Population-based differences in treatment outcome following anticancer drug therapies

Review

Population-based differences in treatment outcome following anticancer drug therapies Brigette B Y Ma, Edwin P Hui, Tony S K Mok

Population-based differences in toxicity and clinical outcome following treatment with anticancer drugs have an important effect on oncology practice and drug development. These differences arise from complex interactions between biological and environmental factors, which include genetic diversity affecting drug metabolism and the expression of drug targets, variations in tumour biology and host physiology, socioeconomic disparities, and regional preferences in treatment standards. Some well-known examples include the high prevalence of activating epidermal growth factor receptor (EGFR) mutations in pulmonary adenocarcinoma among northeast (China, Japan, Korea) and parts of southeast Asia (excluding India) non-smokers, which predict sensitivity to EGFR kinase inhibitors, and the sharp contrast between Japan and the west in the management and survival outcome of gastric cancer. This review is a critical overview of population-based differences in the four most prevalent cancers in the world: lung, breast, colorectal, and stomach cancer. Particular attention is given to the clinical relevance of such knowledge in terms of the individualisation of drug therapy and in the design of clinical trials.

It has long been recognised that different populations have different treatment outcomes following anticancer drug therapy (figure). These disparities often manifest as characteristic patterns of epidemiology, treatment outcome, drug response, and toxicities that are clustered among specific ethnic groups or geographical regions. It has long been thought that such differences exist as a result of complex interactions between biological and environmental factors, but it was not until the Human Genome Project and advances in molecular biology that some of these biological factors have gradually come to light. There is now an expanding body of work that describes differences between ethnic groups in the distribution of genetic polymorphisms that affect DNA-repair enzymes, drug-metabolising enzymes, and cellular transporters of cytotoxic chemotherapy. These genetic variants are often single nucleotide polymorphisms (SNPs) and haplotypes (combinations of SNPs that are inherited together), which can be phenotypically associated with altered enzymatic activity and pharmacokinetics of anticancer drugs. Inter-ethnic variability also extends to the expression of drug targets and receptors that are relevant to predicting the efficacy of target-based drugs. It is therefore timely to review these new findings in light of their relevance to the contemporary practice of oncology and the development of new anticancer drugs. The aim of this review is to explore the biological and environmental bases of population-based differences in tumour biology and treatment outcome following cytotoxic chemotherapy and target-based drugs. Biological factors will be discussed in the context of genomic (inherited or acquired) and molecular factors, while environmental factors will be discussed in terms of disparities in access to health-care resources, public health policy, and treatment standards. The term “population” used in this Review refers to people of different sex, ethnic origin, or those living in certain www.thelancet.com/oncology Vol 11 January 2010

geographical regions within a country or across different continents. This review is limited to the four most prevalent cancers in the world—lung, breast, colorectal, and gastric cancers1—and will focus only on sporadic cancers: therefore excluding hereditary cancer syndromes.

State Key Laboratory in Oncology in South China, Sir Y K Pao Centre for Cancer, Department of Clinical Oncology, Hong Kong Cancer Institute, and Li Ka Shing Institute of Health Sciences, Chinese University of Hong Kong (B B Y Ma FRACP, E P Hui MRCP, Prof T S K Mok FRCPC) Correspondence to: Prof Tony Mok, Department of Clinical Oncology, Ngan Shing Street, Prince of Wales Hospital, Shatin, New Territories, Hong Kong SAR, China [email protected]

Lung cancer Lung cancer accounts for over 1·3 million deaths per year worldwide, and continues to be the most prevalent and one of the most fatal malignancies.1 Epidemiological studies have projected a decline in the incidence of lung cancer in the UK and USA, while an increase is expected in China.2 Ethnic origin and sex have been assessed as possible predisposing factors for lung cancer, and it has been estimated that Asian Americans (Americans of Chinese, Korean, or Japanese background or heritage) have a four-fold higher risk of lung cancer than white

Science Photo Library

Introduction

Lancet Oncol 2010; 11: 75–84

Figure: Different populations will have different treatment outcomes following anticancer therapy

75

Review

Americans after adjusting for age and smoking habits.3 By contrast, the question of whether women are more susceptible to lung cancer than men has been a subject of debate for many years. The largest prospective cohort study published to date reported similar incidence rates of lung cancer between sexes in the USA, after adjusting for smoking exposure.4

Biological and genetic differences in drug responses and susceptibility to cancer By far the most striking difference in the distribution of genetic mutations in lung cancer is the high prevalence of somatic mutations of the epidermal growth factor receptor (EGFR) among Chinese, Korean, and Japanese female non-smokers with lung adenocarcinoma. The mutational sites are clustered between exons 18 and 21 of the EGFR gene, in the tyrosine kinase binding domain.5 Patients with lung cancers that harbour base-pair deletions at exon 19 and the L858R mutation at exon 21 respond very well to EGFR tyrosine kinase inhibitors (TKIs) such as gefitinib and erlotinib.6 In addition to being a powerful predictive biomarker of response to EGFR TKIs, activating EGFR mutations also predict better clinical outcome after gefitinib therapy. In the IPASS study, 1217 patients from China, Hong Kong, Taiwan, Japan, Thailand, Indonesia, Phillipines, Singapore, and Malaysia were randomly assigned to either gefitinib alone (n=609) or carboplatin–paclitaxel (n=608) for the first-line treatment of incurable lung adenocarcinoma. The gefitinib group had longer progression-free survival (PFS) than the carboplatin–paclitaxel group (hazard ratio [HR] 0·74, 95% CI 0·65–0·85; p<0·0001), but this benefit was more pronounced in a subgroup analysis of patients with tumours that were EGFR-mutant (EGFRmut) versus EGFR-wild-type (EGFRwt). In the EGFRmut group, patients who were randomised to gefitinib experienced a significantly higher objective response rate (71·2% vs 47·3%, p=0·0001) and PFS (HR 0·48, 95% CI 0·36–0·64; p<0·0001) than those who had carboplatin–paclitaxel. Conversely, in the EGFRwt group, PFS was better in patients who were randomly assigned to carboplatin– paclitaxel than those who received gefitinib (HR 2·85, 95% CI 2·05–3·98; p<0·0001).7 This suggests that the EGFRmut group benefits more from using gefitinib in the first-line setting, whereas the EGFRwt group benefits from receiving chemotherapy first. This difference in response is not confined to Asians, as studies in white Europeans and white American populations have also reported similar findings. However, activating EGFR mutations could be identified in around 60% of Asians who were non-smokers or light-smokers with adenocarcinoma,7 but only 13·3% amongst white patients enrolled in the Spanish Lung Cancer Group study.8 The Spanish study, which selected 2312 patients who were EGFRmut for erlotinib alone as the first or second-line treatment for lung adenocarcinoma, reported a median overall 76

survival of 24 months and an overall response rate of 73%.8 This result is impressive considering that historically, unselected patients with lung cancers have a median overall survival of just 8 months and a response rate of 20% when treated with first-line platinum-based chemotherapy.9 Given the powerful predictive and prognostic effect of EGFR mutational status, it is now necessary to include this factor when deciding on the use of EGFR TKIs in the first-line setting, and in the design of any multi-national clinical trials of lung cancer. Moreover, when analysing survival in a comparative study for lung cancer, the data should be adjusted for any imbalances that arise from the proportion of patients with EGFR mutations, and from the salvage use of EGFR TKIs following study withdrawal, as these factors will strongly affect treatment outcome.

