You are here

Article Text

Article menu

Download PDFPDF

Recent advances in basic science
Biomarkers of response to therapy in oesophago-gastric cancer
  1. K R Fareed1,
  2. P Kaye2,
  3. I N Soomro2,
  4. M Ilyas2,
  5. S Martin1,
  6. S L Parsons3,
  7. S Madhusudan1
  1. 1
    Laboratory of Molecular Oncology, School of Molecular Medical Sciences, Nottingham University Hospitals NHS Trust, Nottingham, UK
  2. 2
    Department of Pathology, School of Molecular Medical Sciences, Nottingham University Hospitals NHS Trust, Nottingham, UK
  3. 3
    Department of Surgery, Nottingham University Hospitals NHS Trust, Nottingham, UK
  1. Dr S Madhusudan, Laboratory of Molecular Oncology, School of Molecular Medical Sciences, Academic Unit of Oncology, Faculty of Medicine & Health Sciences, University of Nottingham, Nottingham University Hospitals NHS Trust, Nottingham NG5 1PB, UK; srinivasan.madhusudan{at}nottingham.ac.uk

Abstract

Cancer of the oesophagus, gastro-oesophageal junction (GOJ) and stomach remains a major health problem worldwide. The evidence base for the optimal management of patients with operable oesophago-gastric cancer is evolving. Accepted approaches include preoperative chemotherapy followed by surgery (oesophageal cancer), chemo-radiotherapy alone (oesophageal cancer) and perioperative chemotherapy (gastric and gastro-oesophageal adenocarcinomas). The underlying principles behind neoadjuvant therapy are to improve resectability of the tumour by tumour shrinkage/downstaging and to treat occult metastatic disease as early as possible. The response rate to cytotoxic therapy is about 40% in oesophago-gastric cancer. Available evidence suggests that a favourable histopathological response to cytotoxic therapy may be a useful positive predictive marker in oesophago-gastric cancer. However, the ability to predict tumour response in routine clinical practice is difficult and is an area of intense investigation. There is evolving evidence for the role of predictive biomarkers in cancer in general and oesophago-gastric cancer in particular. We provide an overview on the current status of radiological and biological predictive biomarkers. We have focussed on clinical translational investigations and, where appropriate, provided pre-clinical insights. Whether predictive markers will be routinely incorporated in clinical practice remains to be seen as biomarker research is expensive and the data generated from these investigations are complex. It is clear that a concerted international effort between academia and industry is critical if personalised medicine as a practical reality for our cancer patients is to be realised.

https://doi.org/10.1136/gut.2008.155861

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Western countries have seen a significant recent increase in the incidence of distal oesophageal, gastro-oesophageal junction (GOJ) and proximal gastric adenocarcinoma. The overall prognosis for gastro-oesphageal tumours remains poor.1–5 Although surgery plays a central role in the overall management of operable disease,6–9 it is clear that additional therapy is required to improve patient outcomes.10 11 The United Kingdom Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial demonstrated a 25% reduction in the risk of death and a significant improvement in 5 year survival in patients given perioperative chemotherapy compared to those treated with surgery alone.10 Surrogate markers of tumour response and downstaging, such as smaller primary tumours (3 cm vs 5 cm), early-stage tumours (T1 and T2 vs T3 and T4), and less advanced nodal status (N0 or N1 vs N2 or N3) were significantly more frequent in the perioperative chemotherapy group compared to the primary surgery group in that study.10 Similar tumour downstaging and survival benefit has also been demonstrated in the MRC OEO2 trial of surgical resection with or without preoperative chemotherapy (consisting of two cycles of cisplatin and 5-fluorouracil (5-FU) chemotherapy) in oesophageal cancer.11

    The predictive/prognostic significance of histopathological response to cytotoxic therapy was reported by Mandard et al12 in oesophageal cancer specimens (predominantly squamous cell carcinomas) following neoadjuvant chemo-radiotherapy. Tumour regression grade (TRG) analysis, as a marker of treatment response was undertaken in that study and scored from complete regression (TRG1) to absent regression (TRG5) (fig 1). Of the 93 patients in that study, 42% were TRG1–2, 20% TRG3 and 33% were TRG4–5. Tumour size, pathological lymph node status, TRG and oesophageal wall involvement correlated highly with disease-free survival. On multivariate analysis, however, only TRG (TRG1–3 vs TRG4–5) remained a significant (p<0.001) predictor of disease-free survival.12 A similar study in gastric cancer patients receiving neoadjuvant chemotherapy also demonstrated survival benefit in patients who achieved histopathological evidence of response to cytotoxic therapy.13

    Figure 1
    Figure 1 Histopathological tumour regression in response to platinum-based chemotherapy in gastric and gastro-oeasophageal tumours. Tumour regression grade (TRG) was assigned using Mandard’s criteria (see text for further details). SCC, squamous cell carcinoma.

    Current evidence suggests that preoperative treatment does not negatively impact on surgical outcomes.10 However, one of the arguments against preoperative therapy has been the risk of delaying surgery in those patients who do not respond to chemotherapy. Predictive markers of response would therefore be invaluable in individualising patient treatment as it would enable discrimination of those patients likely to respond to combination therapy from those likely to be non-responsive and may progress during therapy. The major focus of this article is to review the current status of predictive biomarkers in oesophago-gastric cancer. We initially provide a brief overview on the evolving role of functional imaging and then focus on molecular markers of tumour response.

    FUNCTIONAL IMAGING IN OESOPHAGO-GASTRIC CANCER

    Conventional contrasted computed tomography (CT) scans, endoscopy and endoscopic ultrasound do not provide functional information in tumour tissue. However, recent evidence suggests that positron emission tomography (PET) with the glucose analogue 18F-fluorodeoxyglucose (FDG-PET) may have a role in the prediction of tumour response during the early phase of chemotherapy in addition to providing prognostic information in patients14 15 The recently published Metabolic response evalUation for Individualisation of neoadjuvant Chemotherapy in oesOphageal and oesophagogastric adeNocarcinoma (MUNICON) trial prospectively evaluated the feasibility and potential effect on prognosis of administering PET response-guided chemotherapy to patients with locally advanced adenocarcinoma of the oesophagus and oesophago-gastric junction.16 All patients recruited into the trial received 2 weeks of induction platinum and fluorouracil-based chemotherapy. A decrease of 35% or more in the tumour glucose standardised uptake value (SUV) was defined as metabolic response, a surrogate marker of tumour response. The metabolic responders continued to receive neoadjuvant chemotherapy whereas the non-responders proceeded directly to surgery following 2 weeks of induction chemotherapy. Metabolic responders survived substantially longer compared to non-responders in that study. In addition, of the 50 patients who were classed as metabolic responders, 29 (58%) had major histological remissions (<10% residual tumour) but no histological response was noted in metabolic non-responders.16 This study therefore suggests that metabolic response may correlate with tumour response and may translate to improved survival in patients. The predictive role of functional PET imaging has also been demonstrated in other recently published studies in oesophago-gastric cancer.14 17 18 Ott et al19 recently updated their results in locally advanced gastric cancer and identified a further sub-group of FDG non-avid tumours that have poor prognosis similar to the metabolic non-responders.

    Key points 1

    • The incidence of proximal gastric, gastro-oesophageal junction and distal oesophageal adenocarcinoma is increasing in the UK

    • Preoperative chemotherapy (oesophageal cancer), chemo-radiotherapy alone (oesophageal cancer), perioperative chemotherapy (preoperative chemotherapy followed by surgery and postoperative chemotherapy–gastric and gastro-oesophageal adenocarcinoma) are accepted multi-modality approaches in the UK

    • Response rate to chemotherapy is about 40%, at the expense of significant toxicity

    • There are no established predictive biomarkers in oesophago-gastric cancer to individualise patient treatment

    PREDICTIVE BIOMARKERS IN OESOPHAGO-GASTRIC CANCER

    The evidence presented above clearly suggests that histopathological and radiological response to cytotoxic therapy is a marker of patient benefit and improved survival. However, it has remained a challenge for decades to identify reliable biomarkers that may be able to predict who will respond and who will progress during cytotoxic therapy. Current clinical parameters such as TNM staging, tumour location, sex and histological subtype are unable to predict responders from non-responders. An added complexity to this problem is the use of conventional protocols to treat gastro-oesophageal tumours based on their site of origin rather than taking into account their biological characteristics. This may largely account for differential response to therapy in individual tumours. For example, advances in molecular biology confirm that the cancerous phenotype is a sum total of genetic and epigenetic changes that confer selective survival advantage to clones of cells we call cancer. Similarly, the ultimate response to cytotoxic therapy is also likely to be dictated by the genetic make-up of tumours and the various cellular pathways they may initiate in response to cytotoxic therapy.

    Whilst prognostic biomarkers are measurements available at the time of diagnosis or surgery which are associated with recurrence, death or other clinical outcomes and determine how patients will fare irrespective of treatment, predictive biomarkers are measures that help determine which patients do well with particular types of treatment. An ideal predictive marker should be reliable, readily available, and detectable by reasonably acceptable laboratory techniques. The role of predictive biomarkers has been well established in solid tumours such as in breast cancer (Her-2 over-expression predicts response to herceptin and oestrogen receptor positivity predicts tamoxifen response), chronic myeloid leukaemia (breakpoint cluster region–Abelson murine leukaemia oncogene (BCR–ABL)) translocation predicts response to imatinib mesylate), lung cancer (epidermal growth factor receptor (EGFR) kinase domain mutations predict response to erlotinib or geftinib), brain tumours (EGFR mutations in the extracellular domain and response EGFR inhibitors) and colonic cancer (mutations in KRAS predict response to EGFR specific antibody therapy). Mining predictive markers in oesophago-gastric cancer at a genetic, epigenetic, transcriptional and translational level is an area of ongoing research and, as such, this will be the focus of the rest of the article. It is important, however, to remember that published studies on molecular predictors in oesophago-gastric cancer are largely retrospective, small (typically fewer than 150 patients), non-randomised and often consist of patients treated with an array of different treatment regimens broadly including fluoropyrimidines and platinum compounds. Although these studies provide valuable insights, molecular marker studies should be embedded into prospective randomised control trials to assess the true predictive value of these markers.

