Detection of Thymidylate Synthase Expression in Lymph Node Metastases of Colorectal Cancer Can Improve the Prognostic Information
http://www.100md.com
《临床肿瘤学》
the Department of Oncology at Radiumhemmet and Department of Surgery, Karolinska University Hospital, Stockholm, Sweden
ABSTRACT
PURPOSE: The level of thymidylate synthase (TS) in primary colorectal cancer (CRC) has been reported as a prognostic marker. The purpose of this study was to determine whether TS expression in lymph node metastases of Dukes' C CRC is a prognostic marker.
PATIENTS AND METHODS: TS expression in the primary tumor and lymph node metastases from 348 patients with Dukes' C CRC was retrospectively assessed using immunohistochemistry and the monoclonal antibody TS 106. The patients had all been enrolled onto our previous study of 862 CRC patients who were included in Nordic trials that randomly assigned the patients to either surgery alone or surgery plus adjuvant chemotherapy.
RESULTS: TS expression in lymph node metastases was a distinct prognostic marker in the entire study group for overall survival (OS; P = .02) and disease-free survival (DFS; P = .04). A low TS expression in the lymph node metastases correlated with a better clinical outcome. In the subgroup of patients treated with surgery alone, the expression of TS in lymph node metastases also had a prognostic value for OS (P = .04) and DFS (P = .03), but this was not the case for the other subgroup who received adjuvant fluorouracil-based chemotherapy (OS, P = .5; DFS, P = .2). The expression of TS in the primary tumor only had a significant prognostic value among patients who were treated with surgery alone (OS, P = .03; DFS, P = .03) and not among the entire patient population.
CONCLUSION: These data show that TS expression in lymph node metastases is a prognostic marker for patients with Dukes' C CRC.
INTRODUCTION
Colorectal cancer (CRC) is one of the most common causes of cancer death in the Western world. Despite improved diagnostic methods and surgical techniques and new chemotherapeutic agents, the CRC mortality rate is still high. The pathologic feature of stage of disease is still the most important prognostic factor in CRC.1 However, recent studies have revealed the potential use of molecular and biochemical markers to predict outcome and response to chemotherapy.2,3
The expression of thymidylate synthase (TS), which is the target enzyme for fluorouracil (FU) -based chemotherapy, has been established as a promising prognostic marker. Several studies have demonstrated that TS intensity in the primary tumor is a significant prognostic indicator in patients with CRC.4-9 In advanced CRC, TS expression in the metastases predicts a response to FU-based chemotherapy.10-18 To date, there is only one study that has shown a significant association between TS expression in the primary tumor and response to FU chemotherapy in metastatic disease.19 Unfortunately, there are still no useful molecular or biochemical markers that can identify those patients with CRC who may benefit from adjuvant FU chemotherapy.
Contradictory results have been reported as to whether TS expression in the primary tumor can predict response to FU-based chemotherapy in the adjuvant setting.6,9,20-23 We have previously presented the results from 862 CRC patients enrolled onto a randomized trial, in which TS in the primary tumor was analyzed with immunohistochemical techniques.6 In this trial, only patients with the highest TS expression (grade 3) had a longer disease-free survival (DFS) after adjuvant FU chemotherapy. These results are in accordance with some reports20-22 but at odds with some other studies.9,23
Currently, the presence of lymph node metastases in CRC is a pathologic feature that helps us to decide whether a patient should receive adjuvant chemotherapy or not. The micrometastases can have other tumor characteristics than the primary tumor. There is also often a TS heterogeneity in the primary tumor.24 In lymph nodes, occult metastases can occur that are so small that they will be overlooked in the hematoxylin and eosin–stained sections.25-27 Because the main reason for adjuvant chemotherapy is to kill micrometastases, it is important to find out more about the mechanism of relapse in local and advanced CRC.
Assessment to prognostic markers in radically operated CRC is based mostly on analysis of the primary tumor.3 The aims of this study were to define whether the TS expression in lymph node metastases correlates with the TS level in the primary tumor and to evaluate whether the TS expression in the lymph node metastases has a prognostic accuracy and, consequently, can predict outcome of adjuvant FU chemotherapy.