Differences in susceptibility to toxicities of cytotoxic chemotherapy Differences exist between populations of different ethnic origins in terms of their response and incidence of toxicity to cytotoxic chemotherapy such as taxanes. Sekine and colleagues10 reported a higher incidence of neutropenia and neutropenic fever among Japanese than US patients who were treated with platinum and taxane combinations. In another study in which Asians (primarily of Chinese descent) and Australians of European ancestry were treated similarly with docetaxel (75 mg/m²) and carboplatin (area-under-the-curve [AUC] of 6), Asian patients experienced a higher incidence of myelo-toxicities and higher response rates (65% vs 31%) than Australians of European ancestry.11 Such differences in drug efficacy and toxicity might be related to variations in genes involved in the metabolism and transport of docetaxel, such as CYP3A4 and MDR1 (also known as ABCB1).12 For these reasons, some Japanese investigators favour the use of taxanes at lower doses or on a weekly schedule instead of the 3-weekly dosing regimen used in studies of North American and European patients when designing clinical trials for non-small cell lung cancer.13

Breast cancer Breast cancer is the second most common cancer worldwide, and its incidence is rising in east Asia.1 In multiethnic US populations, the rising incidence of breast cancer is mainly accounted for by increased incidence in white Americans, while the mortality rates of African Americans are 37% higher than white and Asian (Chinese, Korean, Japanese) Americans.14 This poorer outcome might be due in part to population-based disparities in tumour biology, as some ethnic groups seem to be predisposed to tumours with poor prognostic features, such as high tumour grade, HER2 (also known as ERBB2) overexpression, absent hormone-receptor expression, and the basal-like subtypes known as triple-negative (hormone-receptor negative and HER2 www.thelancet.com/oncology Vol 11 January 2010

Review

negative) breast cancers. This applies especially to Hispanic and African Americans,15,16 even when socioeconomic factors have been taken into consideration.16 For instance, the prevalence of triple-negative breast cancer was reportedly 39% in premenopausal African Americans,17 compared with 8% in the equivalent Japanese population.18

Differences in tolerability and response to hormonal therapy The CYP P450 family of enzymes, including carbonyl reductases and sulfotransferases, have crucial roles in the metabolism of anticancer drugs. Inter-ethnic variability in the activity of these enzymes has long been recognised, and has been attributed to some functional polymorphisms in patients with breast cancer. For instance, polymorphic variants of CYP2D6 and SULT1A1 (sulfotransferase 1A1) have been associated with altered pharmacokinetics and clinical efficacy of selective oestrogen-receptor modulators (SERMs) in some ethnic groups.19–22 Variants in the CYP2D6 allele (eg, CYP2D6*4, CYP2D6*5, CYP2D6*10, and CYP2D6*41) and SULT1A1*2 have been linked with the impaired formation of active metabolites of tamoxifen, and therefore contribute to the poorer survival of carriers of such variants who were treated with tamoxifen in the adjuvant21,22 and palliative settings20 compared with non-carriers. The Japanese have a higher prevalence of a SNP in the CYP2D6*10 allele, which is correlated with higher recurrence rates after treatment with adjuvant tamoxifen.19 By contrast, CYP2C19*17 variants have been associated with higher enzymatic activity and possibly better survival.22 Punglia and colleagues23 addressed the clinical significance of CYP2D6 polymorphisms by investigating the effect of genotype variants on the survival of patients treated with adjuvant tamoxifen versus aromatase inhibitors in two multicentre phase 3 trials.23 Although these phase 3 studies found that aromatase inhibitors improved disease-free survival compared with tamoxifen, Punglia23 and colleagues did not find any differences in survival among patients treated with either drug who were carriers of wild-type CYP2D6. Collectively, these studies suggest that CYP2D6 genotyping as a tool for individualising endocrine therapy in patients with breast cancer should be further prospectively evaluated.

Differences in tolerability and response to cytotoxic chemotherapy Both Chinese and African patients are known to experience a higher incidence of myelotoxicity from adjuvant anthracycline or docetaxel for breast cancer than white patients.12,24 For African Americans and African Caribbeans, this susceptibility has been attributed to the generally lower levels of pre-treatment neutrophil (and also platelet) counts compared with white patients, and even among healthy individuals.25 Chinese patients might www.thelancet.com/oncology Vol 11 January 2010

be more prone to myelotoxicity because of their lower pharmacokinetic clearance of docetaxel compared with white patients;12 a trend that has not been noted in African Americans.26 To explain these differences in pharmacokinetics, researchers have tried to correlate altered drug clearance and exposure (eg, AUC) with the presence of functional variations in genes that encode drug-metabolising enzymes and membrane transporters. To date, such phenotype–genotype associations have been reported in people of Chinese, Malaysian, and Indian ancestry treated with doxorubicin.27–29 Polymorphic variants of the PXR*1B haplotype clusters have been linked to reduced activity of the PXR (pregnane X receptor; encoded by the NR1I2 gene), and its downstream target enzymes, CYP3A4 and ABCB1 (also known as MDR1), which resulted in lower clearance of doxorubicin.27 Furthermore, people of Chinese, Malaysian, and Indian ancestry who carried variants of the multidrug-resistance gene (ABCB1) and the CBR1 D1 (carbonyl reductase 1) gene29,30 experienced an increased clearance of doxorubicin and reduced peak plasma levels of doxorubicinol (an active metabolite of doxorubicin). By contrast, SNPs in SLC22A16 (solute carrier family 22) are associated with an increased AUC of doxorubicin and doxorubicinol concentrations, and are more frequent in Chinese than in individuals.28 Compared with studies on doxorubicin, studies of the phenotype–genotype association between SNPs and the pharmacokinetics of taxanes have been inconclusive. The presence of SNPs or haplotypes of ABCB1, ABCG2, and CYP2C8 was shown not to be significantly correlated with altered clearance or exposure to paclitaxel (except with some of its metabolites).31,32 Similarly, no association was found between docetaxel clearance and variants of genes that encode PXR, hepatic nuclear factor 4α, constitutive androstane receptor, and other CYP P450 enzymes.12,33

Colorectal cancer Colorectal cancer is most prevalent in western countries, although the incidence in east Asia is approaching that in the west.1 Population-based studies from the US have revealed marked disparities in the prognosis of colorectal cancer in populations of different ethnic origin, such that the mortality rate is 45% higher in African Americans than in the white population.34 By contrast, Asian Americans are more likely to develop smaller and distally located colorectal cancers than African Americans,35 and tend to survive the disease longer than patients from other ethnic groups.36 One of the most debated factors that potentially contributes to the poorer outcome noted in African-American patients with colorectal cancer is whether there is a disparity in the receipt of oncological treatment.37 Although unequal access to the healthcare system was once thought to be a key factor contributing to poor outcome in African-American patients, a large population-based study found that even among medically insured patients who had equal opportunity to 77

Review

receive advice by medical oncologists on adjuvant chemotherapy, older African Americans were less likely to receive chemotherapy than white Americans, suggesting that other factors such as disease stage and social support might be more important.37 This section will focus on those genomic biomarkers of toxicity and response to anticancer drugs for colorectal cancer that have shown population-based differences. Although recent studies have proven the predictive value of KRAS mutation and response to cetuximab,38 and topoisomerase I expression and response to irinotecan and oxaliplatin,39 they will not be discussed here since population-based differences have not yet been reported with these biomarkers.

