| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Imaging, Diagnosis, Prognosis |
1 Thoracic Oncology Program and 2 Division of Cancer Biostatistics, Duke University Medical Center, Durham, North Carolina and 3 Norris Cancer Center, University of Southern California; 4 Response Genetics, Inc., Los Angeles, California
Requests for reprints: David H. Harpole, Jr., Thoracic Oncology Program, Duke University Medical Center, Box 3617, Durham, NC 27710. Phone: 919-681-2249; Fax: 919-668-2417; E-mail: Harpo002{at}mc.duke.edu.
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Experimental Design: The original pretreatment formalin-fixed, paraffin-embedded endoscopic esophageal tumor biopsy material was obtained from 99 patients treated with concurrent cisplatin plus 5-fluorouracil plus 45 Gy radiation followed by resection at Duke University Medical Center (Durham, NC) from 1986 to 1997. cDNA was derived from the biopsy and analyzed to determine mRNA expression relative to an internal reference gene (ß-actin) using fluorescence-based, real-time reverse transcription-PCR. Possible markers of platinum chemotherapy association [glutathione S-transferase 
(GSTP1) and excision cross-complementing gene 1 (ERCC1)] and 5-fluorouracil association [thymidylate synthase 1 (TS1)] were measured.
Results: Cox proportional hazards model revealed a significant inverse, linear effect for TS1 with respect to survival (P = 0.007). An inverse relationship between TS1 expression and treatment response was also detected (P 
0.001). Univariate analysis identified an association with decreased survival for GSTP1 
3.0 (P = 0.05). In multivariate analyses, TS1 >6.0, ERCC1 >3, and GSTP1 >3 were statistically significant predictors of decreased survival (P = 0.007). Additionally, the presence of ERCC1 >3.0 or TS1 >6.0 was associated with an 
2-fold increase in the risk of cancer recurrence (P = 0.086 and 0.003, respectively).
Conclusion: The measurement of relative gene expression of molecular markers associated with chemoresistance in endoscopic esophageal tumor biopsies may be a useful tool in assessing outcome in patients with trimodality-treated esophageal cancer. These data should be validated further in larger prospective studies.
Key Words: esophageal cancer • chemoresistance • trimodality therapy
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
10%. Those patients who have undergone surgical resection with curative intent recognize only a limited increase in survival at 2 years (20-30% survival). Such a poor outcome is most likely attributed to the presence of locally advanced and undetected metastatic disease at the time of diagnosis (1, 2). Previous studies in the treatment of esophageal cancer have yielded conflicting results regarding the benefit of neoadjuvant chemotherapy with or without radiation followed by resection. This may be due to problems with the selection criteria of the subjects. These problems may be a lack of consistent pretreatment staging (no computed tomography or endoesophageal ultrasound requirement), inadequate treatment dose secondary to either the study design or the limited patient tolerance and poor performance status, or use of agents that have shown limited activity in esophageal cancer.
More recent advances in chemotherapy and radiotherapy have shown a slight improvement (2). Prospective, phase II, single institution data suggest that the neoadjuvant therapy (chemotherapy followed by resection) results in a 20% to 30% rate pathologic complete response. Only these patients enjoy an improved survival, whereas those without a pathologic complete response have a similar survival as patients treated with resection alone (3). Unfortunately, one cannot predict which patients will fall into which subset. There have been several recent prospective, randomized trials comparing chemotherapy with or without radiation followed by resection compared with resection alone. All but one has shown no significant difference in survival, but each has shown a similar complete response rate for the treatment arm (15-30%). These patients make up the only significant group of long-term survivors (4–8). Therefore, if one could identify the patients likely to respond to a particular regimen, therapy could subsequently be tailored accordingly and the survival rate could therefore be improved.