PATIENTS AND METHODS
As described in detail in a previous report, different Nordic groups enrolled 2,211 patients with CRC (Dukes' B and C) onto prospective clinical trials from September 1991 to December 1996.6 The patients were randomly assigned to either surgery alone or surgery plus adjuvant chemotherapy with a regimen containing FU according to the Moertel schedule with levamisole,28 the Mayo regimen with or without levamisole,29 or the Nordic schedule with or without levamisole.30 Only patients up to age 75 years and with no history of other malignancies within the last 5 years were included. The adjuvant chemotherapy was started within 10 weeks after surgery. In a previous study with 862 of the 2,211 CRC patients enrolled onto the Nordic trials, we reported the role of immunohistochemically detected TS expression in the primary tumor.6 Of the 862 trial participants, 49% had Dukes' C CRC, which was the same percentage as in the 2,211 patients in the whole Nordic trials cohort.
Among this group of 862 CRC patients, tissue samples from primary tumors and lymph node metastases were available from 348 (82%) of 423 patients with Dukes' C CRC. These samples were obtained from 28 hospitals in Sweden and two hospitals in Denmark. Patient characteristics are listed in Table 1. In the study population, 206 patients were male, and 142 were female. The median age of the patients was 64 years (range, 31 to 75 years). Parameters of clinical outcome, such as development of local recurrence, distant metastases, and date of death, were obtained from the Centers of Epidemiological Oncology in Sweden and Denmark.
Laboratory Methods
Paraffin-embedded, formalin-fixed samples from resected primary tumors and lymph node metastases were used for this study. The analysis of TS protein expression was carried out using the standard avidin-biotin peroxidase complex (Vectasin Elite ABC kit; Vector Laboratories Inc, Burlingame, CA), which is a method that has been described in detail before.24 The tumor sections were deparaffinized in xylene and then hydrated in graded ethanol. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in distilled water for 10 minutes. To reduce nonspecific background staining, the sections were exposed to 20% horse serum for 30 minutes. The tissue sections were incubated with the TS 106 monoclonal antibody31 at room temperature for 90 minutes. Thereafter, the tumor sections were rinsed and incubated with biotinylated horse antimouse secondary antibodies for 30 minutes, followed by further incubation with avidin-biotin complexes. Before counterstaining with a modified Harri's hematoxyline, the sites of bound peroxidase were visualized using 0.05% 3,3'-diaminobenzidine tetrahydrochloride.
Scoring
All tumor samples were examined under a light microscope and scored by two independent observers blinded to both clinical and pathologic data. TS expression was quantified using a grading system based on the intensity of the staining arbitrarily graded from 0 to 3. Each time a new set of tumor samples was stained, reference slices were included from tumors that had previously been classified into the four groups (0 to 3) according to the intensity of their TS staining. Low-intensity staining was defined as 0 and 1, and high-intensity staining was defined as 2 and 3.20 The highest staining intensity found in the primary tumor and the lymph node metastases was used for the classification, even if the high-staining area was small. If heterogeneous levels of TS expression were found in the tissue samples from the same patient, the levels of TS expression in the lymph node metastases and the primary tumors were defined according to the highest TS score that was recorded. The agreement in TS evaluation between the two observers was more than 90%. In case of disagreement, the final TS score was determined by consensus after re-examination.
Statistics
The Gehan-Wilcoxon univariate test32 was used to examine the possible correlation between DFS, overall survival (OS), tumor site, treatment, and TS expression. For DFS, an event was considered as recurrence of disease and death from any cause. Multivariate analyses were performed using Cox regression models.33 Survival distributions were estimated using the Kaplan-Meier method.34 The 2 test of association was used to calculate the differences in distribution between TS expression in the primary tumor and lymph node metastases.