Variations in tolerance and response to irinotecan Irinotecan is commonly used for the treatment of colorectal cancer. In colorectal cancer, variability in tolerance to irinotecan has been well described between populations, and has a pharmacogenomic basis. Uridine diphosphate glucuronosyltransferase (UGT1A1) is responsible for the metabolism of SN38, the active metabolite of irinotecan, and functional variants of the UGT1A1 gene have been associated with reduced activity of this enzyme and altered pharmacokinetics of irinotecan.40 These gene variants arise from insertions of varying numbers of TA repeats at the TATA box sequence of the UGT1A1 promoter, resulting in reduced gene transcription. The UGT1A1*28 polymorphism represents the presence of a (TA)7TAA sequence in the promoter region, instead of the wild-type (TA)6TAA. Carriers can be heterozygote TA(6/7) or homozygote TA(7/7). The UGT1A1*28 allele variants have been the most studied in colorectal cancer, and can be found in 20–50% of white40,41 and Japanese individuals,42 respectively, but are less frequent in Chinese (<20%) individuals.43 Retrospective and prospective studies in white and Japanese patients have shown that homozygotes of UGT1A1 TA(7/7) genotype are seven to nine times more likely to experience severe diarrhoea44 and neutropenia42 than heterozygous TA(6/7) or wild-type TA(6/6) genotypes after treatment with irinotecan.40,41 This might be explained by the observation that patients who are homozygous for UGT1A1 TA(7/7) have a higher AUC of SN3841 and lower SN38 glucuronidation rate.45 Based on these findings, the US Food and Drug Administration (FDA) updated the drug label of irinotecan by including pharmacogenetic information in 2005. By contrast, the association between UGT1A1 polymorphisms and response to irinotecan is less clear. In a study of 250 white patients with metastatic colorectal cancer, homozygotes of the UGT1A1 TA(7/7) had better response rates to irinotecan-containing regimens than those with wild-type genotypes.41 However, this difference was not detected in a study of Chinese patients, in which most patients (79·7%, 102 out of 128 patients) had the wild-type genotype.43 78

In addition to UGT1A1 polymorphisms, SNPs or haplotypes of other alleles or genes have shown differences in distribution between ethnic groups. SNPs of the UGT1A1*6 allele have been linked to altered SN38 metabolism and risk of neutropenia in Chinese, Malays, and Indians.46 The haplotype structure of promoter variants at UGT1A1 and its transcriptional regulator PBREM (phenobarbital-responsive enhancer module gene) is different between white and African-American individuals.47 The functional significance of these variants remains to be determined.

Variations in tolerance and response to fluoropyrimidines In a recent pooled analysis of three multinational trials in patients with colorectal cancer, patients from outside the USA (particularly east Asian) reported higher incidence rates of severe toxicity and drug discontinuation than patients from sites in the USA treated with fluorouracil or capecitabine.48 In an attempt to explain this disparity, a Japanese study compared the pharmacokinetics of capecitabine and its metabolites between 20 Japanese and 24 white recipients, but failed to find any differences.49 Other researchers have focused their attention on the inter-ethnic variations in fluorouracil-metabolising genes (eg, dihydropyrimidine dehydrogenase [DPD], encoded by DPYD) and fluorouracil targets (eg, thymidylate synthase). It has been estimated that around 3% of the general global population have partial DPD enzyme deficiency, and might be at risk of developing severe fluorouracil-induced toxicity.50 Some of these deficiencies have been attributed to the presence of DPYD point mutations, especially at the DYPD*2 allele,50 which can be found in 1% of the general population. However, white patients are more likely to be homozygous (0·9%) carriers50 than Japanese or Chinese (<0·2%) patients.51,52 Significant inter-ethnic heterogeneity is also seen in the distribution of other DPYD SNPs and haplotypes;53 however, their clinical significance is yet to be determined. Thymidylate synthase is the inhibitory target of fluorouracil, and a meta-analysis showed that increased thymidylate synthase expression is a negative prognostic factor in patients with advanced colorectal cancer.54 Thymidylate synthase polymorphisms arising from allelic repeats at the promoter enhancer region (TSER) and the 3´-untranslated region (3´-UTR) have been associated with increased toxicity to fluorouracil and capecitabine in a European55 and a Korean populations.56 The TSER contains a polymorphic tandem repeat sequence (2R or 3R) and a SNP (G>C) within the second repeat of the 3R alleles. The low enzyme expression-related genotypes (2R/2R, 2R/3RC, and 3RC/RC) are found in 5–32% of Europeans, while the high enzyme expression-related genotypes (2R/3RG, 3RG/3RC, and 3RG/3RG) are found in around 15% of Europeans.55 A small Spanish study reported that 49 patients with the low enzyme-expressionrelated genotypes had a better clinical response to fluorouracil than 40 patients with high enzymewww.thelancet.com/oncology Vol 11 January 2010

Review

expression-related genotypes.57 Unlike European populations, Chinese and Japanese populations have a much higher prevalence of the triple repeat allele (TSER 3R/3R; 63–67%),58,59 but no differences in clinical outcome between genotypes are reported in Japanese patients.59 In addition to genomic factors, the question of whether women with Dukes’ stage C colorectal cancer are more likely to benefit from adjuvant fluorouracil than men remains controversial. A subset analysis of a phase 3 study60 and a retrospective study61 suggested that women have better clinical outcome following adjuvant fluorouracil-based chemotherapy than men. By contrast, a subset analysis of another phase 3 study found that older men were more likely to benefit from fluorouracil-based chemotherapy,62 while a large prospective analysis of pooled data from 85 934 patients found no difference in benefit from adjuvant chemotherapy between the different sexes.63

Gastric cancer There are several marked differences in the epidemiology and prognosis of gastric cancer between east Asian (especially Japan) and western populations. The incidence rates of adenocarcinoma of the oesophagus and gastrooesophageal junction are increasing at a dramatic rate in the west, whereas cancers of the middle distal third of the stomach are more prevalent in Japan. Although the age-standardised incidence rate and crude mortality rates of gastric cancer for all genders in Japan and some western countries have decreased over the past two decades,64 the reported treatment outcome in Japan is still better than in the west. The 5-year overall survival of resectable gastric cancer is around 70% in specialist Japanese centres,65 compared with only around 35% in specialist centres in Europe and the UK.66,67 The relative contribution of biological versus environmental factors in explaining this disparity in outcome has been a subject of intense debate. Proponents of the theory that tumours in east Asians are less aggressive proposed that this difference had a genetic basis; however, studies published to date have failed to find a genetic difference between clinically comparable gastric cancers in Japan and the west.68 Population-based studies in North America have had mixed results in terms of ascertaining Asian (Chinese, Japanese, Vietnamese, Fillipino, Korean) ethnic origin as an independent predictor of survival in gastric cancer. A study of over 2000 Asian migrants in Canada found that although Asians were more likely to have proximal gastric cancers, Asian ethnic origin did not predict survival.69 This contrasts with a study based on the US National Cancer Data Base, which found that Asian ethnic origin, female sex, and distal tumours are among the predictors of better survival in gastric cancer.70 Female sex has also been reported as a good prognostic factor in European studies,71,72 although sex has not been shown to affect the efficacy of adjuvant chemotherapy in a subset analysis of a phase 3 Japanese study.73 www.thelancet.com/oncology Vol 11 January 2010