    The cytotoxicity of systemic chemotherapy and radiotherapy is predominantly due to their ability to induce DNA damage in cancer cells. Platinating agents (cisplatin and oxaliplatin), 5-FU, capecitabine, anthracyclins and ionising radiation are the predominant cytotoxic agents used in the treatment of oesophago-gastric cancer. The cellular responses involved in mediating the cytotoxicity of these agents are complex (fig 2). Mammalian cells have highly conserved DNA damage sensor mechanisms that result in several possible cellular processes to such potentially cytotoxic insults. These include (1) induction of apoptosis to eliminate heavily damaged cells; (2) activation of DNA damage checkpoints and modulation of cell cycle progression to allow time for repair and to prevent transmission of damaged or incompletely replicated chromosomes; (3) transcriptional response, which causes changes in the transcriptional profile that may be beneficial to the cell; (4) tolerance of damage; and (5) initiation of DNA repair and removal of damage. Collectively, these responses determine cell fate and hence response to therapy; in particular, whether to survive or whether to initiate programmed cell death.20 For example, DNA repair mechanisms that operate in cancer cells to rectify DNA-damaging lesions induced by cytotoxic agents contribute to therapeutic resistance. On the other hand, sub-optimal DNA repair in normal tissue may negatively impact on normal tissue tolerance. Genetic variations that alter protein function may include mutations and single nucleotide polymorphisms (SNPs). SNPs can alter the amino acid sequence of the encoded proteins, alter RNA splicing and gene transcription resulting in either increased or decreased expression or activity of the encoded proteins. Polymorphisms of DNA repair genes that confer sub-optimal DNA repair in tumour tissue may enhance anti-tumour activity but sub-optimal repair in normal tissue could lead to normal tissue toxicity. The evolving field of biomarker discovery has essentially focussed on analyses of critical pathways described above which dictate cellular responses. Though a detailed discussion of individual cellular pathways is beyond the scope of this article, we provide a brief summary of the potential mechanisms of action of the various cytotoxic agents and the predominant pathway involved in processing damaging lesions to facilitate later discussion of individual factors as predictive markers in oesophago-gastric cancer. Recent progress made in our understanding of molecular mechanisms has enabled identification of critical proteins that determine cytotoxicity of anti-cancer agents used in oesophago-gastric cancer. Whilst some of these factors are themselves potential drug targets for modulation of cytotoxicity, many of the proteins are emerging as potential predictive markers of disease response. The current evidence for the role of biomarkers in predicting response to therapy in oesophago-gastric cancer is evolving. We have adopted a pathway-specific approach to describe individual studies. Predictive biomarkers have been investigated at a genetic level (eg, polymorphisms), transcriptional level (mRNA expression) or at the protein level and we have taken this approach to describe individual studies.

    Figure 2
    Figure 2 Cellular responses to cytotoxic agents.

    5-FLUOROURACIL METABOLISM PATHWAY

    A summary of studies relating to the 5-FU metabolism pathway is given in table 1.

    5-Fluorouracil and related drugs

    The metabolic pathway of 5-FU is shown in fig 3. 5-FU belongs to the family of drugs known as anti-metabolites and principally acts via inhibition of DNA and RNA synthesis.21 22 Upon entering cells, 5-FU is converted to 5-fluorodeoxyuridine (5-FUDR) by the enzyme thymidine phosphorylase (TP). 5-FUDR is converted to 5-fluorodeoxyuridine monophosphate (5-FdUMP) by thymidine kinase (TK). 5-FdUMP competes with endogenous deoxyuridine monophosphate (dUMP) for binding to thymidylate synthase (TS) in a complex that is stabilised by 5,10-methylenetetrahydrofolate (CH2THF) (the enzyme methylenetetrahydrofolate reductase (MTHFR) converts 5,10-methylenetetrahydrofolate (CH2THF) to 5-methyltetrahydrofolate). Binding of 5-FdUMP to TS depletes the thymidine nucleotide pool and hence DNA synthesis. This “thymineless death” is considered to be the predominant mechanism for the cytotoxicity of 5-FU. Though the exact molecular mechanisms are unclear, recent studies suggest that depletion of dTMP leads to imbalances in the deoxynucleotide pool and accumulation of dUMP and dUTP which may be incorporated into the DNA. Base excision repair machinery may be involved in the repair of uracil misincorporation into the DNA.23 In addition, 5-FU is also converted to 5-fluorouridine monophosphate (5-FUMP), by the action of orotate phosphoribosyltransferase (OPRT). 5-FUMP is converted to 5-fluorouridine diphosphate (5-FUDP) and then to 5-fluorouridine triphosphate (5-FUTP), which inhibits RNA synthesis. Dihydropyrimidine dehydrogenase (DPD) is the initial and rate-limiting enzyme in 5-FU catabolism. DPD converts 5-FU to dihydrofluorouracil (DHFU) and thereby inactivates it. Capecitabine is also an anti-metabolite belonging to the fluoropyrimidine carbonate class. It is an orally administered precursor of 5-FU and is converted to 5-FU by carboxyesterase (CE), cytidine deaminase (CD) and TP. The daily oral administration of capecitabine mimics the continuous intravenous infusion of 5-FU.24 S-1 is an orally available 5-FU derivative that is approved for use in gastric cancer in Japan. S-1 combines three pharmacological agents: tegafur, a pro-drug of 5-FU; gimeracil, which reversibly inhibits DPD (the enzyme involved in inactivation of 5-FU); and potassium oxonate, which reversibly inhibits the enzyme responsible for phosphorylation of 5-FU (orotate phosphoribosyltransferase) thereby reducing the gastrointestinal side effects of 5-FU.25

    Figure 3
    Figure 3 The metabolic pathway of 5-fluorouracil (5-FU). A detailed description is given in the text. Predictive factors within this pathway are shown in green. CD, cytidine deaminase; CE, carboxyesterase; 5-DFUR, 5′-deoxy-5-fluorouridine; DPD, dihydropyrimidine dehydrogenase; DHFU, dihydrofluorouracil; 5-FU, 5-fluorouracil; OPRT, orotate phosphoribosyltransferase; PO, potassium oxonate; TK, thymidine kinase; TP, thymidine phosphorylase; TS, thymidylate synthase.
    View this table:
    Table 1 Studies relating to the metabolism of 5-fluorouracil

    Decreases in the activity of TP, TK and OPRT are associated with reduced metabolic activation and hence 5-FU resistance. Conversely, over-expression of enzymes involved in 5-FU metabolism (TP, TK, OPRT) may increase active metabolites of 5-FU and hence increase its cytotoxicicity. Over-expression of TS may lead to relative resistance as it may reduce 5-FdUMP binding and hence lead to 5-FU resistance. Over-expression of DPD may increase 5-FU catabolism to inactive metabolites and reduce cytotoxicity. Accordingly, several studies have investigated factors within the 5-FU pathway as potential predictive biomarkers.

    Polymorphisms

    Genetic polymorphisms that may alter the biochemical activity of TS may impact upon the therapeutic activity of 5-FU chemotherapy and hence efficacy. Advanced gastric cancer patients with TS genotypes 2R/2R, 2R/3RC or 3RC/3RC survived better compared to unfavourable genotypes (10.2 months vs 6 months) in a recently published study.26 Lu et al27 analysed the blood samples of 106 patients with advanced gastric cancer for the relationship between polymorphism in the 3′ untranslated region (3′-UTR) of the TS gene and sensitivity to 5-FU-based chemotherapy. Polymorphisms in the TS 3′-UTR were classified into three groups based on the presence or absence of a 6 bp nucleotide fragment. The response rate of patients with −6/−6 bp and −6/+6 bp was 19%, which was significantly higher than patients with +6/+6 bp, which had 0% response rate.27 In a similar study of 146 patients with oesophageal adenocarcinoma, focussing on 3′-UTR of TS, an association between 6/6 bp genotype and decreased risk of loco-regional recurrence and a higher 3 year probability of locoregional control compared to patients with other genotypes was reported.28 However, another study of TS tandem repeat polymorphism in gastric cancer patients showed no association between TS tandem repeat polymorphism and response to 5-FU-based neoadjuvant chemotherapy.29

    Messenger RNA expression

    Clinical relevance of TS mRNA expression levels in gastric cancer has been investigated in several studies. High TS mRNA expression is associated with poor response to chemotherapy.30–32 However, these results have not been reproduced by other investigators33–36 but a high TS mRNA level was associated with poor survival in two studies33 35 suggesting that TS over-expression may reflect tumour progression and hence may have prognostic implication rather than of predictive significance in gastric cancer. Another study in gastric cancer suggests that high DPD mRNA levels may predict low sensitivity to 5-FU.34

    Studies of mRNA expression levels of 5-FU metabolism genes in oesophageal cancer suggest that alterations of expression may affect treatment outcomes. In 21 patients with locally advanced oesophageal adenocarcinoma treated with neoadjuvant cisplatin and 5-FU chemotherapy, mRNA expression levels of TS in addition to TP, DPD, methylenetetrahydrofolate reductase (MTHFR), multidrug resistance-associated protein 1 (MRP1) and multidrug-resistance gene 1 (MDR1) were analysed in pre-treatment endoscopic biopsies and in tumour tissue specimens after surgical resection. Downregulation of TS and MRP1 mRNA expression levels after chemotherapy was associated with tumour response. This suggests that downregulation of chemotherapy metabolism-associated genes occurs after neoadjuvant chemotherapy and may modulate tumour response to chemotherapy.37 There was also a significant post-chemotherapy reduction in the expression levels of TP and MRP1 in that study.37 In patients receiving chemo-radiotherapy, an increase in TS and the excision repair cross-complementing-group 1 gene (ERCC1) mRNA expressions was associated with poorer survival.38 In a similar study of chemo-radiotherapy, reduction in the expression levels of TS in addition to DPD, ERCC1, (GSTPi), ERBB2/neu and EGFR was associated with tumour response.39 In another study, higher expression of methylenetetrahydrofolate reductase (MTHFR) was associated with good histopathological responses in contrast to tumours with low MTHFR expression.40

    Protein expression

    In gastric cancer, high TS protein expression predicted resistance to high-dose 5-FU and leucovorin chemotherapy and correlated with poorer survival.41 Low TS level, on the other hand, was associated with tumour response.42 43 TP expression in tumours was associated with tumour response compared to TP negative tumours.44 Similar positive correlation between TP expression and response was also demonstrated in other studies.45 46 A low DPD level has also been shown to be associated with tumour response.42

    DNA REPAIR PATHWAY

    A summary of studies relating to the DNA repair pathway is shown in table 2.