RESULTS
Patient Characteristics
This TS study included 348 patients with Dukes C CRC who were all enrolled onto our previous report of 862 CRC patients.6 Two hundred thirty-two patients (67%) had a primary tumor located in the colon, and 116 (33%) had a primary tumor located in the rectum (Table 1). In each patient, a median of five lymph nodes (range, one to 20 lymph nodes) were resected, and the median number of lymph nodes containing tumor cells was two (range, one to 20 lymph nodes). TS expression was analyzed in the primary tumor and the lymph node metastases for all patients included in this report. Tumor recurrence was observed in 159 patients (45.7%). During a median follow-up time of 4 years (range, 0.2 to 8.1 years), 178 patients (51.1%) died, of whom 147 (82.6%) died with disseminated CRC.
TS Expression in Lymph Node Metastases and Prognosis
Entire study group. TS expression in lymph node metastases had a prognostic value for the entire study group with regard to OS (P = .02; Fig 1A) and DFS (P = .04; Fig 1B). A longer DFS and OS were observed in one third of the patients who had a low level of TS in their lymph node metastases. Among patients whose lymph node metastases expressed high TS levels, the clinical outcome was worse. In 37% of the patients in whom more than one lymph node metastasis was analyzed, we found a heterogeneous TS expression with various TS levels in different lymph node metastases from the same patient. There was a statistically significant correlation between TS expression and the number of analyzed lymph node metastases (P = .03). Sixty-six percent of the patients with a high expression of TS in the lymph node metastases had two lymph node metastases analyzed compared with 46% of the patients with a low TS expression in the lymph node metastases. However, this difference was solely a result of the number of tissue samples that were obtainable from the treating hospitals.
Irrespective of the TS expression, more than three lymph node metastases correlated with a worse clinical outcome (OS, P = .04; DFS, P = .05). However, in multivariate analyses, the level of TS expression in lymph node metastases was the only factor that remained a prognostic marker (OS, P = .02; DFS, P = .04). Heterogeneity could be studied only among those patients in whom TS was high (staining score of 2 and 3) and analyzed in two different lymph node metastases (n = 148). Patients with TS heterogeneity in their lymph node metastases had a tendency for a better prognosis than patients with TS homogeneity with respect to DFS (P = .06) but not to OS (P = .1).
Subgroup treated with surgery alone. TS expression in lymph node metastases in the subgroup of patients treated with surgery alone had a prognostic value for OS (P = .04) and DFS (P = .03), where a low level of TS correlated with a better prognosis.
Subpopulation treated with adjuvant chemotherapy. In the subgroup of patients who were randomly assigned to receive adjuvant FU-based chemotherapy, TS expression in lymph node metastases did not give any prognostic information (OS, P = .5; DFS, P = .2).
TS expression and clinical benefit of adjuvant chemotherapy. There was no difference in the clinical outcome between the patients treated with adjuvant chemotherapy compared with the patients treated with surgery alone, irrespective of the TS expression in the lymph nodes metastases.
TS Expression in the Primary Tumor and Prognosis
Entire study group. In the entire study group, the level of TS expression in the primary tumors did not have a significant prognostic value with regard to OS (P = .2; Fig 2A) and DFS (P = .2; Fig 2B). The level of TS expression was correlated with the number of primary tumor samples available for analysis (P = .04). In the group of patients with a high TS expression in the primary tumor, there were two tumor samples analyzed in 91% of the patients compared with 82% of the patients who had low TS expression in the primary tumor. However, this result was simply a function of the tumor material that was submitted by the treating institutions and available for TS analysis. Among patients with high TS expression in their primary tumors, TS heterogeneity did not change the clinical outcome.
Subgroup treated with surgery alone. In the subgroup of 178 patients who were treated with surgery alone, TS expression in the primary tumor was a prognostic marker (OS, P = .03; DFS, P = .03).
Subpopulation treated with adjuvant chemotherapy. Among the 170 patients who were randomly assigned to adjuvant treatment with FU, the level of TS did not provide any prognostic information (OS, P = .8; DFS, P = .5).