The other more commonly held opinion—especially among Japanese researchers73,74—is that disparity in gastric cancer treatment outcomes between east-Asian and western countries are related to the use of population screening in Japan, differences in the approach to surgery and adjuvant therapy, and differences in efficacy and tolerance to anticancer drugs against gastric cancer. Since the introduction of a nationwide screening programme in 1983, a progressive stage migration of gastric cancer towards an early stage at diagnosis has been noted in Japan, where over 40% of cases are localised at diagnosis compared with around 25% in the USA.74,75

Differences in surgical practice between Japan and western countries Endoscopic mucosal resection, a minimally invasive technique pioneered by Japanese surgeons, has been widely used in Japan for treating up to 50% of stage IA gastric cancers detected at screening. Such an approach has limited applicability in the west because most patients are diagnosed when their disease is at an advanced stage.74 Gastrectomy with the systematic dissection of lymph nodes, known as D2 dissection, is the standard surgical technique used in Japan, whereas D1 dissection is the standard in most western countries. Randomised and non-randomised studies in the west have been unable to reproduce76 the excellent 5-year survival of up to 69%65 reported in Japan with D2 dissection. Differences in surgical technique and outcomes can affect how adjuvant trials are done and interpreted. Randomised trials from western regions have shown a significant survival benefit of adjuvant chemoradiation or intensive peri-operative combination chemotherapy with the epirubicin, cisplatin, and fluorouracil (ECF) regimen. However, baseline surgical quality and outcomes were quite different from those in Japan, and Japanese oncologists have not accepted the western results.74 This is because in Japan, most patients present at an early stage that is amendable to primary surgery without the need for downstaging with neoadjuvant chemotherapy,74 therefore primary surgical resection and post-operative chemotherapy alone is considered the standard. This is in sharp contrast with practice in the UK and Europe, where neoadjuvant strategies are more popular,77 while in the US, D1 dissection followed by post-operative chemoradiation is favoured.78

Differences in tolerance to chemotherapy for gastric cancer Differences in tolerability to postoperative chemotherapy exist between Japanese and non-Japanese populations. Physiological differences and the lower incidence of postoperative complications might explain why Japanese patients are known to tolerate adjuvant chemotherapy much earlier after surgery than non-Asian patients. Another possible reason is the preference for less toxic adjuvant chemotherapy regimens in Japan, such as the use 79

Review

of single agent intravenous or oral fluorouracil,73 by contrast with the use of multi-agent chemotherapy in the west. This difference in preference for certain chemotherapeutic agents has also led to disparities in the results of randomised trials of new drugs for advanced gastric cancer. The control group in most non-Japanese studies often consists of cisplatin-based doublets or triplets, whereas single-agent fluorouracil or S-1 alone is used in Japanese trials. The platinum-based triplets DCF (docetaxel, cisplatin, and fluorouracil)79 and ECF77 are now the standard of care in the USA and UK, respectively, based on their survival benefit in randomised studies. However, such regimens are often considered by Japanese oncologists to be too toxic and to confer only modest benefits. S-1 is an oral fluoropyrimidine composed of tegafur, 5-chloro-2,4-dihydroxypyridine, and potassium oxonate at a molar ratio of 1:0·4:1. This drug was developed and is widely used in Japan, but is not yet approved in the west. Interestingly, phase 1 studies of S-1 in Japan and the USA80–82 have reported regional differences in its dose-limiting toxicity, where myelosuppression was more Variant

Prevalence of allelic types

common among Japanese patients and diarrhoea in western patients, with the overall risk of toxicity higher in Americans and Europeans. This disparity might be due to a difference in pharmacokinetics, as Americans and Europeans have a higher AUC of fluorouracil (the active metabolite of S-1) following treatment with much lower doses of S-1 than Japanese patients. In turn, this might be due to the fact that the conversion rate of tegafur to fluorouracil differs in Asian (Japanese) and white (Dutch) patients because of polymorphic differences in the CYP2A6 gene.83 Another possibility is that American patients tend to have a higher body surface area than Japanese patients.84 Interestingly, the pharmacokinetic data in a US phase 1 study found that potassium oxonate, the agent in S-1 which supposedly reduces diarrhoea, can be metabolised to a potentially toxic metabolite.80

Implications for clinical practice Pharmacogenomic studies have shown that there is a high degree of genetic diversity in drug-metabolising genes between different ethnic groups (table 1), thus raising the Phenotype association

Drug target expression EGFR (lung cancer)

Activating mutations of tyrosine-kinase-binding domain (exon 18–21)

East Asians 27–60%;5,7 Euoropeans 8–13·3%;5,8 females 42%; Predicts better response to EGFR tyrosine kinase inhibitors in lung cancer males 14%;5 never smokers 50%; smokers 10%5

Gene encoding drug-metabolising enzymes CYP2D6 (breast cancer)

CYP2D6*4; CYP2D6*5; CYP2D6*10; CYP2D6*41

Europeans have four allelic types:22 no null alleles (*4/*5: 60%); one only (*4/*5: 25%); two (*4/*5; 7%); one (*4/*5) and one (*10/*41; 8%); or two (*10/*41: 7%) East Asians: *10/*10 ( 24%)19,20

Absent null alleles associated with intact formation of active tamoxifen metabolites; harbouring one or more null alleles *4/*5 linked to impaired formation of active tamoxifen metabolites and poor survival after adjuvant tamoxifen;22 *10/*10 allele confers increased risk of disease recurrence after adjuvant tamoxifen19 and impaired formation of active tamoxifen metabolites20

CYP2C19 (breast cancer)

CYP2C19 *1/*2/*3 (wt); CYP2C19*17

Europeans have three allelic types:22 wt/wt (55%); wt/*17 (36%); and *17/*17 (9%)

Carriers of *17/*17 have increased enzyme activity and better survival after adjuvant tamoxifen compared with carriers of other alleles22

SULT 1A1 (breast cancer)

SULT1A1*2

Europeans have three allelic types: *1/*1 (43%); *1/*2 (43%); *2/*2 (14%)21 African Americans have three allelic types: *1/*1 (43%); *1/*2 (47%); *2/*2 (10%)

*2/*2 triples the risk of death with adjuvant tamoxifen compared with other genotypes21

PXR*1B (breast cancer)

PXR*1B haplotype clusters

Chinese (29%); Malay (35%); Indian (19%)27

Reduced activity of PXR and its downstream targets CYP3A4 and ABCB1; lower clearance of doxorubicin27

SLC22A16 (breast cancer)

c.146GG; c.1226T>C

Chinese and Malay28 c.146GG (13–18%); Europeans28 c.146GG (9%)

c.146GG variant: associated with increased AUC of doxorubicin and doxorubicinol concentrations28

UGT1A1*28 UGT1A1 (colorectal cancer)

Carriers of TA(7/7) are 7–9 times more likely to experience severe diarrhoea44 Three allelic types: wt TA(6/6); heterozygote TA(6/7); homozygote TA(7/7) and neutropenia42 after irinotecan;40,41 they also have a higher AUC of SN38,41 Prevalence of UGT1A1*28 polymorphisms: North American and lower SN38 glucuronidation rates than non-carriers45 and Europeans 40–50%;40,41 Japanese42 and Chinese43 <20%

DPYD DYPD*2 point mutation (colorectal cancer)

Prevalence of homozygotes: Europeans 0·9%;50 Japanese/ Chinese <0·2%51,52

TYMS Double repeat TSER*2, 2R/2R; Europeans: low TS expression genotypes (2R/2R, 2R/3RC, 3RC/RC), 5–32%; high TS expression genotypes (2R/3RG, (colorectal cancer) triple repeat TSER*3C, *3G 3RG/3RC, 3RG/3RG), 15%;55 Chinese and Japanese: 3R/3R, 63–67%58,59