    View this table:
    Table 2 Studies relating to the DNA repair pathway

    DNA-damaging agents

    Some agents that damage DNA are shown in fig 4. Cisplatin and oxaliplatin are widely used in the treatment of cancer. Cytotoxicity of platinating agents is predominantly due to their ability to bind to DNA and produce cross-links.47 Generation of free radicals may also contribute to their cytotoxicity.48 Although the molecular mechanisms of resistance to platinum are complex49 50 it is clear that the DNA repair capacity may have a major bearing upon the therapeutic efficacy in cancer cells.50 51 The nucleotide excision repair (NER) pathway is the predominant DNA repair pathway that is involved in the repair of bulky DNA-damaging lesions induced by platinating agents.52 53 NER is a highly conserved, versatile and robust DNA repair pathway that deals with a wide range of DNA lesions such as those that distort the DNA double helix, DNA base pairing and interfere with DNA replication and transcription. The most common lesions processed by NER include those induced by platinating agents. The molecular mechanism of mammalian NER is complex and involves several proteins with distinct functions and interacting partners. Detailed discussion is beyond the scope of this article and we provide a summary of the key players in NER. Two sub-pathways of NER have been described: (1) the transcription-coupled nucleotide excision repair (TCR) pathway that targets lesions specifically in the transcribed strand of expressed genes; and (2) the global genome nucleotide excision repair (GGR) pathway that deals with lesions in the rest of the genome. The basic steps of global genome nucleotide excision repair are DNA damage recognition by the xeroderma pigmentosum complementation group C–RAD23 homologue B (XPC–HR23B) complex (damaged DNA binding factor (DDB) may also be involved at this stage), lesion demarcation and verification by TFIIH complex (cdk7, cyclin H, maturation-associated protein 1 (MAT1), XPB, XPD, p34, p44, p52 and p62), assembly of a pre-incision complex (replication protein A (RPA), XPA and XPG), DNA opening by XPB and XPD helicases, dual incision by ERCC1–XPF, XPG endonucleases typically a few bases away from the lesion to form an oligomer of 24–32 nucleotides, release of the excised oligomer, repair synthesis to fill in the resulting gap (RPA, replication factor C (RFC), proliferating cell nuclear antigen (PCNA), Pol δ/ϵ), and ligation (ligase I). The TCR pathway probably utilises translocating RNA polymerase II to detect lesions in the template. A role for CSA and CSB has also been suggested in DNA damage recognition in TCR. Subsequent steps in the TCR pathway are similar to GGR.49 54 Free radicals generated either by radiation or platinating agents may also induce oxidative DNA base damage that are predominantly repaired in base excision repair (BER) machinery.55 Although there is more than one sub-pathway of BER, in most cases excision of a damaged base by a DNA glycosylase enzyme leads to formation of a potentially cytotoxic apurinic/apyrimidinic (AP) site intermediate. This is a target for an AP endonuclease, which cleaves the phosphodiester backbone on the 5′ side of the AP site via a hydrolytic mechanism. The action of APE1 on an AP site generates a strand break with a 3′-hydroxyl terminus, which can prime DNA repair synthesis, and a 5′-deoxyribose phosphate (5′-dRp) terminus. The 5′-dRp residue is removed by the dRp lyase domain present in DNA polymerase β, the enzyme that also performs the task of filling in the single base gap thus formed. Repair is then completed by ligation of the nick, which is generally catalysed by DNA ligase III in association with its binding partner x ray cross-complementing group 1 (XRCC1).54 Among the various factors involved in BER, the role of XRCC1 is particularly relevant to this article. XRCC1 acts as a “scaffold” for recruiting BER proteins.55 Ionising agents induce double-strand DNA breaks that are repaired by non-homologous end joining, homologous recombination and single-strand annealing.20 Glutathione S-transferases (GSTs) are crucial enzymes involved in detoxification of platinating agents. Among the various isoforms of GSTs, GST pi has been shown to bind to platinum and also allows its export from the cytosol using an MDR energy-dependent efflux pump mechanism. Other chemotherapeutic agents used in the treatment of oesophago-gastric cancer include docetaxel (a taxane analogue that inhibits microtubule disassembly)56 and irinotecan (a topoisomerase I inhibitor).57

    Figure 4
    Figure 4 Pathways involved in the repair of platinating agents and ionising radiation are summarised here. Predictive markers within the pathways are shown in bold. AP, apurinic/apyrimidinic; ERCC, excision repair cross-complementing group; FEN-1, flap structure specific endonuclease 1; PCNA, proliferating cell nuclear antigen; RAD23B, RAD23 homolog B; RPA, replication protein A; RFC, replication factor C; XPA and XPC, xeroderma pigmentosum complementation groups A and C; XRCC1, x ray cross-complementation group 1. The TFIIH complex is cdk7, cyclin H, maturation-associated protein 1 (MAT1), xeroderma pigmentosum complementation groups B and D (XPB and XPD), p34, p44, p52 and p62.

    Proficient DNA repair in cancer cells may lead to enhanced repair of DNA damage induced by cytotoxic agents thereby contributing to therapeutic resistance. On the other hand, sub-optimal DNA repair in normal tissue may negatively impact on normal tissue tolerance. Alterations in the DNA repair capacity has been described at a genetic level (eg, polymorphisms), transcriptional levels or at protein expression level.

    Polymorphisms

    XRCC1 Arg399Gln were measured in 47 patients with gastric cancer treated with platinum-based chemotherapy and the median overall survival time was longer in patients with favourable allele G in codon 399 of XRCC1 (40 of 47 patients) than in others.58 However, another study in patients with advanced gastric cancer receiving platinum-based chemotherapy, no association with XRCC1 Arg399Gln was seen.59 In oesophageal cancer, however, XRCC1 Arg399Gln polymorphism was significantly associated with the absence of pathological complete response.60 Polymorphism of the ERCC2 (also known as XPD) gene (Lys751Gln) may be associated with functional alterations in NER. A study by Zarate et al61 investigated ERCC2 Lys751Gln polymorphism in 44 patients with gastric cancer who had received adjuvant chemo-radiotherapy (5-FU and leucovorin plus radiation). Of the 16 patients who had relapse of their disease, 12 showed the Lys/Lys genotype, implying that increased DNA repair capacity associated with the Lys/Lys phenotype may reduce the cytotoxic effect of chemo-radiotherapy. The Lys polymorphism was an independent predictor of high-risk relapse free survival. This study suggests that ERCC2 Lys751Gln polymorphism may be useful in predicting patients who will benefit from adjuvant chemo-radiotherapy.61 However, another study in 52 patients with advanced gastric cancer treated with chemotherapy, polymorphism of ERCC2 (Gln751Lys) did not correlate with histopathological response or survival.26 This study also investigated the role of other polymorphisms such as GSTP1- Ile105Val, GSTT1 and GSTM1 deletion, TS-5′ UTR 2R/3R; TS-5′ G/C; TS-3′ UTR 1494del6, MTHFR-C677T, ERCC1-C118T and ERCC2-Gln751Lys.26 Similarly, another study by Ruzzo et al59 of 13 polymorphisms in nine genes (TS, MTHFR, XPD, ERCC1, XRCC1, XRCC3, GSTPI, GSTTI AND GSTMI) showed no correlation with clinical outcomes. Although it is too early to draw firm conclusions, the current data are in some instances contradictory, but do provide a framework on which to base future larger, prospective studies, aimed at clarifying the role of DNA repair and redox regulating genetic polymorphisms in oesophago-gastric cancer therapy.

    Messenger RNA expression

    In a recent study of 76 patients with advanced gastric cancer treated with oxaliplatin-based chemotherapy, median survival time of patients with low ERCC1 levels was significantly longer than those with high levels of ERCC1 (15.8 months vs 6.2 months) implying that alternative chemotherapy regimes should be considered for patients with high ERCC1 mRNA expressing tumours.62 In another study of 140 patients with unresectable gastric cancer, low ERCC1 correlated with a higher response rate (55.6% vs 18.8%) in patients receiving platinum-based chemotherapy. In addition, low MTHFR expression correlated with a higher response rate (44.9% vs 6.3%) in patients who received S-1 monotherapy. High ERCC1, high DPD, low EGFR and an elevated serum alkaline phosphatase level were significant predictors of poor survival in that study.63 Similarly, in another study, median mRNA levels of ERCC1 in tumours that showed objective response to neoadjuvant cisplatin/5-FU chemotherapy were 2-fold lower that than in tumours that were not responsive to chemotherapy. The median survival in the low ERCC1 mRNA group was substantially more than the high ERCC1 mRNA group.64 These studies in gastric cancer therefore provide consistent evidence that ERCC1 expression levels may predict response to platinum-based chemotherapy. In oesophageal cancer, however, the evidence for ERCC1 is less convincing. In a study by Langer et al,40 the mRNA expression levels of ERCC1 did not correlate with neoadjuvant chemotherapy response. In another study of 36 oesophageal tumours treated with neoadjuvant chemo-radiotherapy, relative expression levels of ERCC1 mRNA predicted minor histopathological response following chemo-radiotherapy.65 A study by Leichman et al66 suggests that a higher level of XPA was related to shorter survival in this study and the expression of ERCC-1 declined from initial diagnosis to the end of treatment. Another study suggests that higher ERCC1 expression in tumours may predict poor response to chemo-radiotherapy.67 High ERCC1 mRNA expression was a statistically significant predictor of decreased survival and was also associated with an approximately 2-fold increase in the risk of cancer recurrence in another study.38

    Protein expression

    Immunohistochemical analyses of ERCC1 expression in tumours of 64 patients with advanced gastric cancer treated with 5-FU/oxaliplatin chemotherapy suggest that those patients without ERCC1 expression were more likely to respond to chemotherapy and this was also associated with significantly longer median overall survival.68 Similarly, in a study of oesophageal cancer, ERCC1 negative or TS negative tumours were more likely to achieve a pathological major response. On multivariate analysis, ERCC1 was found to be the only independent variable predicting pathological response. In addition, patients with ERCC1 negative tumours showed improved overall survival and disease-free survival.69 A study by Noguchi et al70 investigated the role of DNA PKcs (that is involved in DNA double-strand break repair) expression in patients with oesophageal cancer who received either definitive chemo-radiotherapy or neoadjuvant chemo-radiotherapy.70 Tumours with high DNA PKcs expression showed an increased sensitivity to chemo-radiotherapy than the low expression group in that study.