Correlation Between TS Expression in the Primary Tumor and Lymph Node Metastases
There was a significant correlation between TS expression in the primary tumor and the expression of TS in the lymph node metastases (P = .001). However, in 28% of the patients, we found different levels of TS expression in the lymph node metastases compared with the primary tumor. Among the 92 patients with a low TS expression in the primary tumor, the TS level in the lymph node metastases was an additional significant prognostic marker for cancer-specific survival (P = .02, Fig 3). Patients who expressed a low level of TS both in the primary tumor and the lymph node metastases had a better cancer-specific survival, but the difference was not statistically significant with respect to OS and DFS.
DISCUSSION
TS expression in the primary tumors of CRC has a prognostic value according to previous studies in which low TS levels in the primary tumor correlated with a better clinical outcome.4-9,35 However, some studies have reported conflicting results that are not further discussed here.23,36
It has also been reported that TS expression in CRC metastases is a predictor of responsiveness to FU chemotherapy in patients with disseminated disease. Distant metastases with a low level of TS expression respond better to FU-based chemotherapy than distant metastases with high TS levels.10-18
Nevertheless, we still lack reliable predictive factors for CRC patients in the adjuvant setting. Because the primary tumor is often heterogeneous, the characteristics found in the primary tumor can be different from the characteristics found in lymph node metastases. It seems reasonable to believe that the grade of aggressiveness in lymph node metastases has a prognostic value.
In a study with 56 rectal cancer patients (Dukes' C), we have previously reported a correlation between the TS level in the primary tumor and the level of TS expression in the lymph node metastases.4 However, other studies have failed to show a correlation in TS expression between the primary tumor and the lymph node metastases.37,38
This present study with 348 patients has revealed that TS expression in lymph node metastases of CRC has a distinct prognostic value; patients whose lymph node metastases express a low level of TS have a longer DFS and OS. The prognostic value was independent of age, sex, site of tumor, and treatment. TS in lymph node metastases was a prognostic marker for patients who were treated with surgery alone and not for patients who received adjuvant chemotherapy. One reason for this could be a change in outcome as a result of the adjuvant chemotherapy, which might have an impact on the survival differences between patients with high compared with low TS levels in the lymph node metastases. If this is true, then adjuvant FU-based chemotherapy tends to reduce any differences in outcome between patients with high versus low TS expression. However, we did not consider risk factors such as impaired performance status, poor differentiation, mucinous pathology, perforation, and proximal and distal margins.
Another possible bias to consider is the number of tumor samples available for TS analyses from each patient because the possibility of detecting areas of higher expression would be expected to increase in proportion to the number of areas examined. In this study, there was a significant correlation between the number of analyzed tissue samples and the expression of TS in the lymph node metastases (P = .03) and the primary tumors (P = .04). The number of obtainable lymph node metastases and primary tumors is a result of the material submitted by the treating institutions. For every case included in the study, each available lymph node metastasis and primary tumor was analyzed for TS expression. The mean number of lymph node metastases was the same in both the patients with high TS expression in the primary tumor and the patients with low TS expression in the primary tumor.
A tendency towards a better prognosis with respect to DFS was observed among patients with a high TS expression and a TS heterogeneity in their lymph node metastases. However, patients with a high TS expression in their primary tumors and TS heterogeneity in the primary tumors did not have a better clinical outcome. Further studies are needed to evaluate whether the intratumoral heterogeneity of TS expression can be a marker for genomic instability and prognosis.
There was also no clinical benefit of adjuvant FU chemotherapy in this subgroup of patients, which is contrary to previously and recently published studies.28,39-45 However, the primary object of this retrospective study was not to evaluate the effect of FU-based adjuvant chemotherapy. The lack of a chemotherapy effect may represent inadequate statistical power in this study, which makes it hard to draw any conclusions. The effect of adjuvant chemotherapy in CRC is also a result of the accumulated dose of chemotherapy administered to each patient. Therefore, it is hard to evaluate the efficacy of the adjuvant chemotherapy because we do not know the number of patients who received a whole chemotherapy schedule without interruption or dose reduction. Furthermore, the patients in our study also received five different treatment schedules of adjuvant chemotherapy, which might have had an impact on the results. The anomalous results with chemotherapy may provide another window through which it may be possible to examine the biology of tumor metastasis and progression, as well as adjuvant chemotherapy.