Unclear relationship between DYPD*2 point mutation and toxicity to fluoropyrimidines51,52 2R/2R, 2R/3RC and 3RC/RC are associated with reduced enzyme expression and increased response to fluorouracil57

Drug transporters MDR1/ABCB1 (breast cancer)

c.1236T; c.2677T; c.2677A; c.3435T

Chinese:29 c.1236T (60%); c.2677T (38%); c.2677A (7%); c.3435T (22%)

Patients with CC-GG-CC genotypes have lower doxorubicin exposure levels compared with patients with CT-GT-CT and TT-TT-TT genotypes;29 c.3435T is associated with reduced p-glycoprotein activity29

EGFR=epidermal growth factor receptor. SULT 1A1=sulfotransferase 1A1. SLC22A16=solute carrier family 22. UGT1A1=uridine diphosphate glucuronosyltransferase. wt=wild-type. PXR=pregnane X receptor. DPYD=dihydropyrimidine dehydrogenase; MDR1 (ABCB1)=multidrug-resistance gene. TYMS=thymidylate synthase gene. CBR1=carbonyl reductase 1 gene. AUC=area-under-curve concentration.

Table 1: Inter-ethnic differences in the prevalence of selected genetic variants that affect response and/or toxicity to anticancer drugs

80

www.thelancet.com/oncology Vol 11 January 2010

Review

Biological

Environmental/epidemiological

Lung cancer

High prevalence of EGFR tyrosine kinase mutations in Asian female non-smokers High incidence of taxane-related neutropenia in Asians

Higher prevalence of pulmonary adenocarcinoma in Asian women Declining incidence in the west, rising incidence in China

Breast cancer

Predisposition to cancers with poor prognostic features in Hispanic and African Americans Pharmacogenomic variations affecting clinical outcome with hormonal therapy, and toxicity from chemotherapy Higher incidence of doxorubicin and docetaxel-related myelotoxicity in Asians

Rising incidence in white women in the USA Higher mortality rates in African Americans than in white individuals, due in part to disparities in access to medical care and socioeconomic status.

Colorectal cancer

Higher incidence of severe toxicities from fluoropyrimidines in trial centres outside the USA (eg, east Asia); pharmacogenomic variations affecting toxicity from irinotecan and fluoropyrimidines

Rising incidence in economically developed parts of east Asia

Gastric cancer

Tolerance of post-operative chemotherapy better in Japanese studies than in western studies, possibly due to differences in body-surface area Higher incidence of S-1 neutropenia in patients from USA and Europe than those from Japan

Rising incidence of adenocarcinoma of the oesophagus and gastrooesophageal junction in the west Differences in surgical standards (extended vs limited lymphadenectomy), drug preferences (eg, S-1, ECF, DCF), and treatment outcome between Japanese and western centres Nationwide screening programme in Japan since 1983

EGFR=epidermal growth factor receptor; ECF=epirubicin–cisplatin–fluorouracil; DCF=docetaxel-cisplatin-fluorouracil.

Table 2: Population-based differences in drug tolerance and treatment outcome following anticancer drug therapies, and the associated causes of such differences

prospect of individualising drug therapy based on pharmacogenomic profiling of cancer patients. However, opinions are divided as to the clinical value of such an approach. Diversity begets complexity, and studies have shown that variations of multiple genes or haplotypes (rather than single genes) are often involved in influencing the expression of a clinically relevant phenotype. Furthermore, only some variants are functionally associated with altered enzyme activity, but even then they do not consistently predict clinically significant events. For example, DYPD polymorphisms account for only a few of the reported cases of severe fluorouracil toxicities in white patients,50 while most cases of severe irinotecan-related toxicity do not have a molecular basis. This could be because of the relatively low incidence of functionally significant gene polymorphisms in some ethnic groups (eg, UGT1A1*28 allelic variants in Chinese and Japanese individuals), and interactions with dominant non-genetic factors. For example, severe irinotecan toxicities are affected by liver function (eg, bilirubin level),85 dosage and schedule of irinotecan,86 and the type of drug used in combination with irinotecan. To illustrate these points, the pharmacogenomic data on the UGT1A1*28 polymorphism which led to the change of drug label of irinotecan were based mainly on data derived from studies using the 3-weekly schedule of irinotecan (doses 300–350 mg/m²), but not with weekly or 2-weekly dosing regimens (doses 180 mg/m²).87 Even patients with colorectal cancer who were TA(7) homozygotes were able to tolerate doses and cycles of irinotecan similar to other patients, and experienced a similar frequency of dose interruptions to other genotypes in one study.88 Finally, there are no clinical guidelines concerning how dose adjustments should be made with dosing regimens of irinotecan. In a recent consensus statement published by the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group, it is considered that “the evidence is currently insufficient to recommend for or against the routine use of UGT1A1 genotyping in www.thelancet.com/oncology Vol 11 January 2010

Possible implications Lung cancer

Clinical trial design and data interpretation should include patient’s EGFR mutation status, given its effect on treatment outcome Asians might require lower dosing regimens

Breast cancer

Improve access to medical care for certain ethnic groups Clinical trial data interpretation should include patient’s ethnic status, given its effect on tumour biology and treatment outcome Further prospective studies on the pharmacogenomic profiling for individualising hormonal therapy and chemotherapy

Colorectal Clinical trial data interpretation should include patient’s ethnic origin, given its effect on toxicity cancer Further prospective studies on the pharmacogenomic profiling for individualising treatment with irinotecan and fluoropyrimidine Gastric cancer

Clinical trial design: flexibility in study treatment groups, separate registrational studies in different regions; dose modification of S-1 might be necessary in European patients

Table 3: Practical implications of population-based differences in anticancer drug tolerance and treatment outcome

patients with metastatic colorectal cancer who are to be treated with irinotecan, with the intent of modifying the dose as a way to avoid adverse drug reactions”.89 It should also be emphasised that while the main distinction between populations lies in the prevalence of these predictive genotypes, this observed relationship between specific genotypes and response to treatment is expected to be similar across populations. Most of the pharmacogenomic studies described in this review are retrospective, case–control studies, the sample sizes of which are often inadequately powered for allowing multiple comparisons across many different SNPs and haplotypes. This results in inconclusive and sometimes contradictory results. The Gastrointestinal Scientific Leadership Council of the Coalition of Cancer Cooperative Groups has highlighted the need for more prospective pharmacogenomic studies in colorectal cancer, developing more advanced bioinformatics tools for evaluating genotype–phenotype relationships, and establishing the population-based tissue and blood archives and clinical databases that are necessary for adequately-powered studies on inter-ethnic variations in pharmacogenomic biomarkers.90

For more on the EGAPP Working Group see http:// www.egappreviews.org/

81

Review

Search strategy and selection criteria Data for this Review were identified by searches of PubMed, and references from relevant articles using the search terms “ethnic differences/disparities”, “population”, “racial differences/disparities”, “pharmacogenomics”, “biology”, “treatment outcome”, and “lung/breast/gastric/colorectal cancer”. Abstracts and reports from meetings were included only if they were recently reported, phase 3 trials or large cohort studies with potentially important, innovative, or previously unpublished results, and only if they were essential for illustrating specific points. Papers published in English between January, 1980, and March, 2009, were included.