Advances in molecular pharmacology have refined the understanding of the mechanisms of action of drugs and resistance to chemotherapy. Several mechanisms of resistance have been identified among the most commonly used agents in the treatment of patients with esophageal cancer. Assessment of the probability of chemotherapy resistance by molecular marker expression may allow for the selection of a more effective chemotherapeutic regimen. The authors of the current study have previously published immunohistochemically measured protein expression for several possible markers of chemoresistance in trimodality-treated patients with esophageal cancer. Associations with decreased survival were obtained for relative overexpression of markers of platinum resistance [glutathione S-transferase (GSTP1) and multidrug resistance] and 5-fluorouracil [5-FU; thymidylate synthase 1 (TS1); ref. 9]. However, immunohistochemical determination of protein expression may have limited predictive value because it is subjective and semiquantitative. Due do these limitations, reverse transcription-PCR (RT-PCR) measuring gene expression may be a more accurate method.
In this study, we measured gene expressions of TS1, GSTP1, and excision cross-complementing gene 1 (ERCC1) by quantitative RT-PCR in pretreatment biopsies of esophageal tumor tissue specimens from patients who were scheduled to undergo neoadjuvant therapy with a regimen that included 5-FU, cisplatin, and radiation. High expressions of these genes were found to be significantly associated with patients' survival.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RNA Extraction and cDNA Synthesis. RNA was isolated from formalin-fixed, paraffin-embedded specimens using a novel, proprietary procedure (Response Genetics, Los Angeles, CA, U.S. Patent No. 6,248,535). After RNA isolation, cDNA was derived from each sample according to a previously described procedure (10).
PCR Quantification of mRNA Expression. Target cDNA sequences were amplified by quantitative PCR using a fluorescence-based real-time detection method [ABI PRISM 7700 Sequence Detection System (Taqman), Applied Biosystems, Foster City, CA] as described previously (11, 12). The 25 µL PCR reaction mixture contained 600 nmol/L of each primer (Table 1); 200 nmol/L each of dATP, dCTP, and dGTP; 400 µmol/L dUTP; 5.5 mmol/L MgCl2; and 1x Taqman buffer A containing a reference dye (all reagents were supplied by Applied Biosystems). PCR conditions were 50°C for 10 seconds and 95°C for 10 minutes followed by 42 cycles at 95°C for 15 seconds and 60°C for 1 minute.
|
2 method of Miller and Siegmund (13) and Halpern (14) was used along with defined quartiles of relative gene expression values. The Cox proportional hazards model was used to assess the prognostic importance of each marker on survival as well as any linear relationship. Variables with Ps < 0.2 were used for further multivariate analysis. |
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
0.001) was detected.
|
GST1 mRNA expression was detectable by quantitative RT-PCR in all of the 93 samples in which material was available for analysis. The median GST1 mRNA expression was 1.52 (range, 0.10-8.02). Figure 2C is a univariate analysis of GSTP1 expression with respect to survival. These data show that GSTP1 expression levels of >3.0 are associated with a decrease in survival (P = 0.05). A linear effect (continuous variable) was noted with increasing relative GSTP1 gene expression and decreasing survival.
Survival Stratified by the Number of Genes Expressed. Figure 3A shows the effect on survival of having at least one of the markers present in the determined positive at the high ranges defined in Fig. 2A-C compared with all markers being in the low ranges. Data were obtained on at least one of the markers analyzed for each of the 99 patients included in the study. A significant association between an increase in the number of markers expressed and a decrease in survival (P = 0.00035) was observed.
|
Table 6 summarizes multivariate model that incorporates TS1, ERCC1, and GSTP1 gene expressions into survival prediction. Stepwise regression Cox analysis for survival using all variables with Ps < 0.2 (GSTP1, ERCC1, TS1, and stage). Using the 2 test for the entire model, the overall significance was P = 0.007.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The method of action and treatment resistance for many therapies in various tumor types have been described by many investigators. These include cisplatin, 5-FU, and radiotherapy. However, this information is limited in esophageal cancer. This study assessed the possible relationship of molecular markers of chemotherapy resistance to survival. To date, it is the largest of its type and was done on pretreatment tissue collected from a well-defined patient population. Before therapy, all patients underwent clinical staging that included at least a chest/abdomen/pelvic computed tomogram with p.o. and i.v. contrast. Table 2 shows the stages observed before therapy. Note, the majority of the patients were T3 or N1 positive. Treatment consisted of two cycles of 5-FU given as 800 mg/m2/d x 5 or 1,000 mg/m2/d x 4 plus 75 mg/m2 cisplatin with concurrent 45 Gy radiation (radiation field included the celiac axis) followed by resection.