    FACTORS INVOLVED IN APOPTOSIS

    A summary of studies related to factors involved in apoptosis and biomarker response is shown in table 3. The main aim of cytotoxic therapy is to kill cancer cells. However, the mechanism by which cancer cells die is complex and may include necrosis, mitotic catastrophe and apoptosis (programmed cell death). It is widely accepted that chemotherapy and irradiation induce cell death mainly through apoptosis. There are at least two pathways of apoptosis in cells: (1) the extrinsic pathway that is mediated by death receptors on the cell surface; and (2) the intrinsic pathway that is mediated by the mitochondria. Chemotherapy and irradiation initiate apoptosis through the mitochondrial (intrinsic) pathway. However, this is tightly controlled by pro-apoptotic and anti-apoptotic factors. For example, over-expression of inhibitors of apoptosis (such as survivin) or downregulation of pro-apoptotic factors (such as Bax) may lead to relative resistance to chemotherapy in cancer. Therefore factors involved in apoptosis may be important predictive markers in oesophago-gastric cancer.71 72

    View this table:
    Table 3 Studies involving factors in apoptosis and biomarker response

    Survivin

    Survivin is an important member of the inhibitor-of-apoptosis (IAP) family that inhibits the activation of downstream effectors of apoptosis such as the caspases.73 Survivin expression may cause resistance to cisplatin-induced apoptopsis during chemotherapy. In a study of 42 patients with gastric cancer the mRNA expression of survivin and its relationship between expression and sensitivity of cisplatin (CDDP) was examined.74 Survivin was frequently upregulated in gastric cancer tissue and was negatively associated with overall survival in patients in that study.74 A study of 51 patients with oesophageal squamous cell cancer receiving cisplatin and 5-FU neoadjuvant chemotherapy showed that survivin expression in the cancer tissue in patients who achieved a partial response was significantly lower than that in patients with no change or with progressive disease.75 However, Warnecke-Eberz et al76 who examined the potential of survivin mRNA expression to predict histopathological tumour response following neoadjuvant chemo-radiotherapy in 51 patients with oesophageal cancer failed to demonstrate a correlation between survivin expression and response.

    Bax

    A study by Kang et al 77 in 63 patients with locally advanced oesophageal cancer treated by definitive chemo-radiotherapy examined protein expression of Bax, p53, Bcl-2 and galectin-3 in pre treatment biopsy specimens. Low expression of Bax was significantly correlated with a lack of clinical complete response. In addition the low expression of Bax was associated with a poor overall median survival of 8 months vs 16 months in other patients.

    View this table:
    Table 4 Studies relating to transcription factors

    Cyclooxygenase-2

    Over-expression of cyclooxygenase-2 (COX-2) may be associated with resistance to apoptosis. Although the molecular mechanism is not clear, COX-2 may be involved in the upregulation of factors that promote cell survival. High levels of COX-2 have been reported in oesophageal78 and gastric cancers.79 In a study of 52 patients with resectable oesophageal cancer following neoadjuvant chemo-radiotherapy, high COX-2 protein expression levels seem to be significantly associated with poor prognosis and a poor histopathological response to chemotherapy.80

    TRANSCRIPTION FACTORS

    A summary of studies relating to transcription factors is given in table 4.

    TP53

    TP53 (p53) is a critical transcription factor involved in key cellular functions such as in cell cycle regulation, apoptosis and DNA repair. TP53 is frequently mutated in cancer. Mutant p53 tends to be resistant to degradation and hence p53 protein expression in tumours is considered a surrogate marker of p53 mutation and hence may correlate with tumour resistance. In gastric cancer, p53 negativity has been correlated with good tumour response in four published studies.81–84 In contrast, one study of 25 patients with metastatic gastric cancer who underwent high-dose chemotherapy followed by autologous bone marrow transplantation, over-expression of p53 and the presence of p53 mutations in exons 5–9 in pre-treatment biopsy specimens was significantly associated with increased overall survival. In addition, p53 protein expression and mutation status were the only parameters associated with objective tumour regression and histological response in that study.85 Similarly, in oesophageal cancer, the results have been conflicting. Whilst one study suggests that p53 negativity may correlate positively to tumour response,86 other studies did not find any correlation between p53 protein expression and response to cytotoxic therapy.87 88 Another study also suggests that p53 mutations may correlate with complete pathological response and improved survival.89

    Nuclear factor kappa B

    Nuclear factor kappa B (NF-κB) protein expression was studied in 43 patients with oesophageal cancer (both squamous cell and adenocarcinomas) who underwent neoadjuvant chemo-radiotherapy. NF-κB positivity correlated with lack of tumour response and poor overall survival.90

    TUMOUR HYPOXIA AND ANGIOGENESIS

    Data relating to tumour hypoxia and angiogenesis are given in table 5. In the complex multistep process of carcinogenesis, genetic and epigenetic changes in cells are essential for malignant transformation. In addition, the ability to propagate and progress is critically dependent on the induction of a tumour vasculature. In fact, to maintain adequate supply of oxygen and nutrients and to remove metabolic waste products, tumours need to initiate angiogenesis to grow beyond a size of ∼2 mm3 (ie, 105 to 106 cells).91 92 Cells that are located within about 100 μm from blood vessels are normoxic, whereas tumour cells located more than 100 μm away from blood vessels become hypoxic. In tumours, hypoxia sets in early because rapidly proliferating tumour cells outgrow the capacity of the host vasculature. Hypoxic clones may survive by initiating an angiogenic pathway. Clones that are unable to initiate angiogenesis remain dormant, sometimes for months to years before they can switch to an angiogenic phenotype.93 Tumour hypoxia is an important inducer of angiogenesis within tumours. In addition, the hypoxic microenvironment selects cells that are capable of evading apoptosis, and survive in the absence of normal oxygen availability. The hypoxia-inducible transcription factor-1 (HIF-1) mediated pathway is critical for tumour angiogenesis.94 Whilst the exact molecular mechanisms of induction of HIF-1 is beyond the scope of this article, it is well know that HIF-1 transcription factors induce the expression of several genes that regulate various biological processes critical for tumour formation, such as cell proliferation, apoptosis, immortalisation, migration and angiogenesis.94 In particular, HIF-1 activates the expression of vascular endothelial growth factor (VEGF), a critical pro-angiogenic factor. In addition, HIF-1 also induces the expression of vascular endothelial growth factor receptor 1 (VEGFR1) on endothelial cells.94 HIF also regulates many other angiogenic factors, proteases and macrophage attractants, which could contribute to angiogenesis.94

    View this table:
    Table 5 Studies relating to tumour hypoxia and angiogenesis

    An immunohistochemical study of 65 patients with squamous cell carcinoma of the oesophagus has shown that HIF-1α expression was negatively related to response to chemo-radiotherapy independently of p53 and p21 expression.95 In a recent study in gastric cancer, patients with VEGF negative tumours had response rates very similar to VEGF positive tumours. However, median survival in VEGF negative tumours was significantly longer than in VEGF positive tumours. This suggests that VEGF is of prognostic significance in gastric cancer.96 Similarly, Fukuda et al42 did not find an association between VEGF and response to neoadjuvant chemotherapy.

    Immunohistochemical analyses of VEGF in pre-treatment biopsies of 52 patients with oesophageal carcinoma treated with chemo-radiotherapy have shown that VEGF over-expression may be associated with resistance to treatment. In addition, TP over-expression was also associated with poor response in that study.97 In another immunohistochemical study, lower VEGF expression was associated with complete tumour response. This study also investigated the levels of MIB-1 (indicator of proliferative activity) and CD34 (capillary density) expression and tumours with a VEGF/MIB-1 ratio of 1:6 or less prior to chemo-radiotherapy were highly predictive of the best responders.98 However, Kulke et al99 failed to demonstrate an association between VEGF and tumour response.

    MISCELLANEOUS PATHWAYS

    Studies relating to miscellaneous pathways are listed in table 6.

    View this table:
    Table 6 Studies relating to miscellaneous pathways

    Multi-drug resistance related protein-1 and multi-drug resistance related gene-1

    Chemotherapy resistance may be due to over-expression of multi-drug resistance proteins belonging to the ATP binding cassette drug transporter family. These carrier proteins cause an efflux of drugs through an energy (ATP) consuming process. MRP-1 has been shown to be responsible for resistance to a number of chemotherapy agents including cisplatin.100 However, surprisingly, a study in oesophageal adenocarcinoma suggests that high MRP-1 gene expression may be associated with higher response to chemotherapy.40 This intriguing study also suggests that MRP-1 levels may also be associated with prolonged survival. Another study revealed significant downregulation of MRP-1 and no change in MDG-1 expression after chemotherapy.37

    Signal transduction

    Receptor tyrosine kinases have critical roles in cancer pathogenesis and their altered expressions are frequently seen in epithelial tumours.101 In addition, cytotoxic therapy induced signal transduction that promotes cell survival may impair therapeutic efficacy of chemotherapy and radiation. Messenger RNA expression levels of c-erb-B1 and c-erb-B2 were examined in 36 consecutive patients with oesophageal cancer receiving neoadjuvant chemo-radiotherapy. Low intratumoural expression levels of c-erb-B1 were significantly associated with better histopathological response but the mRNA expression levels of c-erb-B2 in pre-treatment biopsies did not predict response to chemo-radiotherapy in that study.67 In a study by Gotoh et al102 of 62 oesophageal squamous cell carcinomas treated with definitive chemo-radiotherapy, EGFR protein expression (by immunohistochemistry) was significantly associated with complete response rate.

    Caldesmon

    Caldesmon is a regulatory factor involved in the assembly and stabilisation of microfilaments. Caldesmon may be related to cisplatin resistance in preclinical studies.103 Expression of the caldesmon gene was significantly higher in responding than non-responding patients with oesophageal adenocarcinoma treated with neoadjuvant cisplatin and 5-FU suggesting a role for caldesmon as a predictive factor.40

    Polymorphisms of glutathione S-transferases

    Glutathione S-transferases (GSTs) are involved in essential cellular functions such as phase II metabolism, stress response, cell proliferation, apoptosis, cancer pathogenesis and drug resistance.104 GSTP1 protein is involved in glutathione conjugation (and hence inactivation) of several chemotherapeutic agents. GSTP1-105 polymorphism, which may alter GSTP1 functional capacity, was studied in 50 patients with advanced gastric cancer treated with platinum-based chemotherapy. The GSTP1-105 Val/Val genotype showed a significant response rate of 67% compared to only 21% in patients harbouring at least one GSTP1-105Ile allele.26

    GADD153, p21 and c-Jun

    GADD153 (involved in cellular growth response to stress signals), p21 (a regulator of cell cycle progression) and c-Jun (early response transcription factor) were the candidate genes in a gastric cancer study that investigated mRNA expression levels during chemotherapy. Upregulation of GADD153, p21 and c-Jun was a good predictor of therapeutic outcome in that study.105

    Key points 2

    • Predictive markers are measures that help determine which patients do well with particular types of treatment

    • Computed tomography scanning is not a reliable marker of response, but recent advances in functional positron emission tomography imaging may hold promise

    • The cytotoxicity of chemotherapeutic agents used in cancer is directly related to their ability to induce DNA damage in cancer cells

    • 5-Fluorouracil-based agents are widely used in chemotherapy protocols. Factors involved in the 5-fluorouracil metabolism pathway are promising biomarkers

    • Platinating agents (cisplatin, oxaliplatin) induce bulky DNA lesions repaired by the nucleotide excision repair (NER) pathway. Factors involved in NER are an area of intense investigation

    • Other pathways involved in cellular responses to cytotoxic agents such as apoptosis, transcriptional regulation, hypoxia and angiogenesis and signal transduction factors have been investigated and are potential predictive markers.