Implications in drug development For more on the International Conference on Harmonisation see http://www.ich.org/cache/ compo/276-254-1.html

82

As documented in the International Conference on Harmonisation—Ethnic Factors in the Acceptability of Foreign Clinical Data, there is a practical need to acknowledge differences between the uses of medicines in different regions of the world. The approval of new oncology drugs depends on the attainment of key efficacy and safety endpoints such as survival, response rate, and toxicity. Therefore, judicious selection of study patients and study treatment (in terms of dosage, schedule, choice of comparator groups) is crucial for the successful translation of new drugs. Population-based differences in biological factors such as target expression can have a significant effect on study outcome and data interpretation. The clinical development of gefitinib in lung cancer is a good example of the importance of how careful patient selection based on their demographic (eg, Asian non-smokers), pathological (eg, adenocarcinoma) and molecular profiles (eg, activating EGFR tyrosine kinase mutations), can make the difference between meeting7 or not meeting study endpoints.91 Differences in the incidence of drug toxicity and altered pharmacokinetics between groups of different ethnic origin point to the increasing need for the systematic prospective collection of blood samples for the assessment of pharmacogenomics, and also points to the need for clinical trials from the phase 1–3 level outside western countries. Molecular epidemiological studies using clinical data and creating large tumour-sample repositories from specific populations are also required. These activities rely on support from multicentre cooperative groups, academic bodies, the government, and the pharmaceutical industry. However, multinational cooperative studies can be complicated by regional differences in tumour biology, standards of care, and drug-approval requirements. These complications might be resolved by introducing more flexibility in trial design (such as the use of “dealer’s choice” of control group), or by doing parallel registrational studies in western and Asian populations.

Conclusion This review has highlighted some key population-based differences in the tolerance and clinical outcome to anticancer drugs (table 2). The many factors that contribute to these differences can be regarded as either biological (eg, variations in physiology, pharmacogenomics, and tumour biology) or environmental (eg, socioeconomic factors, regional differences in treatment standards). These differences can have important implications for clinical practice and the development of new drugs in oncology (table 3). In summary, population-based differences have become an essential consideration in the development of modern anticancer drugs, and are an important factor in clinical decision-making with regards to implementing new therapeutic recommendations. These differences reflect the diversity and complex interplay of biological and socio-cultural characteristics of the human race and human diseases. Contributors TTSM conceived the manuscript and wrote the section on lung cancer. BBYM wrote the sections on breast and colorectal cancer, and was responsible for the overall organisation and revision of the manuscript. EPH wrote the section on gastric cancer. All authors participated equally to the writing of the manuscript, the preparation of tables, and editing. Conflicts of interest TTSM is involved in consultancy (with honorarium) from Astra Zeneca, Hoffman-La Roche, Eli Lilly, and Pfizer. All other authors declared no conflicts of interest. References 1 Boyle P, Levin B. World Cancer Report 2008. Geneva: World Health Organization, 2008. 2 Alberg AJ, Brock MV, Samet JM. Epidemiology of lung cancer: looking to the future. J Clin Oncol 2005; 23: 3175–85. 3 Epplein M, Schwartz SM, Potter JD, Weiss NS. Smoking-adjusted lung cancer incidence among Asian-Americans (United States). Cancer Causes Control 2005; 16: 1085–90. 4 Bain C, Feskanich D, Speizer FE, et al. Lung cancer rates in men and women with comparable histories of smoking. J Natl Cancer Inst 2004; 96: 826–34. 5 Shigematsu H, Lin L, Takahashi T, et al. Clinical and biological features associated with epidermal growth factor receptor gene mutations in lung cancers. J Natl Cancer Inst 2005; 97: 339–46. 6 Paez JG, Janne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004; 304: 1497–500. 7 Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatinpaclitaxel in pulmonary adenocarcinoma. N Engl J Med 2009; 361: 947–57. 8 Porta R, Queralt C, Cardenal F, et al. Erlotinib customization based on epidermal growth factor receptor mutations in stage IV non-small-cell lung cancer patients. Proc Am Soc Clin Oncol 2008; 26 (suppl): Abstr 8038. 9 Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 2002; 346: 92–98. 10 Sekine I, Yamamoto N, Nishio K, Saijo N. Emerging ethnic differences in lung cancer therapy. Br J Cancer 2008; 99: 1757–62. 11 Millward MJ, Boyer MJ, Lehnert M, et al. Docetaxel and carboplatin is an active regimen in advanced non-small-cell lung cancer: a phase II study in Caucasian and Asian patients. Ann Oncol 2003; 14: 449–54. 12 Goh BC, Lee SC, Wang LZ, et al. Explaining interindividual variability of docetaxel pharmacokinetics and pharmacodynamics in Asians through phenotyping and genotyping strategies. J Clin Oncol 2002; 20: 3683–90

www.thelancet.com/oncology Vol 11 January 2010

Review

13

14 15

16

17

18

19

20

21

22

23

24

25 26

27

28

29

30

31

32

33

34

Mukohara T, Takeda K, Miyazaki M, Takifuji N, Terakawa K, Negoro S. Japanese experience with second-line chemotherapy with low-dose (60 mg/M2) docetaxel in patients with advanced non-small-cell lung cancer. Cancer Chemother Pharmacol 2001; 48: 356–60. Smigal C, Jemal A, Ward E, et al. Trends in breast cancer by race and ethnicity: update 2006. CA Cancer J Clin 2006; 56: 168–83. Chlebowski RT, Chen Z, Anderson GL, et al. Ethnicity and breast cancer: factors influencing differences in incidence and outcome. J Natl Cancer Inst 2005; 97: 439–48. Watlington AT, Byers T, Mouchawar J, Sauaia A, Ellis J. Does having insurance affect differences in clinical presentation between Hispanic and non-Hispanic white women with breast cancer? Cancer 2007; 109: 2093–99. Carey LA, Perou CM, Livasy CA, et al. Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 2006; 295: 2492–502. Kurebayashi J, Moriya T, Ishida T, et al. The prevalence of intrinsic subtypes and prognosis in breast cancer patients of different races. Breast 2007; 16 (suppl 2): S72–77. Kiyotani K, Sasa M, Bando Y, et al. Impact of CYP2D6*10 on recurrence-free survival in breast cancer patients receiving adjuvant tamoxifen therapy. Cancer Sci 2008; 99: 995–99. Lim HS, Ju Lee H, Seok Lee K, Sook Lee E, Jang IJ, Ro J. Clinical implications of CYP2D6 genotypes predictive of tamoxifen pharmacokinetics in metastatic breast cancer. J Clin Oncol 2007; 25: 3837–45. Nowell S, Sweeney C, Winters M, et al. Association between sulfotransferase 1A1 genotype and survival of breast cancer patients receiving tamoxifen therapy. J Natl Cancer Inst 2002; 94: 1635–40. Schroth W, Antoniadou L, Fritz P, et al. Breast cancer treatment outcome with adjuvant tamoxifen relative to patient CYP2D6 and CYP2C19 genotypes. J Clin Oncol 2007; 25: 5187–93. Punglia RS, Burstein H, Winer EP, Weeks JC. Pharmacogenomic variation of CYP2D6 and the choice of optimal adjuvant endocrine therapy for postmenopausal breast cancer: a modeling analysis. J Natl Cancer Inst 2008; 100: 642–48. Hershman D, Weinberg M, Rosner Z, et al. Ethnic neutropenia and treatment delay in African American women undergoing chemotherapy for early-stage breast cancer. J Natl Cancer Inst 2003; 95: 1545–48. Grann VR, Bowman N, Joseph C, et al. Neutropenia in 6 ethnic groups from the Caribbean and the US. Cancer 2008; 113: 854–60. Lewis LD, Miller AA, Rosner GL, et al. A comparison of the pharmacokinetics and pharmacodynamics of docetaxel between African-American and Caucasian cancer patients: CALGB 9871. Clin Cancer Res 2007; 13: 3302–11. Sandanaraj E, Lal S, Selvarajan V, et al. PXR Pharmacogenetics: Association of haplotypes with hepatic CYP3A4 and ABCB1 messenger RNA expression and doxorubicin clearance in Asian breast cancer patients. Clin Cancer Res 2008; 14: 7116–26. Lal S, Wong ZW, Jada SR, et al. Novel SLC22A16 polymorphisms and influence on doxorubicin pharmacokinetics in Asian breast cancer patients. Pharmacogenomics 2007; 8: 567–75. Lal S, Wong ZW, Sandanaraj E, et al. Influence of ABCB1 and ABCG2 polymorphisms on doxorubicin disposition in Asian breast cancer patients. Cancer Sci 2008; 99: 816–23. Lal S, Sandanaraj E, Wong ZW, et al. CBR1 and CBR3 pharmacogenetics and their influence on doxorubicin disposition in Asian breast cancer patients. Cancer Sci 2008; 99: 2045–54. Marsh S, Somlo G, Li X, et al. Pharmacogenetic analysis of paclitaxel transport and metabolism genes in breast cancer. Pharmacogenomics J 2007; 7: 362–65. Saito Y, Katori N, Soyama A, et al. CYP2C8 haplotype structures and their influence on pharmacokinetics of paclitaxel in a Japanese population. Pharmacogenet Genomics 2007; 17: 461–71. Tham LS, Holford NH, Hor SY, et al. Lack of association of single-nucleotide polymorphisms in pregnane X receptor, hepatic nuclear factor 4α, and constitutive androstane receptor with docetaxel pharmacokinetics. Clin Cancer Res 2007; 13: 7126–32. American Cancer Society. Colorectal Cancer Facts and Figures 2008–2010. http://www.cancer.org/downloads/STT/F861708_ finalforweb.pdf (accessed Nov 2, 2009).