We recently published a study of 862 CRC patients in whom we analyzed the predictive value of TS in the primary tumor.6 Unexpectedly, we found that the group of patients with a low level of TS expression in the primary tumor had a worse outcome with adjuvant chemotherapy (P = .01), but this observation did not persist in our present study of 348 CRC patients who were all included in this previous report. No other published studies have found any association between low TS expression and a deleterious effect of adjuvant chemotherapy.
There are conflicting results reported on the role of TS expression in primary CRC as a predictive factor for patients treated in the adjuvant setting. The use of different technologies for the assessment of intratumoral TS levels might be a reason for this discrepancy. To date, four studies have shown a predictive value of TS expression in the primary tumor.6,20-22 In these studies, only patients whose cancer had a high TS expression were found to benefit from adjuvant treatment. Another group has reported that CRC patients whose tumors had a low TS gene polymorphism, which is associated with low TS levels, derived greater benefit from adjuvant chemotherapy.46 However, these results are discordant with some other reports using immunohistochemistry to assess TS expression, which did not show any predictive value of TS.9,23 In one of these studies, TS was assessed using two separate scoring systems.23 These different methodologies for evaluating TS yielded different results. Newer techniques using microdissection and polymerase chain reaction may prove more enlightening in the future.
This study indicates that TS assessment in locoregional lymph node metastases of CRC improves the prognostic precision compared with TS expression in the primary tumor. The patients included in our study underwent surgery between 1991 and 1996, and the median number of lymph nodes found in the surgical specimens was only five (range, one to 20 lymph nodes), which might have had an impact on the categorization of the TS cohorts. At present, colorectal surgeons are more aware of the importance of resecting a high number of lymph nodes. Pathologists have also learned the great significance of finding as many lymph nodes as possible in resected tumor specimens.
A thorough tumor staging has a crucial impact on the prognostic and predictive value of the TS expression assessed in lymph node metastases. The use of techniques to detect sentinel nodes might be one way to improve future tumor staging.47,48 Further prospective randomized studies measuring the TS expression in lymph node metastases are needed to evaluate the role of TS as a predictor of benefit from adjuvant chemotherapy and a prognostic marker in patients with locally advanced CRC.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Professor Anders Jacobsen and Professor Bengt Glimelius for valuable assistance during this project and Bo Nilsson for contributing to the statistical analysis of the data. We also thank Associate Professor Carlos Rubio and Associate Professor Johan Lindholm at the Division of Pathology, Karolinska University Hospital, Stockholm, Sweden, for support.
NOTES
Supported by grants from the King Gustaf V:s Jubilee Foundation, the Cancerf?reningen, Stockholm, the Swedish Cancer Society, the Bengt Ihre Foundation of the Swedish Society of Medicine, and the Ruth and Rickard Juhlin Foundation.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer: Adopted on May 17, 1996 by the American Society of Clinical Oncology. J Clin Oncol 14:2843-2877, 1996
Johnston PG, Allegra CJ: Colorectal cancer biology: Clinical implications. Semin Oncol 22:418-432, 1995
McDermott U, Longley DB, Johnston PG, et al: Molecular and biochemical markers in colorectal cancer. Ann Oncol 13:235-245, 2002 (suppl 4)
Edler D, Hallstrom M, Johnston PG, et al: Thymidylate synthase expression: An independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 6:1378-1384, 2000
Edler D, Kressner U, Ragnhammar P, et al: Immunohistochemically detected thymidylate synthase in colorectal cancer: An independent prognostic factor of survival. Clin Cancer Res 6:488-492, 2000
