S, et al: Sensitive method for the quantification of urinary pyrimidine metabolites in healthy adults by gas chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 791:371-380, 2003

6 Ezzeldin HH, Lee AM, Mattison LK, et al: Methylation of the DPYD promoter: An alternative mechanism for dihydropyrimidine dehydrogenase deficiency in cancer patients. Clin Cancer Res 11:8699-8705, 2005

7 Toxicity of fluorouracil in patients with advanced colorectal cancer: Effect of administration schedule and prognostic factors—Meta-Analysis Group In Cancer. J Clin Oncol 16:3537-3541, 1998

8 Raida M, Schwabe W, Hausler P, et al: Prevalence of a common point mutation in the dihydropyrimidine dehydrogenase (DPD) gene within the 5'-splice donor site of intron 14 in patients with severe 5-fluorouracil (5-FU)- related toxicity compared with controls. Clin Cancer Res 7:2832-2839, 2001

9 Harrell FE: Regression Modeling Strategies, with Applications to Linear Models, Survival Analysis and Logistic Regression. New York, NY, Springer, 2001

10 Mattison LK, Soong R, Diasio RB: Implications of dihydropyrimidine dehydrogenase on 5-fluorouracil pharmacogenetics and pharmacogenomics. Pharmacogenomics 3:485-492, 2002

11 van Kuilenburg AB, Haasjes J, Richel DJ, et al: Clinical implications of dihydropyrimidine dehydrogenase (DPD) deficiency in patients with severe 5-fluorouracil-associated toxicity: identification of new mutations in the DPD gene. Clin Cancer Res 6:4705-4712, 2000

12 Collie-Duguid ES, McLeod HL, Cassidy J: Estimation of dihydropyrimidine dehydrogenase activity: does it have a role in cancer therapy? Ann Oncol 11:255-257, 2000

13 Soong R, Diasio RB: Advances and challenges in fluoropyrimidine pharmacogenomics and pharmacogenetics. Pharmacogenomics 6: 835-847, 2005

View larger version (35K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A1. Pathways of fluoropyrimidine and folate metabolism. Schematic overview of the pathways of fluoropyrimidine and folate metabolism (adapted from Soong and Diasio13). Abbreviations: 5-FU, fluorouracil; 5'dFCR, 5'-deoxy-5-fluorocytidine; 5'dFUR, 5'-deoxy-5-fluorouridine; DNApol, DNA polymerase; DPD, dihydropyrimidine dehydrogenase; dTMP, deoxythymidine monophosphate; dUMP, deoxyuridine monophosphate; FBAL, fluoro-beta-alanine; FDHU, fluoro-dihydrouracil; FdUDP, fluoro-deoxyuridine diphosphate; FdUrd, fluoro-deoxyuridine; FdUTP, fluoro-deoxyuridine triphosphate; FUrd, fluoro-uridine; FUTP, fluoro-uridine triphosphate; Met, methionine; MS, methionine synthase; MTHFR, methylene tetrahydrofolate reductase; RNApol, RNA polymerase; THF, tetrahydrofolate; TS, thymidylate synthase.

View larger version (16K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A2. Influence of sex and DPYD genotype on dihydropyrimidine dehydrogenase (DPD) protein expression and metabolic capacity in human liver. DPD protein content (A, C; in relative units) was determined by immunoblot analysis and dihydrothymine formation (B, D; in pmol formed per minute per milligram of cytosolic protein) was determined mass-spectrometrically in cytosol fractions prepared from 82 surgical human liver samples. All liver donors were genotyped for DPYD*2A. Dihydrothymine formation and DPD protein content varied 4.8 fold (55.7 to 272 pmol/min*mg) and 8.2-fold (10.6 to 87.1 rel U), respectively, among all livers analyzed.