    GENE EXPRESSION PROFILING IN OESOPHAGO-GASTRIC CANCER

    The studies described above have focussed on a single marker or a set of markers targeting a specific pathway or a set of pathways that may be involved in cellular response to cytotoxic agents. However, it is clear that the cancer phenotype is a sum total of genetic and epigenetic alterations and, therefore, the ultimate response to cytotoxic therapy in cancer is also likely to be dictated by these genetic and epigenetic changes involving perhaps several thousands of genes within the cancer genome. In addition, the germ-line make-up of the host is also likely to have a significant impact on treatment response and toxicity. Recent advances in genomics, proteomics and metabolomics have for the first time provided tools to explore these possibilities.106–108 Whilst gene expression analysis is an area of intense investigation in tumours such as breast cancer,106 such analysis is less well investigated in oesophago-gastric cancer. Nevertheless, we provide some evidence for this approach in oesophago-gastric cancer. Two recent studies in patients with oesophageal cancer treated with neoadjuvant chemo-radiotherapy provide valuable insights.109 110 In one study of nineteen patients,110 pre-treatment endoscopic biopsies were subjected to gene expression analyses and the molecular profile was correlated with pathological response. Unsupervised hierarchical cluster analyses segregated tumours into two molecular subtypes and approximately 400 genes were differentially expressed between the two groups. Pathological complete response was seen in 32% of patients (six out of nine patients); five of the six patients clustered into subtype I suggesting that gene signatures may predict response. In addition, survival was inferior in molecular subtype II. Genes involved in apoptosis, calcium homeostasis, stress response and proliferation were differentially expressed in both groups.110 In addition, downregulation of epidermal differentiation complex (EDC) genes (S100A2 and SPRR3) at chromosome 1q21 was associated with resistance in that study. A subsequent study by the same group has provided further evidence that expression levels of genes mapping within and close to the EDC may determine sensitivity to chemo-radiotherapy.109 A study in 33 oesophageal squamous cell carcinomas treated with chemo-radiotherapy identified 57 genes that correlated with short-term survival, and 120 genes with long-term survival. Genes involved in immune response were upregulated in long-term survivors and, additionally, immunohistochemical staining with an anti-CD8 antibody in these long-term survivors suggested that the effects of chemo-radiotherapy were correlated with CD8 positive staining.111 Duong et al112 were able to identify a 32-gene classifier that could predict response to neoadjuvant chemo-radiotherapy in squamous cell carcinomas.112 Of particular interest from this study was the gene Siah2, which is known to enhance hypoxia-inducible transcription factors. A decreased expression of Siah2 was seen amongst patients with squamous cell carcinoma who had a complete response assessed by CT, endoscopy/biopsy and FDG-PET.112

    The gene expression studies discussed above provide preliminary evidence that this technique may have a role as a predictive marker in oesophago-gastric cancer. However, published studies are limited and patient numbers are too small to provide convincing evidence that this approach is feasible in routine clinical practice.

    CONCLUSIONS AND FUTURE DIRECTIONS

    Cancer of the oesophagus, the gastro-oesophageal junction (GOJ) and stomach remains a major health problem worldwide. The evidence base for the optimal management of patients with operable oesophago-gastric cancer is evolving. Accepted approaches include preoperative chemotherapy followed by surgery (oesophageal squamous cell carcinoma (SCC)), chemo-radiotherapy alone (oesophageal SCC) and perioperative chemotherapy (gastric, GOJ and lower oesophageal adenocarcinoma). The underlying principles behind neoadjuvant therapy are to improve resectability of the tumour by tumour shrinkage/downstaging and to treat occult metastatic disease as early as possible. The response rates to cytotoxic therapy are between 40% and 60% in oesophago-gastric cancer. Available evidence suggests that a favourable histopathological response to cytotoxic therapy may be a useful positive predictive marker in oesophago-gastric cancer. However, the ability to predict tumour response in routine clinical practice is difficult and hence an area of intense investigation. Advances in cancer biology have enabled identification of critical pathways that are involved in processing cytotoxic lesions induced by therapeutic agents. Accordingly, several of the studies presented above have predominantly investigated selected factors as predictive biomarkers. The majority of investigations described above have been small retrospective studies and hence the scientific validity remains limited. Hence there is an urgent need for large prospective studies, ideally in the context of a clinical trial. However, the technical complexities involved in the collection of tissue from across different academic centres, data management and costs negatively impact on the conduct of such studies.

    Though pre-treatment biopsy tissue is a valuable resource for genetic, epigenetic and proteomic investigations, there is an argument that these biopsies may be restricted to the superficial aspects of the tumour and therefore may not be representative of the tumour tissue as a whole. There is also considerable debate as to which marker should be investigated (DNA analyses vs gene expression vs protein expression). Immunohistochemistry is cost effective but not quantitative, not objective and potentially time consuming. The recent advent of high-throughput tissue microarray technology may help in addressing these issues and may improve rapid analyses of several samples and several markers in a short time. Messenger RNA expression level is highly sensitive and suitable for use with very small amounts of biopsy material, but there is lack of control of sample morphology and intratumoural heterogeneity. In addition, optimisation of tissue collection protocols across different academic centres remains a formidable challenge.

    Though chemotherapy and irradiation still remain an important treatment modality in the management of patients with oesophago-gastric cancer, the recent advances in the use of targeted agents in solid tumours hold much promise for incorporation of these agents in oesophago-gastric cancer. The role of biomarkers that predict response to molecular targeted agents has been well established in breast cancer, lung cancer, brain tumours and colonic cancer. In chronic myeloid leukaemia, for example, BCR–ABL translocation predicts response to imatinib mesylate. It is clear from these studies that predictive biomarkers may be critical for identifying those patients who may benefit from these expensive drugs, which are being increasingly used in cancer. However, cost effectiveness precludes their widespread use and in this context the use of predictive biomarkers is anticipated to produce huge benefits to both the patient and the health economy. The role of such targeted agents is less well investigated in oesophago-gastric cancer. Incorporation of targeted agents in the context of clinical trials has been slow in oesophago-gastric cancer. However, there is now considerable enthusiasm as evidenced by the recently initiated multi-centre phase III trial of bevacizumab (monoclonal antibody that targets VEGF) in combination with cytotoxic chemotherapy in operable gastric cancer in the UK.

    It has been previously argued that a “data-driven” (unbiased) approach to prognostic biomarker discovery may not be applicable to predictive marker discovery as there are considerable challenges with regards to obtaining serial tumour tissue in solid tumours in general.106 However, oesophago-gastric cancer provides an ideal situation for identifying such “data-driven” gene expression signatures for predicting response as illustrated by gene expression studies described above. However, this approach has not been widely adopted and the current emphasis remains on pathway specific approaches to biomarker discovery.

    The post-genomic era has opened new avenues for investigation which promise to provide a global view of the cancer cell to enable personalised cancer medicine.106–108 In parallel, it has to be emphasised that new imaging technologies also provide the practising physician with an opportunity to visualise biological processes and potential biomarkers to plan treatment decisions.113 Functional imaging is a relatively non-invasive technology as evidenced by studies conducted using PET scanning in oesophago-gastric cancer patients presented above. The potential for PET scanning to investigate anti-cancer drug resistance is immense.114 For example, tumour hypoxia is an important resistance mechanism to cytotoxic therapy. The use of radiolabelled 2-nitroimidazole probes (eg, 18F-misonidazole (18F-MISO)) that undergo reduction selectively in hypoxic cells enables the visualisation and quantification of hypoxic areas in tumour cells.114 The potential role of 18F-MISO PET scanning was recently demonstrated in patients with head and neck cancer undergoing radiotherapy.115

    In conclusion, there is evolving evidence for the role of predictive biomarkers in cancer in general and oesophago-gastric cancer in particular. However, whether predictive markers will be routinely incorporated in clinical practice remains to be seen. Biomarker research is expensive and the data generated from these investigations are complex. It is clear that a concerted international effort between academia and industry is critical if personalised medicine as a practical reality for our cancer patients is to be realised.

    Acknowledgments

    We thank members of the Laboratory of Molecular Oncology, School of Molecular Medical Sciences, University of Nottingham for useful discussions.