Edler D, Glimelius B, Hallstrom M, et al: Thymidylate synthase expression in colorectal cancer: A prognostic and predictive marker of benefit from adjuvant fluorouracil-based chemotherapy. J Clin Oncol 20:1721-1728, 2002
Lenz HJ, Danenberg KD, Leichman CG, et al: p53 and thymidylate synthase expression in untreated stage II colon cancer: Associations with recurrence, survival, and site. Clin Cancer Res 4:1227-1234, 1998
Van Triest B, Loftus BM, Pinedo HM, et al: Thymidylate synthase expression in patients with colorectal carcinoma using a polyclonal thymidylate synthase antibody in comparison to the TS 106 monoclonal antibody. J Histochem Cytochem 48:755-760, 2000
Allegra CJ, Paik S, Colangelo LH, et al: Prognostic value of thymidylate synthase, Ki-67, and p53 in patients with Dukes' B and C colon cancer: A National Cancer Institute-National Surgical Adjuvant Breast and Bowel Project collaborative study. J Clin Oncol 21:241-250, 2003
Kornmann M, Link KH, Lenz HJ, et al: Thymidylate synthase is a predictor for response and resistance in hepatic artery infusion chemotherapy. Cancer Lett 118:29-35, 1997
Aschele C, Debernardis D, Casazza S, et al: Immunohistochemical quantitation of thymidylate synthase expression in colorectal cancer metastases predicts for clinical outcome to fluorouracil-based chemotherapy. J Clin Oncol 17:1760-1770, 1999
Johnston PG, Lenz HJ, Leichman CG, et al: Thymidylate synthase gene and protein expression correlate and are associated with response to 5-fluorouracil in human colorectal and gastric tumors. Cancer Res 55:1407-1412, 1995
Lenz HJ, Hayashi K, Salonga D, et al: p53 point mutations and thymidylate synthase messenger RNA levels in disseminated colorectal cancer: An analysis of response and survival. Clin Cancer Res 4:1243-1250, 1998
Leichman CG, Lenz HJ, Leichman L, et al: Quantitation of intratumoral thymidylate synthase expression predicts for disseminated colorectal cancer response and resistance to protracted-infusion fluorouracil and weekly leucovorin. J Clin Oncol 15:3223-3229, 1997
Gonen M, Hummer A, Zervoudakis A, et al: Thymidylate synthase expression in hepatic tumors is a predictor of survival and progression in patients with resectable metastatic colorectal cancer. J Clin Oncol 21:406-412, 2003
Cascinu S, Aschele C, Barni S, et al: Thymidylate synthase expression in advanced colon cancer: Correlation with the site of metastasis and the clinical response to leucovorin-modulated bolus 5-fluorouracil. Clin Cancer Res 5:1996-1999, 1999
Davies MM, Johnston PG, Kaur S, et al: Colorectal liver metastasis thymidylate synthase staining correlates with response to hepatic arterial floxuridine. Clin Cancer Res 5:325-328, 1999
Salonga D, Danenberg KD, Johnson M, et al: Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase. Clin Cancer Res 6:1322-1327, 2000
Paradiso A, Simone G, Petroni S, et al: Thymidylate synthase and p53 primary tumour expression as predictive factors for advanced colorectal cancer patients. Br J Cancer 82:560-567, 2000
Johnston PG, Fisher ER, Rockette HE, et al: The role of thymidylate synthase expression in prognosis and outcome of adjuvant chemotherapy in patients with rectal cancer. J Clin Oncol 12:2640-2647, 1994
Takenoue T, Nagawa H, Matsuda K, et al: Relation between thymidylate synthase expression and survival in colon carcinoma, and determination of appropriate application of 5-fluorouracil by immunohistochemical method. Ann Surg Oncol 7:193-198, 2000
Yamachika T, Nakanishi H, Inada K, et al: A new prognostic factor for colorectal carcinoma, thymidylate synthase, and its therapeutic significance. Cancer 82:70-77, 1998
Allegra CJ, Parr AL, Wold LE, et al: Investigation of the prognostic and predictive value of thymidylate synthase, p53, and Ki-67 in patients with locally advanced colon cancer. J Clin Oncol 20:1735-1743, 2002
Edler D, Blomgren H, Allegra CJ, et al: Immunohistochemical determination of thymidylate synthase in colorectal cancer-methodological studies. Eur J Cancer 33:2278-2281, 1997
Calaluce R, Miedema BW, Yesus YW, et al: Micrometastasis in colorectal carcinoma: A review. J Surg Oncol 67:194-202, 1998