In the present study, we attempted to determine an association between gene expression and response to therapy. However, quantifying response to treatment of patients enrolled in a clinical phase II trial can be difficult. Radiological and clinical staging before treatment is often inaccurate, in that the presence of locally advanced and metastatic disease are not detected at the time of diagnosis, thus making it difficult to assess a clinical response when compared with the pathologic results after therapy. Although for this project we used the best clinical staging modalities available for esophageal cancer, the task was even more problematic due to a lack of a standard scale to quantify a therapy response in esophageal cancer. For example, a patient with a large T3N1 tumor may be downsized to a small T1N1 after therapy and resection but recur within 12 months of the initiation of therapy. (Thus, any measured response would have little clinical relevance.) Therefore, response was simply determined using the pathologic diagnosis at the time of resection and was defined as pathologic complete resection or no pathologic complete resection. Using this method to indicate response to therapy, the investigators determined a significant correlation between increased survival and a pathologic complete response (P = 0.007). Due to the number of deaths in the follow-up interval of causes unrelated to esophageal cancer >60 days after discharge (n = 25), mortality by any cause was not used as an end point. Therefore, the end point of this study was cancer-specific survival.
The markers analyzed have been associated previously with the biological pathway of the two chemotherapy drugs administered (5-FU and cisplatin) and data exist with respect to likely drug resistance in other tumor models, such as colorectal and gastric carcinoma. Platinum compounds, such as cisplatin and carboplatin, react with cellular DNA. This results in platinum-DNA adducts. By binding to DNA, cisplatin inhibits replication and can lead to arrest in the G2 phase of the cell cycle, thus resulting in apoptosis (20). There are two main mechanisms of cisplatin resistance. One leads to the reduction in the formation of cytotoxic DNA lesions and the other minimizes their impact. GSTP1 not only aids in detoxification of oxygen free radicals (one method of radiation injury) but also actively binds to platinum and allows it to be removed from the cytosol. In a previous study, the current authors observed an association between the overexpression of GSTP1 protein, measured by immunohistochemistry, with a significant decrease in survival (P = 0.02) in a similar population of esophageal cancer patients treated with trimodality therapy. In the current study, univariate analysis indicate that patients with a relative GSTP1 gene expression level of >3 had a decrease in survival (P = 0.05).
The nucleoside excision repair system is one of the major DNA repair systems. It consists of integrated recognition and excision of a broad variety of DNA damage, including UV lesions and bulky chemical adducts, such as those that are the result of cross-linking by platinum agents. The damaged DNA strand is incised at both sides of the lesion and is removed. The gap is filled by DNA synthesis and the remaining nick is ligated (20). To determine the underlying mechanism of the increase in nucleoside excision repair tumors resistant to platinum chemotherapy, the genes and gene products involved in nucleoside excision repair have to be analyzed. All of the nucleoside excision repair genes have been cloned and named with the xeroderma pigmentosa complementation group, with the exception of the ERCC1 gene. Thus, mutations in the xeroderma pigmentosa genes and ERCC1 are used as markers for nucleoside excision repair deficiency and treatment resistance (21, 22). Although not statistically significant, univariate analysis of relative ERCC1 gene expression in relation to survival showed that those patients with a level of >3 tended to have a decrease in survival (P = 0.086).
5-FU is an antimetabolite that interferes with DNA and RNA synthesis during the cell cycle. Its metabolite (5-fluoro-dUMP) is a tight-binding inhibitor of TS1. Overexpression of TS1 seems to be a major method of resistance to 5-FU and data from colorectal cancer patients suggest an association with TS1 and resistance to 5-FU (23, 24). Previously, we showed that overexpression of TS1 protein using immunohistochemistry was associated with a decreased survival in esophageal cancer (P = 0.04). In the current study, univariate analysis of relative TS1 gene expression and survival showed that patients with TS1 expression levels of >3.32 and >6.0 had an increase in survival (P = 0.026 and 0.003, respectively). These results prompted further investigation into a possible linear effect of an increase in gene expression of each gene with survival as a continuous variable. Significant correlation between increasing TS1 expression and decreasing survival (P = 0.007) was observed, whereas a linear effect approaching statistical significance was also noted, indicating that decreasing survival was associated with increasing gene expression of GST1 (P = 0.064) and ERCC1 (P = 0.071).