    REFERENCES

    1. ↵
      1. Devesa SS,
      2. Blot WJ,
      3. Fraumeni JF Jr
      . Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049–53.
      OpenUrlCrossRefPubMedWeb of Science
      1. Blot WJDS,
      2. Kneller RW
      . Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991;265:1287–9.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Holmes RS,
      2. Vaughan TL
      . Epidemiology and pathogenesis of esophageal cancer. Semin Radiat Oncol 2007;17:2–9.
      OpenUrlCrossRefPubMedWeb of Science
      1. Wanebo HJ
      , Kennedy KB, Chmiel J. Cancer of the stomach. A patient care study by the American College of Surgeons. Ann Surg 1993;218:583–92.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Scartozzi M,
      2. Galizia E,
      3. Freddari F,
      4. et al
      .Molecular biology of sporadic gastric cancer: prognostic indicators and novel therapeutic approaches. Cancer Treat Rev 2004;30:451–9.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      1. Law SY
      , Fok M, Cheng SW, Wong J. A comparison of outcome after resection for squamous cell carcinomas and adenocarcinomas of the esophagus and cardia. Surg Gynecol Obstet 1992;175:107–12.
      OpenUrlPubMedWeb of Science
      1. Lerut TDLP,
      2. Coosemans W,
      3. Van Raemdonck D,
      4. et al
      .Surgical strategies in esophageal carcinoma with emphasis on radical lymphadenectomy. Ann Surg 1992;216:583–90.
      OpenUrlCrossRefPubMedWeb of Science
    8. Orringer MB. Multimodality therapy for esophageal carcinoma – an update. Chest 1993;103:406S–9S.
      OpenUrl
    9. ↵
      1. Lieberman MDSC,
      2. Bleckner S,
      3. Butt M
      . Carcinoma of the esophagus. Prognostic significance of histologic type. J Thoracic Cardiovas Surg 1995;109:130–8.
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Cunningham D,
      2. Allum WH,
      3. Stenning SP,
      4. et al
      .Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med 2006;355:11–20.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Bancewicz J,
      2. Clark PI,
      3. Smith DB,
      4. et al
      .Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. Lancet 2002;359:1727–33.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Mandard AM,
      2. Dalibard F,
      3. Mandard JC,
      4. et al
      .Pathologic assessment of tumor regression after preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer 1994;73:2680–6.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Becker K,
      2. Mueller JD,
      3. Schulmacher C,
      4. et al
      .Histomorphology and grading of regression in gastric carcinoma treated with neoadjuvant chemotherapy. Cancer 2003;98:1521–30.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Ott K,
      2. Weber WA,
      3. Lordick F,
      4. et al
      .Metabolic imaging predicts response, survival, and recurrence in adenocarcinomas of the esophagogastric junction. J Clin Oncol 2006;24:4692–8.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Weber WA,
      2. Ott K,
      3. Becker K,
      4. et al
      .Prediction of response to preoperative chemotherapy in adenocarcinomas of the esophagogastric junction by metabolic imaging. J Clin Oncol 2001;19:3058–65.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Lordick F,
      2. Ott K,
      3. Krause BJ,
      4. et al
      .PET to assess early metabolic response and to guide treatment of adenocarcinoma of the oesophagogastric junction: the MUNICON phase II trial. Lancet Oncol 2007;8:797–805.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Ott K,
      2. Fink U,
      3. Becker K,
      4. et al
      .Prediction of response to preoperative chemotherapy in gastric carcinoma by metabolic imaging: results of a prospective trial. J Clin Oncol 2003;21:4604–10.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Kim MK,
      2. Ryu JS,
      3. Kim SB,
      4. et al
      .Value of complete metabolic response by (18)F-fluorodeoxyglucose-positron emission tomography in oesophageal cancer for prediction of pathologic response and survival after preoperative chemoradiotherapy. Eur J Cancer 2007;43:1385–91.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Ott K,
      2. Herrmann K,
      3. Lordick F,
      4. et al
      .Early metabolic response evaluation by fluorine-18 fluorodeoxyglucose positron emission tomography allows in vivo testing of chemosensitivity in gastric cancer: long-term results of a prospective study. Clin Cancer Res 2008;14:2012–8.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Madhusudan S,
      2. Hickson ID
      . DNA repair inhibition: a selective tumour targeting strategy. Trends Mol Med 2005;11:503–11.
      OpenUrlCrossRefPubMed
    21. ↵
      1. Pinedo HM,
      2. Peters GF
      . Fluorouracil: biochemistry and pharmacology. J Clin Oncol 1988;6:1653–64.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. Longley DB,
      2. Harkin DP,
      3. Johnston PG
      . 5-fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer 2003;3:330–8.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. An Q,
      2. Robins P,
      3. Lindahl T,
      4. et al
      .5-Fluorouracil incorporated into DNA is excised by the Smug1 DNA glycosylase to reduce drug cytotoxicity. Cancer Res 2007;67:940–5.
      OpenUrlAbstract/FREE Full Text
    24. ↵
      1. Venturini M
      . Rational development of capecitabine. Eur J Cancer 2002;38(Suppl 2):3–9.
      OpenUrlPubMed
    25. ↵
      1. Schoffski P
      . The modulated oral fluoropyrimidine prodrug S-1, and its use in gastrointestinal cancer and other solid tumors. Anti-Cancer Drugs 2004;15:85–106.
      OpenUrlCrossRefPubMed
    26. ↵
      1. Goekkurt E,
      2. Hoehn S,
      3. Wolschke C,
      4. et al
      .Polymorphisms of glutathione S-transferases (GST) and thymidylate synthase (TS) – novel predictors for response and survival in gastric cancer patients. Br J Cancer 2006;94:281–6.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Lu JW,
      2. Gao CM,
      3. Wu JZ,
      4. et al
      .Polymorphism in the 3′-untranslated region of the thymidylate synthase gene and sensitivity of stomach cancer to fluoropyrimidine-based chemotherapy. J Hum Genet 2006;51:155–60.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Liao Z,
      2. Liu H,
      3. Swisher SG,
      4. et al
      .Polymorphism at the 3′-UTR of the thymidylate synthase gene: a potential predictor for outcomes in Caucasian patients with esophageal adenocarcinoma treated with preoperative chemoradiation. Int J Radiat Oncol Biol Phys 2006;64:700–8.
      OpenUrlPubMed
    29. ↵
      1. Ott K,
      2. Vogelsang H,
      3. Marton N,
      4. et al
      .The thymidylate synthase tandem repeat promoter polymorphism: A predictor for tumor-related survival in neoadjuvant treated locally advanced gastric cancer. Int J Cancer 2006;119:2885–94.
      OpenUrlCrossRefPubMed
    30. ↵
      1. Metzger R,
      2. Leichman CG,
      3. Danenberg KD,
      4. et al
      .ERCC1 mRNA levels complement thymidylate synthase mRNA levels in predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy. J Clin Oncol 1998;16:309–16.
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Lenz HJ,
      2. Leichman CG,
      3. Danenberg KD,
      4. et al
      .Thymidylate synthase mRNA level in adenocarcinoma of the stomach: a predictor for primary tumor response and overall survival. J Clin Oncol 1996;14:176–82.
      OpenUrlAbstract
    32. ↵
      1. Ichikawa W,
      2. Takahashi T,
      3. Suto K,
      4. et al
      .Simple combinations of 5-FU pathway genes predict the outcome of metastatic gastric cancer patients treated by S-1. Int J Cancer 2006;119:1927–33.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Napieralski R,
      2. Ott K,
      3. Kremer M,
      4. et al
      .Combined GADD45A and thymidine phosphorylase expression levels predict response and survival of neoadjuvant-treated gastric cancer patients. Clin Cancer Res 2005;11:3025–31.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Ishikawa Y,
      2. Kubota T,
      3. Otani Y,
      4. et al
      .Dihydropyrimidine dehydrogenase and messenger RNA levels in gastric cancer: possible predictor for sensitivity to 5-fluorouracil. Jpn J Cancer Res 2000;91:105–12.
      OpenUrlCrossRefWeb of Science
    35. ↵
      1. Terashima M,
      2. Irinoda T,
      3. Fujiwara H,
      4. et al
      .Roles of thymidylate synthase and dihydropyrimidine dehydrogenase in tumor progression and sensitivity to 5-fluorouracil in human gastric cancer. Anticancer Res 2002;22:761–8.
      OpenUrlPubMed
    36. ↵
      1. Kodera Y,
      2. Ito S,
      3. Fujiwara M,
      4. et al
      .Gene expression of 5-fluorouracil metabolic enzymes in primary gastric cancer: correlation with drug sensitivity against 5-fluorouracil. Cancer Lett 2007;252:307–13.
      OpenUrlCrossRefPubMed
    37. ↵
      1. Langer R,
      2. Specht K,
      3. Becker K,
      4. et al
      .Comparison of pretherapeutic and posttherapeutic expression levels of chemotherapy-associated genes in adenocarcinomas of the esophagus treated by 5-fluorouracil- and cisplatin-based neoadjuvant chemotherapy. Am J Clin Pathol 2007;128:191–7.
      OpenUrlAbstract/FREE Full Text
    38. ↵
      1. Joshi MB,
      2. Shirota Y,
      3. Danenberg KD,
      4. et al
      .High gene expression of TS1, GSTP1, and ERCC1 are risk factors for survival in patients treated with trimodality therapy for esophageal cancer. Clin Cancer Res 2005;11:2215–21.
      OpenUrlAbstract/FREE Full Text
    39. ↵
      1. Schneider S,
      2. Uchida K,
      3. Brabender J,
      4. et al
      .Downregulation of TS, DPD, ERCC1, GST-Pi, EGFR, and HER2 gene expression after neoadjuvant three-modality treatment in patients with esophageal cancer. J Am Coll Surg 2005;200:336–44.
      OpenUrlCrossRefPubMed
    40. ↵
      1. Langer R,
      2. Specht K,
      3. Becker K,
      4. et al
      .Association of pretherapeutic expression of chemotherapy-related genes with response to neoadjuvant chemotherapy in Barrett carcinoma. Clin Cancer Res 2005;11:7462–9.