Noura S, Yamamoto H, Ohnishi T, et al: Comparative detection of lymph node micrometastases of stage II colorectal cancer by reverse transcriptase polymerase chain reaction and immunohistochemistry. J Clin Oncol 20:4232-4241, 2002
Yasuda K, Adachi Y, Shiraishi N, et al: Pattern of lymph node micrometastasis and prognosis of patients with colorectal cancer. Ann Surg Oncol 8:300-304, 2001
Laurie JA, Moertel CG, Fleming TR, et al: Surgical adjuvant therapy of large-bowel carcinoma: An evaluation of levamisole and the combination of levamisole and fluorouracil—The North Central Cancer Treatment Group and the Mayo Clinic. J Clin Oncol 7:1447-1456, 1989
Poon MA, O'Connell MJ, Moertel CG, et al: Biochemical modulation of fluorouracil: Evidence of significant improvement of survival and quality of life in patients with advanced colorectal carcinoma. J Clin Oncol 7:1407-1418, 1989
Glimelius B: Biochemical modulation of 5-fluorouracil: A randomized comparison of sequential methotrexate, 5-fluorouracil and leucovorin versus sequential 5-fluorouracil and leucovorin in patients with advanced symptomatic colorectal cancer: The Nordic Gastrointestinal Tumor Adjuvant Therapy Group. Ann Oncol 4:235-240, 1993
Johnston PG, Liang CM, Henry S, et al: Production and characterization of monoclonal antibodies that localize human thymidylate synthase in the cytoplasm of human cells and tissue. Cancer Res 51:6668-6676, 1991
Gehan EA: A generalized Wilcoxon test for comparing arbitrarily single-censored samples. Biometrika 52:202-223, 1965
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Mini E, Biondi C, Morganti M, et al: Marked variation of thymidylate synthase and folylpolyglutamate synthase gene expression in human colorectal tumours. Oncol Res 11:437-445, 1999
Sanguedolce R, Vultaggio G, Sanguedolce F, et al: The role of thymidylate synthase levels in the prognosis and the treatment of patients with colorectal cancer. Anticancer Res 18:1515-1520, 1998
Yamada H, Ichikawa W, Uetake H, et al: Thymidylate synthase gene expression in primary colorectal cancer and metastatic sites. Clin Colorectal Cancer 1:169-174, 2001
Marsh S, McKay JA, Curran S, et al: Primary colorectal tumour is not an accurate predictor of thymidylate synthase in lymph node metastasis. Oncol Rep 9:231-234, 2002
Wolmark N, Fisher B, Rockette H, et al: Postoperative adjuvant chemotherapy or BCG for colon cancer: Results from NSABP protocol C-01. J Natl Cancer Inst 80:30-36, 1988
Moertel CG, Fleming TR, MacDonald JS, et al: Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 322:352-358, 1990
Moertel C, Fleming T, MacDonald J, et al: Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: A final report. Ann Intern Med 122:321-326, 1995
NIH Consensus Conference: Adjuvant therapy for patients with colon and rectal cancer. JAMA 264:1444-1450, 1990
International Multicentre Pooled Analysis of Colon Cancer Trials Investigators: Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. Lancet 345:939-944, 1995
International Multicentre Pooled Analysis of B2 Colon Cancer Trials Investigators: Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer: International Multicentre Pooled Analysis of B2 Colon Cancer Trials (IMPACT B2) investigators. J Clin Oncol 17:1356-1363, 1999
The Colorectal Cancer Chemotherapy Study Group of Japan: Five-year results of a randomized controlled trial of adjuvant chemotherapy for curatively resected colorectal carcinoma. Jpn J Clin Oncol 25:91-103, 1995
Iacopetta B, Gieu F, Joseph D, et al: A polymorphism in the enhancer region of thymidylate synthase promoter influences the survival of colorectal patients treated with 5-fluorouracil. Br J Cancer 85:827-830, 2001
Muslow J, Winter DC, O'Keane JC, et al: Sentinel lymph node mapping in colorectal cancer. Br J Surg 90:659-667, 2003
Trocha SD, Saha SS, Wiese DW, et al: Differential expression of thymidylate synthase in colorectal tumors and matched lymph nodes: Impact on adjuvant treatment. Am Surg 69:918-922, 2003(Katarina ?hrling, David E)
ABSTRACT
PURPOSE: The level of thymidylate synthase (TS) in primary colorectal cancer (CRC) has been reported as a prognostic marker. The purpose of this study was to determine whether TS expression in lymph node metastases of Dukes' C CRC is a prognostic marker.