To determine a relationship between response and gene expression, contingency tables were generated. Fisher's exact test was used to determine if any statistically significant relationship existed between variables. There was not a statistically significant relationship between response and GSTP1 or ERCC1. However, the relationship between response and TS1 expression was approaching statistical significance (P = 0.07). On further analysis, an inverse relationship between levels of TS1 expression and response to therapy (P 0.001) was detected.
GSTP1 not only actively binds to platinum and allows it to be removed from the cytosol but also aids in detoxification of oxygen free radicals, which is one method of radiation injury (25). Because all of these patients were treated with concurrent 45 Gy radiation, the effect GSTP1 expression on radiotherapy should be addressed. According to our findings, increased expression is associated with decreased survival. Due to the design of this study, it is difficult to determine if GSTP1 expression affects the effectiveness of the chemotherapy or radiation alone or in combination. The literature addressing this issue is contradictory. Preclinical studies have reported that decreased expression of the GSTP1 protein GST indicate radiosensitivity (26). However, in studies of resected breast and uterine cervical cancer followed by radiotherapy, overexpression of GST indicated a increased benefit of radiotherapy (27, 28). These contradictions indicate that the complexity of cellular detoxification mechanisms requires further investigation.
Data obtained from the previous immunohistochemistry and current relative gene expression RT-PCR experiments done on formalin-fixed, paraffin-embedded tissue samples indicate that the measurement of TS1, GSTP1, and ERCC1 could be a potential tool for predicting those patients that will most likely respond to neoadjuvant therapy. Immunohistochemistry and RT-PCR are important tools in clinical diagnostics as well as in research. Although easily done on fresh, frozen tissue samples, immunohistochemistry done on formalin-fixed, paraffin-embedded tissue samples is met with some obstacles. The major factors that effect staining results are (a) tissue fixation, (b) antigen unmasking, (c) sensitivity of the detection system, (d) quality control of each assay done, and (e) standardization (29, 30). Some of these same obstacles are also faced when working with RNA extracted from formalin-fixed, paraffin-embedded tissue, but the main difference between these two methodologies is the data that are obtained. Immunohistochemistry allows the location of defined tissue antigens to be visualized through the binding of antibodies to small and unique regions on the antigens (29). Although immunohistochemistry is useful in determining the location of the marker of interest, the scoring of the slides is based on a semiquantitative scale that is not referenced to a known internal protein concentration but to a known positive control tissue. This scale measures the number of tumor cells stained but not the intensity of the staining. This may vary with the age of the paraffin blocks (30). Advances in image cytometry technology may provide a way to further standardize the scoring process allowing for improved quantification. Currently, determination of relative gene expression through quantitative real-time PCR is considered to be a more sensitive and more quantitative methodology than immunohistochemistry (31). This method uses an internal reference gene as a control for DNA quality for each sample assayed. It also acts as internal baseline measurement of the amount of DNA present, thus allowing the ability to differentiate smaller differences between samples and making these results more precise than immunohistochemistry-measured protein expression. However, the universal application of RT-PCR has not arrived in the clinical setting due to the significant cost of reagents, technology, and skilled labor.