      OpenUrlAbstract/FREE Full Text
    41. ↵
      1. Yeh KH,
      2. Shun CT,
      3. Chen CL,
      4. et al
      .High expression of thymidylate synthase is associated with the drug resistance of gastric carcinoma to high dose 5-fluorouracil-based systemic chemotherapy. Cancer 1998;82:1626–31.
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      1. Fukuda H,
      2. Takiguchi N,
      3. Koda K,
      4. et al
      .Thymidylate synthase and dihydropyrimidine dehydrogenase are related to histological effects of 5-fluorouracil and cisplatin neoadjuvant chemotherapy for primary gastric cancer patients. Cancer Invest 2006;24:235–41.
      OpenUrlCrossRefPubMed
    43. ↵
      1. Alexander HR,
      2. Grem JL,
      3. Hamilton JM,
      4. et al
      .Thymidylate synthase protein expression: Association with response to neoadjuvant chemotherapy and resection for locally advanced gastric and gastroesophageal adenocarcinoma. Cancer J Sci Am 1995;1:49–54.
      OpenUrlPubMed
    44. ↵
      1. Koizumi W,
      2. Saigenji K,
      3. Nakamaru N,
      4. et al
      .Prediction of response to 5′-deoxy-5-fluorouridine (5′-DFUR) in patients with inoperable advanced gastric cancer by immunostaining of thymidine phosphorylase/platelet-derived endothelial cell growth factor. Oncology 1999;56:215–22.
      OpenUrlCrossRefPubMedWeb of Science
    45. ↵
      1. Nishina T,
      2. Hyodo I,
      3. Miyaike J,
      4. et al
      .The ratio of thymidine phosphorylase to dihydropyrimidine dehydrogenase in tumour tissues of patients with metastatic gastric cancer is predictive of the clinical response to 5′-deoxy-5-fluorouridine. Eur J Cancer 2004;40:1566–71.
      OpenUrlCrossRefPubMed
    46. ↵
      1. Kikuyama S,
      2. Inada T,
      3. Shimizu K,
      4. et al
      .p53, bcl-2 and thymidine phosphorylase as predictive markers of chemotherapy in patients with advanced and recurrent gastric cancer. Anticancer Res 2001;21:2149–53.
      OpenUrlPubMed
    47. ↵
      1. Rabik CA,
      2. Dolan ME
      . Molecular mechanisms of resistance and toxicity associated with platinating agents. Cancer Treat Rev 2007;33:9–23.
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      1. Masuda H,
      2. Tanaka T,
      3. Takahama U
      . Cisplatin generates superoxide anion by interaction with DNA in a cell-free system. Biochem Biophys Res Commun 1994;203:1175–80.
      OpenUrlCrossRefPubMedWeb of Science
    49. ↵
      1. Gossage L,
      2. Madhusudan S
      . Current status of excision repair cross complementing-group 1 (ERCC1) in cancer. Cancer Treat Rev 2007;33:565–77.
      OpenUrlCrossRefPubMedWeb of Science
    50. ↵
      1. Siddik ZH
      . Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 2003;22:7265–79.
      OpenUrlCrossRefPubMedWeb of Science
    51. ↵
      1. Gossage L,
      2. Madhusudan S
      . Cancer pharmacogenomics: role of DNA repair genetic polymorphisms in individualizing cancer therapy. Mol Diagn Ther 2007;11:361–80.
      OpenUrlPubMed
    52. ↵
      1. Gillet LC,
      2. Scharer OD
      . Molecular mechanisms of mammalian global genome nucleotide excision repair. Chem Rev 2006;106:253–76.
      OpenUrlCrossRefPubMedWeb of Science
    53. ↵
      1. Reardon JT,
      2. Sancar A
      . Nucleotide excision repair. Prog Nucleic Acid Res Mol Biol 2005;79:183–235.
      OpenUrlCrossRefPubMedWeb of Science
    54. ↵
      1. Sancar A,
      2. Lindsey-Boltz LA,
      3. Unsal-Kacmaz K,
      4. et al
      .Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004;73:39–85.
      OpenUrlCrossRefPubMedWeb of Science
    55. ↵
      1. Hegde ML,
      2. Hazra TK,
      3. Mitra S
      . Early steps in the DNA base excision/single-strand interruption repair pathway in mammalian cells. Cell Res 2008;18:27–47.
      OpenUrlCrossRefPubMedWeb of Science
    56. ↵
      1. Deeks ED,
      2. Scott LJ
      . Docetaxel: in gastric cancer. Drugs 2007;67:1893–901.
      OpenUrlCrossRefPubMed
    57. ↵
      1. Farhat FS
      . A general review of the role of irinotecan (CPT11) in the treatment of gastric cancer. Med Oncol 2007;24:137–46.
      OpenUrlCrossRefPubMed
    58. ↵
      1. Wei J,
      2. Lin B,
      3. Zou Z,
      4. et al
      .ERCC1 mRNA expression and XRCC1 polymorphism as predictive factors for overall survival in gastric cancer patients receiving platinum based chemotherapy. J Clin Oncol 2007;25(18S):2512.
      OpenUrl
    59. ↵
      1. Ruzzo A,
      2. Graziano F,
      3. Kawakami K,
      4. et al
      .Pharmacogenetic profiling and clinical outcome of patients with advanced gastric cancer treated with palliative chemotherapy. J Clin Oncol 2006;24:1883–91.
      OpenUrlAbstract/FREE Full Text
    60. ↵
      1. Wu X,
      2. Gu J,
      3. Wu TT,
      4. et al
      .Genetic variations in radiation and chemotherapy drug action pathways predict clinical outcomes in esophageal cancer. J Clin Oncol 2006;24:3789–98.
      OpenUrlAbstract/FREE Full Text
    61. ↵
      1. Zarate RN,
      2. Arias F,
      3. Bandres E,
      4. et al
      .Xeroderma pigmentosum group D 751 polymorphism as a predictive factor in resected gastric cancer treated with chemo-radiotherapy. World J Gastroenterol 2006;12:6032–6.
      OpenUrlPubMedWeb of Science
    62. ↵
      1. Wei J,
      2. Zou Z,
      3. Qian X,
      4. et al
      .ERCC1 mRNA levels and survival of advanced gastric cancer patients treated with a modified FOLFOX regimen. Br J Cancer 2008;98:1398–402.
    63. ↵
      1. Matsubara J,
      2. Nishina T,
      3. Yamada Y,
      4. et al
      .Impacts of excision repair cross-complementing gene 1 (ERCC1), dihydropyrimidine dehydrogenase, and epidermal growth factor receptor on the outcomes of patients with advanced gastric cancer. Br J Cancer 2008;98:832–9.
      OpenUrlCrossRefPubMedWeb of Science
    64. ↵
      1. Metzger R,
      2. Danenberg K,
      3. Leichman CG,
      4. et al
      .High basal level gene expression of thymidine phosphorylase (platelet-derived endothelial cell growth factor) in colorectal tumors is associated with nonresponse to 5-fluorouracil. Clin Cancer Res 1998;4:2371–6.
      OpenUrlAbstract/FREE Full Text
    65. ↵
      1. Warnecke-Eberz U,
      2. Metzger R,
      3. Miyazono F,
      4. et al
      .High specificity of quantitative excision repair cross-complementing 1 messenger RNA expression for prediction of minor histopathological response to neoadjuvant radiochemotherapy in esophageal cancer. Clin Cancer Res 2004;10:3794–9.
      OpenUrlAbstract/FREE Full Text
    66. ↵
      1. Leichman L,
      2. Lawrence D,
      3. Leichman CG,
      4. et al
      .Expression of genes related to activity of oxaliplatin and 5-fluorouracil in endoscopic biopsies of primary esophageal cancer in patients receiving oxaliplatin, 5-flourouracil and radiation: characterization and exploratory analysis with survival. J Chemother 2006;18:514–24.
      OpenUrlPubMed
    67. ↵
      1. Miyazono F,
      2. Metzger R,
      3. Warnecke-Eberz U,
      4. et al
      .Quantitative c-erbB-2 but not c-erbB-1 mRNA expression is a promising marker to predict minor histopathologic response to neoadjuvant radiochemotherapy in oesophageal cancer. Br J Cancer 2004;91:666–72.
      OpenUrlPubMed
    68. ↵
      1. Kwon HC,
      2. Roh MS,
      3. Oh SY,
      4. et al
      .Prognostic value of expression of ERCC1, thymidylate synthase, and glutathione S-transferase P1 for 5-fluorouracil/oxaliplatin chemotherapy in advanced gastric cancer. Ann Oncol 2007;18:504–9.
      OpenUrlAbstract/FREE Full Text
    69. ↵
      1. Kim MK,
      2. Cho KJ,
      3. Kwon GY,
      4. et al
      .ERCC1 predicting chemoradiation resistance and poor outcome in oesophageal cancer. Eur J Cancer 2008;44:54–60.
      OpenUrlCrossRefPubMedWeb of Science
    70. ↵
      1. Noguchi T,
      2. Shibata T,
      3. Fumoto S,
      4. et al
      .DNA-PKcs expression in esophageal cancer as a predictor for chemoradiation therapeutic sensitivity. Ann Surg Oncol 2002;9:1017–22.
      OpenUrlCrossRefPubMed
    71. ↵
      1. Letai AG
      . Diagnosing and exploiting cancer’s addiction to blocks in apoptosis. Nat Rev Cancer 2008;8:121–32.
      OpenUrlCrossRefPubMedWeb of Science
    72. ↵
      1. Igney FH,
      2. Krammer PH
      . Death and anti-death: tumour resistance to apoptosis. Nat Rev Cancer 2002;2:277–88.
      OpenUrlCrossRefPubMedWeb of Science
    73. ↵
      1. Shin S,
      2. Sung BJ,
      3. Cho YS,
      4. et al
      .An anti-apoptotic protein human survivin is a direct inhibitor of caspase-3 and -7. Biochemistry 2001;40:1117–23.
      OpenUrlCrossRefPubMedWeb of Science
    74. ↵
      1. Nakamura M,
      2. Tsuji N,
      3. Asanuma K,
      4. et al
      .Survivin as a predictor of cis-diamminedichloroplatinum sensitivity in gastric cancer patients. Cancer Sci 2004;95:44–51.
      OpenUrlCrossRefPubMed
    75. ↵
      1. Kato J,
      2. Kuwabara Y,
      3. Mitani M,
      4. et al
      .Expression of survivin in esophageal cancer: correlation with the prognosis and response to chemotherapy. Int J Cancer 2001;95:92–5.
      OpenUrlCrossRefPubMedWeb of Science
    76. ↵
      1. Warnecke-Eberz U,
      2. Hokita S,
      3. Xi H,
      4. et al
      .Overexpression of survivin mRNA is associated with a favorable prognosis following neoadjuvant radiochemotherapy in esophageal cancer. Oncol Rep 2005;13:1241–6.
      OpenUrlPubMed
    77. ↵
      1. Kang SY,
      2. Han JH,
      3. Lee KJ,
      4. et al
      .Low expression of Bax predicts poor prognosis in patients with locally advanced esophageal cancer treated with definitive chemoradiotherapy. Clin Cancer Res 2007;13:4146–53.
      OpenUrlAbstract/FREE Full Text
    78. ↵
      1. Zimmermann KC,
      2. Sarbia M,
      3. Weber AA,
      4. et al
      .Cyclooxygenase-2 expression in human esophageal carcinoma. Cancer Res 1999;59:198–204.
      OpenUrlAbstract/FREE Full Text
    79. ↵
      1. Mao XY,
      2. Wang XG,
      3. Lv XJ,
      4. et al