View larger version (9K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Fig A3. DPYD Promoter methylation analysis. CpG methylation analysis at the DPYD promoter was performed as described previously by Ezzeldin et al6 on genomic DNA extracted from 76 human livers and from 28 patients with fluorouracil (FU) -related grade 4 toxicity. (A) Two-peak DHPLC profile of a methylated positive control DNA (UM, unmethylated signal; M, methylated signal); (B) a typical chromatogram of genomic DNA obtained from a patient with FU toxicity, showing absence of methylation; (C) superimposed tracings of liver DNAs.

	ACKNOWLEDGMENTS

 
We thank the patients who participated and the oncologists of the German 5-FU Toxicity Study Group (see Appendix). We also thank Igor Liebermann and Andrea Zwicker for excellent technical assistance.

	NOTES

 
published online ahead of print at www.jco.org on February 25, 2008.

Supported by Grant No. 01 GG 9846 from the German Federal Ministry of Education and Science and by the Robert Bosch Foundation, Stuttgart, Germany.

Presented in part at the 1st Joint Cold Spring Harbor Laboratory/Wellcome Trust Conference on Pharmacogenomics, September 24-28, 2003, Hinxton, United Kingdom.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION PATIENTS AND METHODS RESULTS DISCUSSION AUTHORS' DISCLOSURES OF... AUTHOR CONTRIBUTIONS Appendix REFERENCES

 
1. Lamont EB, Schilsky RL: The oral fluoropyrimidines in cancer chemotherapy. Clin Cancer Res 5:2289-2296, 1999[Abstract/Free Full Text]

2. Meyerhardt JA, Mayer RJ: Systemic therapy for colorectal cancer. N Engl J Med 352:476-487, 2005[Free Full Text]

3. Rich TA, Shepard RC, Mosley ST: Four decades of continuing innovation with fluorouracil: Current and future approaches to fluorouracil chemoradiation therapy. J Clin Oncol 22:2214-2232, 2004[Abstract/Free Full Text]

4. Toxicity of fluorouracil in patients with advanced colorectal cancer: Effect of administration schedule and prognostic factors—Meta-Analysis Group in Cancer. J Clin Oncol 16:3537-3541, 1998[Abstract]

5. Heggie GD, Sommadossi JP, Cross DS, et al: Clinical pharmacokinetics of 5-fluorouracil and its metabolites in plasma, urine, and bile. Cancer Res 7:2203-2206, 1987

6. Diasio RB, Beavers TL, Carpenter JT: Familial deficiency of dihydropyrimidine dehydrogenase. Biochemical basis for familial pyrimidinemia and severe 5-fluorouracil-induced toxicity. J Clin Invest 81:47-51, 1988[Medline]

7. Etienne MC, Lagrange JL, Dassonville O, et al: Population study of dihydropyrimidine dehydrogenase in cancer patients. J Clin Oncol 12:2248-2253, 1994[Abstract/Free Full Text]

8. Lu Z, Zhang R, Diasio RB: Dihydropyrimidine dehydrogenase activity in human peripheral blood mononuclear cells and liver: Population characteristics, newly identified deficient patients, and clinical implication in 5-fluorouracil chemotherapy. Cancer Res 53:5433-5438, 1993[Abstract/Free Full Text]

9. Fleming RA, Milano G, Thyss A, et al: Correlation between dihydropyrimidine dehydrogenase activity in peripheral mononuclear cells and systemic clearance of fluorouracil in cancer patients. Cancer Res 52:2899-2902, 1992[Abstract/Free Full Text]

10. Okuda H, Nishiyama T, Ogura K, et al: Lethal drug interactions of sorivudine, a new antiviral drug, with oral 5-fluorouracil prodrugs. Drug Metab Dispos 25:270-273, 1997[Abstract/Free Full Text]

11. Johnson MR, Wang K, Diasio RB: Profound dihydropyrimidine dehydrogenase deficiency resulting from a novel compound heterozygote genotype. Clin Cancer Res 8:768-774, 2002[Abstract/Free Full Text]

12. Raida M, Schwabe W, Hausler P, et al: Prevalence of a common point mutation in the dihydropyrimidine dehydrogenase (DPD) gene within the 5'-splice donor site of intron 14 in patients with severe 5-fluorouracil (5-FU)- related toxicity compared with controls. Clin Cancer Res 7:2832-2839, 2001[Abstract/Free Full Text]