www.thelancet.com/oncology Vol 11 January 2010

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50 51

52

53

54

55

56

Shavers VL. Racial/ethnic variation in the anatomic subsite location of in situ and invasive cancers of the colon. J Natl Med Assoc 2007; 99: 733–48. Artinyan A, Sun C, Bhatia S, Pigazzi A, Garcia-Aguilar J, Kim J. Stark racial disparities in outcomes for colorectal cancer in Los Angeles County. Proc Am Soc Clin Oncol Gastrointest Cancers Symp 2009; Abstr 324. Baldwin LM, Dobie SA, Billingsley K, et al. Explaining black-white differences in receipt of recommended colon cancer treatment. J Natl Cancer Inst 2005; 97: 1211–20. Linardou H, Dahabreh IJ, Kanaloupiti D, et al. Assessment of somatic k-RAS mutations as a mechanism associated with resistance to EGFR-targeted agents: a systematic review and meta-analysis of studies in advanced non-small-cell lung cancer and metastatic colorectal cancer. Lancet Oncol 2008; 9: 962–72 Braun MS, Richman SD, Quirke P, et al. Predictive biomarkers of chemotherapy efficacy in colorectal cancer: results from the UK MRC FOCUS trial. J Clin Oncol 2008; 26: 2690–98. Innocenti F, Undevia SD, Iyer L, et al. Genetic variants in the UDP-glucuronosyltransferase 1A1 gene predict the risk of severe neutropenia of irinotecan. J Clin Oncol 2004; 22: 1382–88. Toffoli G, Cecchin E, Corona G, et al. The role of UGT1A1*28 polymorphism in the pharmacodynamics and pharmacokinetics of irinotecan in patients with metastatic colorectal cancer. J Clin Oncol 2006; 24: 3061–68. Ando Y, Saka H, Ando M, et al. Polymorphisms of UDP-glucuronosyltransferase gene and irinotecan toxicity: a pharmacogenetic analysis. Cancer Res 2000; 60: 6921–26. Liu CY, Chen PM, Chiou TJ, et al. UGT1A1*28 polymorphism predicts irinotecan-induced severe toxicities without affecting treatment outcome and survival in patients with metastatic colorectal carcinoma. Cancer 2008; 112: 1932–40. Marcuello E, Altes A, Menoyo A, Del Rio E, Gomez-Pardo M, Baiget M. UGT1A1 gene variations and irinotecan treatment in patients with metastatic colorectal cancer. Br J Cancer 2004; 91: 678–82. Iyer L, Das S, Janisch L, et al. UGT1A1*28 polymorphism as a determinant of irinotecan disposition and toxicity. Pharmacogenomics J 2002; 2: 43–47. Jada SR, Lim R, Wong CI, et al. Role of UGT1A1*6, UGT1A1*28 and ABCG2 c.421C>A polymorphisms in irinotecan-induced neutropenia in Asian cancer patients. Cancer Sci 2007; 98: 1461–67. Innocenti F, Grimsley C, Das S, et al. Haplotype structure of the UDP-glucuronosyltransferase 1A1 promoter in different ethnic groups. Pharmacogenetics 2002; 12: 725–33. Haller DG, Cassidy J, Clarke SJ, et al. Potential regional differences for the tolerability profiles of fluoropyrimidines. J Clin Oncol 2008; 26: 2118–23. Reigner B, Watanabe T, Schuller J, et al. Pharmacokinetics of capecitabine (Xeloda) in Japanese and Caucasian patients with breast cancer. Cancer Chemother Pharmacol 2003; 52: 193–201. Wei X, Elizondo G, Sapone A, et al. Characterization of the human dihydropyrimidine dehydrogenase gene. Genomics 1998; 51: 391–400. Yamaguchi K, Arai Y, Kanda Y, Akagi K. Germline mutation of dihydropyrimidine dehydrogenese gene among a Japanese population in relation to toxicity to 5-fluorouracil. Jpn J Cancer Res 2001; 92: 337–42. He YF, Wei W, Zhang X, et al. Analysis of the DPYD gene implicated in 5-fluorouracil catabolism in Chinese cancer patients. J Clin Pharm Ther 2008; 33: 307–14. Maekawa K, Saeki M, Saito Y, et al. Genetic variations and haplotype structures of the DPYD gene encoding dihydropyrimidine dehydrogenase in Japanese and their ethnic differences. J Hum Genet 2007; 52: 804–19. Popat S, Matakidou A, Houlston RS. Thymidylate synthase expression and prognosis in colorectal cancer: a systematic review and meta-analysis. J Clin Oncol 2004; 22: 529–36. 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 2004; 10: 5880–88. Cho HJ, Park YS, Kang WK, Kim JW, Lee SY. Thymidylate synthase (TYMS) and dihydropyrimidine dehydrogenase (DPYD) polymorphisms in the Korean population for prediction of 5-fluorouracil-associated toxicity. Ther Drug Monit 2007; 29: 190–96.