      .COX-2 expression in gastric cancer and its relationship with angiogenesis using tissue microarray. World J Gastroenterol 2007;13:3466–71.
      OpenUrlPubMedWeb of Science
    80. ↵
      1. Xi H,
      2. Baldus SE,
      3. Warnecke-Eberz U,
      4. et al
      .High cyclooxygenase-2 expression following neoadjuvant radiochemotherapy is associated with minor histopathologic response and poor prognosis in esophageal cancer. Clin Cancer Res 2005;11:8341–7.
      OpenUrlAbstract/FREE Full Text
    81. ↵
      1. Boku N,
      2. Chin K,
      3. Hosokawa K,
      4. et al
      .Biological markers as a predictor for response and prognosis of unresectable gastric cancer patients treated with 5-fluorouracil and cis-platinum. Clin Cancer Res 1998;4:1469–74.
      OpenUrlAbstract
    82. ↵
      1. Nagashima F,
      2. Boku N,
      3. Ohtsu A,
      4. et al
      .Biological markers as a predictor for response and prognosis of unresectable gastric cancer patients treated with irinotecan and cisplatin. Jpn J Clin Oncol 2005;35:714–9.
      OpenUrlAbstract/FREE Full Text
    83. ↵
      1. Cascinu S,
      2. Graziano F,
      3. Del Ferro E,
      4. et al
      .Expression of p53 protein and resistance to preoperative chemotherapy in locally advanced gastric carcinoma. Cancer 1998;83:1917–22.
      OpenUrlCrossRefPubMedWeb of Science
    84. ↵
      1. Nakata B,
      2. Chung KH,
      3. Ogawa M,
      4. et al
      .p53 protein overexpression as a predictor of the response to chemotherapy in gastric cancer. Surg Today 1998;28:595–8.
      OpenUrlCrossRefPubMedWeb of Science
    85. ↵
      1. Bataille F,
      2. Rummele P,
      3. Dietmaier W,
      4. et al
      .Alterations in p53 predict response to preoperative high dose chemotherapy in patients with gastric cancer. Mol Pathol 2003;56:286–92.
      OpenUrlAbstract/FREE Full Text
    86. ↵
      1. Okumura H,
      2. Natsugoe S,
      3. Matsumoto M,
      4. et al
      .The predictive value of p53, p53R2, and p21 for the effect of chemoradiation therapy on oesophageal squamous cell carcinoma. Br J Cancer 2005;92:284–9.
      OpenUrlPubMedWeb of Science
    87. ↵
      1. Sarbia M,
      2. Ott N,
      3. Puhringer-Oppermann F,
      4. et al
      .The predictive value of molecular markers (p53, EGFR, ATM, CHK2) in multimodally treated squamous cell carcinoma of the oesophagus. Br J Cancer 2007;97:1404–8.
      OpenUrlCrossRefPubMed
    88. ↵
      1. Akamatsu M,
      2. Matsumoto T,
      3. Oka K,
      4. et al
      .c-erbB-2 oncoprotein expression related to chemoradioresistance in esophageal squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2003;57:1323–7.
      OpenUrlCrossRefPubMed
    89. ↵
      1. Gibson MK,
      2. Abraham SC,
      3. Wu TT,
      4. et al
      .Epidermal growth factor receptor, p53 mutation, and pathological response predict survival in patients with locally advanced esophageal cancer treated with preoperative chemoradiotherapy. Clin Cancer Res 2003;9:6461–8.
      OpenUrlAbstract/FREE Full Text
    90. ↵
      1. Izzo JG,
      2. Malhotra U,
      3. Wu TT,
      4. et al
      .Association of activated transcription factor nuclear factor kappab with chemoradiation resistance and poor outcome in esophageal carcinoma. J Clin Oncol 2006;24:748–54.
      OpenUrlAbstract/FREE Full Text
    91. ↵
      1. Carmeliet P
      . Mechanisms of angiogenesis and arteriogenesis. Nat Med 2000;6:389–95.
      OpenUrlCrossRefPubMedWeb of Science
    92. ↵
      1. Carmeliet P,
      2. Jain RK
      . Angiogenesis in cancer and other diseases. Nature 2000;407:249–57.
      OpenUrlCrossRefPubMedWeb of Science
    93. ↵
      1. Tonini T,
      2. Rossi F,
      3. Claudio PP
      . Molecular basis of angiogenesis and cancer. Oncogene 2003;22:6549–56.
      OpenUrlCrossRefPubMedWeb of Science
    94. ↵
      1. Harris AL
      . Hypoxia–a key regulatory factor in tumour growth. Nature Rev Cancer 2002;2:38–47.
      OpenUrlCrossRefPubMedWeb of Science
    95. ↵
      1. Sohda M,
      2. Ishikawa H,
      3. Masuda N,
      4. et al
      .Pretreatment evaluation of combined HIF-1alpha, p53 and p21 expression is a useful and sensitive indicator of response to radiation and chemotherapy in esophageal cancer. Int J Cancer 2004;110:838–44.
      OpenUrlCrossRefPubMedWeb of Science
    96. ↵
      1. Boku N,
      2. Ohtsu A,
      3. Nagashima F,
      4. et al
      .Relationship between expression of vascular endothelial growth factor in tumor tissue from gastric cancers and chemotherapy effects: comparison between S-1 alone and the combination of S-1 plus CDDP. Jpn J Clin Oncol 2007;37:509–14.
      OpenUrlAbstract/FREE Full Text
    97. ↵
      1. Shimada H,
      2. Hoshino T,
      3. Okazumi S,
      4. et al
      .Expression of angiogenic factors predicts response to chemoradiotherapy and prognosis of oesophageal squamous cell carcinoma. Br J Cancer 2002;86:552–7.
      OpenUrlCrossRefPubMed
    98. ↵
      1. Imdahl A,
      2. Bognar G,
      3. Schulte-Monting J,
      4. et al
      .Predictive factors for response to neoadjuvant therapy in patients with oesophageal cancer. Eur J Cardiothorac Surg 2002;21:657–63.
      OpenUrlAbstract/FREE Full Text
    99. ↵
      1. Kulke MH,
      2. Odze RD,
      3. Mueller JD,
      4. et al
      .Prognostic significance of vascular endothelial growth factor and cyclooxygenase 2 expression in patients receiving preoperative chemoradiation for esophageal cancer. J Thorac Cardiovasc Surg 2004;127:1579–86.
      OpenUrlCrossRefPubMed
    100. ↵
      1. Kruh GD,
      2. Zeng H,
      3. Rea PA,
      4. et al
      .MRP subfamily transporters and resistance to anticancer agents. J Bioenerg Biomembr 2001;33:493–501.
      OpenUrlCrossRefPubMedWeb of Science
    101. ↵
      1. Hynes NE,
      2. Lane HA
      . ERBB receptors and cancer: the complexity of targeted inhibitors. Nat Rev Cancer 2005;5:341–54.
      OpenUrlCrossRefPubMedWeb of Science
    102. ↵
      1. Gotoh M,
      2. Takiuchi H,
      3. Kawabe S,
      4. et al
      .Epidermal growth factor receptor is a possible predictor of sensitivity to chemoradiotherapy in the primary lesion of esophageal squamous cell carcinoma. Jpn J Clin Oncol 2007;37:652–7.
      OpenUrlAbstract/FREE Full Text
    103. ↵
      1. Scherf U,
      2. Ross DT,
      3. Waltham M,
      4. et al
      .A gene expression database for the molecular pharmacology of cancer. Nat Genet 2000;24:236–44.
      OpenUrlCrossRefPubMedWeb of Science
    104. ↵
      1. Lo HW,
      2. Ali-Osman F
      . Genetic polymorphism and function of glutathione S-transferases in tumor drug resistance. Curr Opin Pharmacol 2007;7:367–74.
      OpenUrlCrossRefPubMedWeb of Science
    105. ↵
      1. Akatsu Y,
      2. Saikawa Y,
      3. Kubota T,
      4. et al
      .Predictive value of GADD153, p21 and c-Jun for chemotherapy response in gastric cancer. Cancer Sci 2007;98:707–15.
      OpenUrlCrossRefPubMed
    106. ↵
      1. van ‘t Veer LJ,
      2. Bernards R
      . Enabling personalized cancer medicine through analysis of gene-expression patterns. Nature 2008;452:564–70.
      OpenUrl
      1. Hanash SM,
      2. Pitteri SJ,
      3. Faca VM
      . Mining the plasma proteome for cancer biomarkers. Nature 2008;452:571–9.
      OpenUrl
    108. ↵
      1. Chin L,
      2. Gray JW
      . Translating insights from the cancer genome into clinical practice. Nature 2008;452:553–63.
      OpenUrl
    109. ↵
      1. Luthra MG,
      2. Ajani JA,
      3. Izzo J,
      4. et al
      .Decreased expression of gene cluster at chromosome 1q21 defines molecular subgroups of chemoradiotherapy response in esophageal cancers. Clin Cancer Res 2007;13:912–9.
      OpenUrlAbstract/FREE Full Text
    110. ↵
      1. Luthra R,
      2. Wu TT,
      3. Luthra MG,
      4. et al
      .Gene expression profiling of localized esophageal carcinomas: association with pathologic response to preoperative chemoradiation. J Clin Oncol 2006;24:259–67.
      OpenUrlAbstract/FREE Full Text
    111. ↵
      1. Ashida A,
      2. Boku N,
      3. Aoyagi K,
      4. et al
      .Expression profiling of esophageal squamous cell carcinoma patients treated with definitive chemoradiotherapy: clinical implications. Int J Oncol 2006;28:1345–52.
      OpenUrlPubMedWeb of Science
    112. ↵
      1. Duong C,
      2. Greenawalt DM,
      3. Kowalczyk A,
      4. et al
      .Pretreatment gene expression profiles can be used to predict response to neoadjuvant chemoradiotherapy in esophageal cancer. Ann Surg Oncol 2007;14:3602–9.
      OpenUrlCrossRefPubMedWeb of Science
    113. ↵
      1. Weissleder R,
      2. Pittet MJ
      . Imaging in the era of molecular oncology. Nature 2008;452:580–9.
      OpenUrlCrossRefPubMed
    114. ↵
      1. West CM,
      2. Jones T,
      3. Price P
      . The potential of positron-emission tomography to study anticancer-drug resistance. Nat Rev Cancer 2004;4:457–69.
      OpenUrlCrossRefPubMedWeb of Science
    115. ↵
      1. Lee NY,
      2. Mechalakos JG,
      3. Nehmeh S,
      4. et al
      .Fluorine-18-labeled fluoromisonidazole positron emission and computed tomography-guided intensity-modulated radiotherapy for head and neck cancer: a feasibility study. Int J Radiat Oncol Biol Phys 2008;70:2–13.
      OpenUrlPubMed
    116. ↵
      1. Ott K,
      2. Vogelsang H,
      3. Mueller J,
      4. et al
      .Chromosomal instability rather than p53 mutation is associated with response to neoadjuvant cisplatin-based chemotherapy in gastric carcinoma. Clin Cancer Res 2003;9:2307–15.
      OpenUrlAbstract/FREE Full Text
    117. ↵
      1. Ling FC,
      2. Leimbach N,
      3. Baldus SE,
      4. et al
      .HIF-1alpha mRNA is not associated with histopathological regression following neoadjuvant chemoradiation in esophageal cancer. Anticancer Res 2006;26:4505–9.
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Funding: KRF is funded by Fresenius Biotech, Nottingham University Hospitals Charity and the Institute of Clinical Research, University of Nottingham.

    • Competing interests: None.