13. Soong R, Diasio RB: Advances and challenges in fluoropyrimidine pharmacogenomics and pharmacogenetics. Pharmacogenomics 6:835-847, 2005[CrossRef][Medline]

14. Van Kuilenburg AB, Meinsma R, Zoetekouw L, et al: Increased risk of grade IV neutropenia after administration of 5-fluorouracil due to a dihydropyrimidine dehydrogenase deficiency: High prevalence of the IVS14+1g>a mutation. Int J Cancer 101:253-258, 2002[CrossRef][Medline]

15. Wei X, McLeod HL, McMurrough J, et al: Molecular basis of the human dihydropyrimidine dehydrogenase deficiency and 5-fluorouracil toxicity. J Clin Invest 98:610-615, 1996[Medline]

16. Morel A, Boisdron-Celle M, Fey L, et al: Clinical relevance of different dihydropyrimidine dehydrogenase gene single nucleotide polymorphisms on 5-fluorouracil tolerance. Mol Cancer Ther 5:2895-2904, 2006[Abstract/Free Full Text]

17. Milano G, Etienne MC, Cassuto-Viguier E, et al: Influence of sex and age on fluorouracil clearance. J Clin Oncol 10:1171-1175, 1992[Abstract]

18. Port RE, Daniel B, Ding RW, et al: Relative importance of dose, body surface area, sex, and age for 5-fluorouracil clearance. Oncology 48:277-281, 1991[Medline]

19. Chansky K, Benedetti J, Macdonald JS: Differences in toxicity between men and women treated with 5-fluorouracil therapy for colorectal carcinoma. Cancer 103:1165-1171, 2005[CrossRef][Medline]

20. Sloan JA, Goldberg RM, Sargent DJ, et al: Women experience greater toxicity with fluorouracil-based chemotherapy for colorectal cancer. J Clin Oncol 20:1491-1498, 2002[Abstract/Free Full Text]

21. Robien K, Boynton A, Ulrich CM: Pharmacogenetics of folate-related drug targets in cancer treatment. Pharmacogenomics 6:673-689, 2005[CrossRef][Medline]

22. Horie N, Aiba H, Oguro K, et al: Functional analysis and DNA polymorphism of the tandemly repeated sequences in the 5'-terminal regulatory region of the human gene for thymidylate synthase. Cell Struct Funct 20:191-197, 1995[Medline]

23. Pullarkat ST, Stoehlmacher J, Ghaderi V, et al: Thymidylate synthase gene polymorphism determines response and toxicity of 5-FU chemotherapy. Pharmacogenomics J 1:65-70, 2001[Medline]

24. Adleff V, Hitre E, Koves I, et al: Heterozygote deficiency in thymidylate synthase enhancer region polymorphism genotype distribution in Hungarian colorectal cancer patients. Int J Cancer 108:852-856, 2004[CrossRef][Medline]

25. Lecomte T, Ferraz JM, Zinzindohoue F, et al: Thymidylate synthase gene polymorphism predicts toxicity in colorectal cancer patients receiving 5-fluorouracil-based chemotherapy. Clin Cancer Res 10:5880-5888, 2004[Abstract/Free Full Text]

26. Dotor E, Cuatrecases M, Martinez-Iniesta M, et al: Tumor thymidylate synthase 1494del6 genotype as a prognostic factor in colorectal cancer patients receiving fluorouracil-based adjuvant treatment. J Clin Oncol 24:1603-1611, 2006[Abstract/Free Full Text]

27. Jakobsen A, Nielsen JN, Gyldenkerne N, et al: Thymidylate synthase and methylenetetrahydrofolate reductase gene polymorphism in normal tissue as predictors of fluorouracil sensitivity. J Clin Oncol 23:1365-1369, 2005[Abstract/Free Full Text]

28. Thirion P, Michiels S, Pignon JP, et al: Modulation of fluorouracil by leucovorin in patients with advanced colorectal cancer: An updated meta-analysis. J Clin Oncol 22:3766-3775, 2004[Abstract/Free Full Text]