83

Review

57

58

59

60

61

62

63

64 65

66

67

68

69

70

71

72

73

74

84

Marcuello E, Altes A, del Rio E, Cesar A, Menoyo A, Baiget M. Single nucleotide polymorphism in the 5’ tandem repeat sequences of thymidylate synthase gene predicts for response to fluorouracil-based chemotherapy in advanced colorectal cancer patients. Int J Cancer 2004; 112: 733–37. Marsh S, Collie-Duguid ES, Li T, Liu X, McLeod HL. Ethnic variation in the thymidylate synthase enhancer region polymorphism among Caucasian and Asian populations. Genomics 1999; 58: 310–12. Tsuji T, Hidaka S, Sawai T, et al. Polymorphism in the thymidylate synthase promoter enhancer region is not an efficacious marker for tumor sensitivity to 5-fluorouracil-based oral adjuvant chemotherapy in colorectal cancer. Clin Cancer Res 2003; 9: 3700–04. Moertel CG, Fleming T, Macdonald JS, et al. Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: a final report. Ann Intern Med 1995; 122: 321–26. Elsaleh H, Joseph D, Grieu F, Zeps N, Spry N, Iacopetta B. Association of tumour site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. Lancet 2000; 355: 1745–50. Laurie JA, Moertel CG, Fleming TR, et al. Surgical adjuvant therapy of large-bowel carcinoma: an evaluation of levamisole and the combination of levamisole and fluorouracil. The North Central Cancer Treatment Group and the Mayo Clinic. J Clin Oncol 1989; 7: 1447–56. Jessup JM, Stewart A, Greene FL, Minsky BD. Adjuvant chemotherapy for stage III colon cancer: implications of race/ ethnicity, age, and differentiation. JAMA 2005; 294: 2703–11. Matsuzaka M, Fukuda S, Takahashi I, et al. The decreasing burden of gastric cancer in Japan. Tohoku J Exp Med 2007; 212: 207–19. Sasako M, Sano T, Yamamoto S, et al. D2 lymphadenectomy alone or with para-aortic nodal dissection for gastric cancer. N Engl J Med 2008; 359: 453–62. Cuschieri A, Weeden S, Fielding J, et al. Patient survival after D1 and D2 resections for gastric cancer: long-term results of the MRC randomized surgical trial. Surgical Co-operative Group. Br J Cancer 1999; 79: 1522–30. Bonenkamp JJ, Songun I, Hermans J, et al. Randomised comparison of morbidity after D1 and D2 dissection for gastric cancer in 996 Dutch patients. Lancet 1995; 345: 745–48. McCulloch PG, Ochiai A, O’Dowd GM, Nash JR, Sasako M, Hirohashi S. Comparison of the molecular genetics of c-erb-B2 and p53 expression in stomach cancer in Britain and Japan. Cancer 1995; 75: 920–25. Gill S, Shah A, Le N, Cook EF, Yoshida EM. Asian ethnicity-related differences in gastric cancer presentation and outcome among patients treated at a canadian cancer center. J Clin Oncol 2003; 21: 2070–76. Al-Refaie WB, Tseng JF, Gay G, et al. The impact of ethnicity on the presentation and prognosis of patients with gastric adenocarcinoma. Results from the National Cancer Data Base. Cancer 2008; 113: 461–69. Faivre J, Forman D, Esteve J, Gatta G. Survival of patients with oesophageal and gastric cancers in Europe. EUROCARE Working Group. Eur J Cancer 1998; 34: 2167–75. Maguire A, Porta M, Sanz-Anquela JM, Ruano I, Malats N, Pinol JL. Sex as a prognostic factor in gastric cancer. Eur J Cancer 1996; 32A: 1303–09. Sakuramoto S, Sasako M, Yamaguchi T, et al. Adjuvant chemotherapy for gastric cancer with S-1, an oral fluoropyrimidine. N Engl J Med 2007; 357: 1810–20. Ohtsu A, Yoshida S, Saijo N. Disparities in gastric cancer chemotherapy between the East and West. J Clin Oncol 2006; 24: 2188–96.

75

76

77

78

79

80

81

82

83

84

85

86

87 88

89

90

91

Lambert R, Guilloux A, Oshima A, et al. Incidence and mortality from stomach cancer in Japan, Slovenia and the USA. Int J Cancer 2002; 97: 811–18. McCulloch P, Nita ME, Kazi H, Gama-Rodrigues J. Extended versus limited lymph nodes dissection technique for adenocarcinoma of the stomach. Cochrane Database Syst Rev 2004: CD001964. Cunningham D, Allum WH, Stenning SP, et al. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006; 355: 11–20. Kelsen DP. Postoperative adjuvant chemoradiation therapy for patients with resected gastric cancer: intergroup 116. J Clin Oncol 2000; 18: 32S–34S. Van Cutsem E, Moiseyenko VM, Tjulandin S, et al. Phase III study of docetaxel and cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line therapy for advanced gastric cancer: a report of the V325 Study Group. J Clin Oncol 2006; 24: 4991–97. Ajani JA, Faust J, Ikeda K, et al. Phase I pharmacokinetic study of S-1 plus cisplatin in patients with advanced gastric carcinoma. J Clin Oncol 2005; 23: 6957–65. Hirata K, Horikoshi N, Aiba K, et al. Pharmacokinetic study of S-1, a novel oral fluorouracil antitumor drug. Clin Cancer Res 1999; 5: 2000–05. van Groeningen CJ, Peters GJ, Schornagel JH, et al. Phase I clinical and pharmacokinetic study of oral S-1 in patients with advanced solid tumors. J Clin Oncol 2000; 18: 2772–79. Fujita K, Yamamoto W, Endo S, et al. CYP2A6 and the plasma level of 5-chloro-2, 4-dihydroxypyridine are determinants of the pharmacokinetic variability of tegafur and 5-fluorouracil, respectively, in Japanese patients with cancer given S-1. Cancer Sci 2008; 99: 1049–54. Comets E, Ikeda K, Hoff P, Fumoleau P, Wanders J, Tanigawara Y. Comparison of the pharmacokinetics of S-1, an oral anticancer agent, in Western and Japanese patients. J Pharmacokinet Pharmacodyn 2003; 30: 257–83. Hoskins JM, Marcuello E, Altes A, et al. Irinotecan pharmacogenetics: influence of pharmacodynamic genes. Clin Cancer Res 2008; 14: 1788–96. Hoskins JM, Goldberg R, Qu P, Ibrahim JG, McLeod HL. UGT1A1*28 genotype and irinotecan-induced neutropenia: dose matters. J Natl Cancer Inst 2007; 99: 1290–95. Hahn KK, Wolff JJ, Kolesar JM. Pharmacogenetics and irinotecan therapy. Am J Health Syst Pharm 2006; 63: 2211–17. Kweekel DM, Gelderblom H, Van der Straaten T, Antonini NF, Punt CJ, Guchelaar HJ. UGT1A1*28 genotype and irinotecan dosage in patients with metastatic colorectal cancer: a Dutch Colorectal Cancer Group study. Br J Cancer 2008; 99: 275–82. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group. Recommendations from the EGAPP Working Group: can UGT1A1 genotyping reduce morbidity and mortality in patients with metastatic colorectal cancer treated with irinotecan? Genet Med 2009; 11: 15–20. O’Dwyer P, Eckhardt SG, Haller D, et al. Priorities in colorectal cancer research: recommendations from the gastrointestinal scientific leadership council of the coalition of cancer cooperative groups. J Clin Oncol 2007; 25: 2313–21. Thatcher N, Chang A, Parikh P, et al. Gefitinib plus best supportive care in previously treated patients with refractory advanced non-small-cell lung cancer: results from a randomised, placebo-controlled, multicentre study (Iressa Survival Evaluation in Lung Cancer). Lancet 2005; 366: 1527–37.

www.thelancet.com/oncology Vol 11 January 2010