In conclusion, our data suggest an increase in gene expression of GSTP1, ERCC1, and TS1 and that the increase in the number of the genes expressed is associated with a decrease in survival. These results seem to define a subset of patients who would gain the most benefit from neoadjuvant chemotherapy. The investigators are currently testing this conclusion in a comparison study of gene expression (of GSTP1, ERCC1, and TS1) in patients treated with neoadjuvant chemotherapy plus radiation plus surgery and those treated with surgery alone. These data should be validated in a large prospective cohort of patients.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 7/14/04; revised 11/19/04; accepted 12/23/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
relates to clinical drug resistance in non-small cell lung cancer. Oncology 1999;57:318–23.[Medline]
expression is downregulated in patents with Barrett's esophagus and esophageal adenocarcinoma. J Gastrointest Surg 2002;6:359–67.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  T. C. Krivak, K. M. Darcy, C. Tian, D. Armstrong, B. E. Baysal, H. Gallion, C. B. Ambrosone, and J. A. DeLoia Relationship Between ERCC1 Polymorphisms, Disease Progression, and Survival in the Gynecologic Oncology Group Phase III Trial of Intraperitoneal Versus Intravenous Cisplatin and Paclitaxel for Stage III Epithelial Ovarian Cancer J. Clin. Oncol., July 20, 2008; 26(21): 3598 - 3606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Kim, K.-J. Cho, G. Y. Kwon, S.-I. Park, Y. H. Kim, J. H. Kim, H.-Y. Song, J. H. Shin, H. Y. Jung, G. H. Lee, et al. Patients with ERCC1-Negative Locally Advanced Esophageal Cancers May Benefit from Preoperative Chemoradiotherapy Clin. Cancer Res., July 1, 2008; 14(13): 4225 - 4231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Breen and F. Barlesi The place of excision repair cross complementation 1 (ERCC1) in surgically treated non-small cell lung cancer Eur. J. Cardiothorac. Surg., May 1, 2008; 33(5): 805 - 811. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Yu, X. Zhang, J. Liu, P. Yuan, W. Tan, Y. Guo, T. Sun, D. Zhao, M. Yang, J. Liu, et al. Characterization of Functional Excision Repair Cross-Complementation Group 1 Variants and Their Association with Lung Cancer Risk and Prognosis Clin. Cancer Res., May 1, 2008; 14(9): 2878 - 2886. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Duong, D. M. Greenawalt, A. Kowalczyk, M. L. Ciavarella, G. Raskutti, W. K. Murray, W. A. Phillips, and R. J. S. Thomas Pretreatment Gene Expression Profiles Can Be Used to Predict Response to Neoadjuvant Chemoradiotherapy in Esophageal Cancer. Ann. Surg. Oncol., December 1, 2007; 14(12): 3602 - 3609. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Handra-Luca, J. Hernandez, G. Mountzios, E. Taranchon, J. Lacau-St-Guily, J.-C. Soria, and P. Fouret Excision Repair Cross Complementation Group 1 Immunohistochemical Expression Predicts Objective Response and Cancer-Specific Survival in Patients Treated by Cisplatin-Based Induction Chemotherapy for Locally Advanced Head and Neck Squamous Cell Carcinoma Clin. Cancer Res., July 1, 2007; 13(13): 3855 - 3859. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Ceppi, M Volante, S Novello, I Rapa, K. Danenberg, P. Danenberg, A Cambieri, G Selvaggi, S Saviozzi, R Calogero, et al. ERCC1 and RRM1 gene expressions but not EGFR are predictive of shorter survival in advanced non-small-cell lung cancer treated with cisplatin and gemcitabine Ann. Onc., December 1, 2006; 17(12): 1818 - 1825. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Olaussen, A. Dunant, P. Fouret, E. Brambilla, F. Andre, V. Haddad, E. Taranchon, M. Filipits, R. Pirker, H. H. Popper, et al. DNA repair by ERCC1 in non-small-cell lung cancer and cisplatin-based adjuvant chemotherapy. N. Engl. J. Med., September 7, 2006; 355(10): 983 - 991. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Rubio-Viqueira, A. Jimeno, G. Cusatis, X. Zhang, C. Iacobuzio-Donahue, C. Karikari, C. Shi, K. Danenberg, P. V. Danenberg, H. Kuramochi, et al. An In vivo Platform for Translational Drug Development in Pancreatic Cancer Clin. Cancer Res., August 1, 2006; 12(15): 4652 - 4661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-V. CHIN, L. ALABANZA, K. FUJII, K. KUDOH, T. KITA, Y. KIKUCHI, Z. E. SELVANAYAGAM, Y. F. WONG, Y. LIN, and W. C. SHIH Application of Expression Genomics for Predicting Treatment Response in Cancer Ann. N.Y. Acad. Sci., November 1, 2005; 1058(1): 186 - 195. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