29. Zhang ZG, Rustum YM: Pharmacologic rationale for fluoropyrimidine-leucovorin combination: Biochemical mechanisms. Semin Oncol 19:46-50, 1992[Medline]

30. Cohen V, Panet-Raymond V, Sabbaghian N, et al: Methylenetetrahydrofolate reductase polymorphism in advanced colorectal cancer: A novel genomic predictor of clinical response to fluoropyrimidine-based chemotherapy. Clin Cancer Res 9:1611-1615, 2003[Abstract/Free Full Text]

31. Etienne MC, Formento JL, Chazal M, et al: Methylenetetrahydrofolate reductase gene polymorphisms and response to fluorouracil-based treatment in advanced colorectal cancer patients. Pharmacogenetics 14:785-792, 2004[CrossRef][Medline]

32. Fischer J, Schwab M, Eichelbaum M, et al: Mutational analysis of the human dihydropyrimidine dehydrogenase gene by denaturing high-performance liquid chromatography. Genet Test 7:97-105, 2003[CrossRef][Medline]

33. Harrell FE: Regression Modeling Strategies, with Applications to Linear Models, Survival Analysis and Logistic Regression. New York, NY, Springer, 2001

34. Mattison LK, Fourie J, Desmond RA, et al: Increased prevalence of dihydropyrimidine dehydrogenase deficiency in African-Americans compared with Caucasians. Clin Cancer Res 12:5491-5495, 2006[Abstract/Free Full Text]

35. Ezzeldin HH, Lee AM, Mattison LK, et al: Methylation of the DPYD promoter: An alternative mechanism for dihydropyrimidine dehydrogenase deficiency in cancer patients. Clin Cancer Res 11:8699-8705, 2005[Abstract/Free Full Text]

Submitted December 19, 2006; accepted December 6, 2007.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

Related Editorial

• Predicting Fluorouracil Toxicity: Can We Finally Do It?
  Hany H. Ezzeldin and Robert B. Diasio
  JCO 2008 26: 2080-2082 [Full Text]

Related Correspondence

• Is Fluorouracil-Induced Severe Toxicity in DPYD*2A Individuals Related to Sex or to Treatment Regimen?
  Maarten J. Deenen, Jos H. Beijnen, and Jan H.M. Schellens
  JCO 2008 26: 4997-4998 [Full Text]

This article has been cited by other articles:

				S. Afzal, S. A. Jensen, B. Vainer, U. Vogel, J. P. Matsen, J. B. Sorensen, P. K. Andersen, and H. E. Poulsen MTHFR polymorphisms and 5-FU-based adjuvant chemotherapy in colorectal cancer Ann. Onc., May 22, 2009; (2009) mdp046v1. [Abstract] [Full Text] [PDF]

				R. S. Huang and M. J. Ratain Pharmacogenetics and pharmacogenomics of anticancer agents CA Cancer J Clin, January 1, 2009; 59(1): 42 - 55. [Abstract] [Full Text] [PDF]

				M. J. Deenen, J. H. Beijnen, and J. H.M. Schellens Is Fluorouracil-Induced Severe Toxicity in DPYD*2A Individuals Related to Sex or to Treatment Regimen? J. Clin. Oncol., October 20, 2008; 26(30): 4997 - 4998. [Full Text] [PDF]

				U. M. Zanger, K. Klein, and M. Schwab In Reply J. Clin. Oncol., October 20, 2008; 26(30): 4998 - 4999. [Full Text] [PDF]

				H. H. Ezzeldin and R. B. Diasio Predicting Fluorouracil Toxicity: Can We Finally Do It? J. Clin. Oncol., May 1, 2008; 26(13): 2080 - 2082. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Publisher's Note Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schwab, M. Articles by Eichelbaum, M. Search for Related Content PubMed PubMed Citation Articles by Schwab, M. Articles by Eichelbaum, M. Related Articles Related Editorial Related Correspondence Social Bookmarking                            What's this?