PATIENTS AND METHODS: TS expression in the primary tumor and lymph node metastases from 348 patients with Dukes' C CRC was retrospectively assessed using immunohistochemistry and the monoclonal antibody TS 106. The patients had all been enrolled onto our previous study of 862 CRC patients who were included in Nordic trials that randomly assigned the patients to either surgery alone or surgery plus adjuvant chemotherapy.
RESULTS: TS expression in lymph node metastases was a distinct prognostic marker in the entire study group for overall survival (OS; P = .02) and disease-free survival (DFS; P = .04). A low TS expression in the lymph node metastases correlated with a better clinical outcome. In the subgroup of patients treated with surgery alone, the expression of TS in lymph node metastases also had a prognostic value for OS (P = .04) and DFS (P = .03), but this was not the case for the other subgroup who received adjuvant fluorouracil-based chemotherapy (OS, P = .5; DFS, P = .2). The expression of TS in the primary tumor only had a significant prognostic value among patients who were treated with surgery alone (OS, P = .03; DFS, P = .03) and not among the entire patient population.
CONCLUSION: These data show that TS expression in lymph node metastases is a prognostic marker for patients with Dukes' C CRC.
INTRODUCTION
Colorectal cancer (CRC) is one of the most common causes of cancer death in the Western world. Despite improved diagnostic methods and surgical techniques and new chemotherapeutic agents, the CRC mortality rate is still high. The pathologic feature of stage of disease is still the most important prognostic factor in CRC.1 However, recent studies have revealed the potential use of molecular and biochemical markers to predict outcome and response to chemotherapy.2,3
The expression of thymidylate synthase (TS), which is the target enzyme for fluorouracil (FU) -based chemotherapy, has been established as a promising prognostic marker. Several studies have demonstrated that TS intensity in the primary tumor is a significant prognostic indicator in patients with CRC.4-9 In advanced CRC, TS expression in the metastases predicts a response to FU-based chemotherapy.10-18 To date, there is only one study that has shown a significant association between TS expression in the primary tumor and response to FU chemotherapy in metastatic disease.19 Unfortunately, there are still no useful molecular or biochemical markers that can identify those patients with CRC who may benefit from adjuvant FU chemotherapy.
Contradictory results have been reported as to whether TS expression in the primary tumor can predict response to FU-based chemotherapy in the adjuvant setting.6,9,20-23 We have previously presented the results from 862 CRC patients enrolled onto a randomized trial, in which TS in the primary tumor was analyzed with immunohistochemical techniques.6 In this trial, only patients with the highest TS expression (grade 3) had a longer disease-free survival (DFS) after adjuvant FU chemotherapy. These results are in accordance with some reports20-22 but at odds with some other studies.9,23
Currently, the presence of lymph node metastases in CRC is a pathologic feature that helps us to decide whether a patient should receive adjuvant chemotherapy or not. The micrometastases can have other tumor characteristics than the primary tumor. There is also often a TS heterogeneity in the primary tumor.24 In lymph nodes, occult metastases can occur that are so small that they will be overlooked in the hematoxylin and eosin–stained sections.25-27 Because the main reason for adjuvant chemotherapy is to kill micrometastases, it is important to find out more about the mechanism of relapse in local and advanced CRC.
Assessment to prognostic markers in radically operated CRC is based mostly on analysis of the primary tumor.3 The aims of this study were to define whether the TS expression in lymph node metastases correlates with the TS level in the primary tumor and to evaluate whether the TS expression in the lymph node metastases has a prognostic accuracy and, consequently, can predict outcome of adjuvant FU chemotherapy.
PATIENTS AND METHODS
As described in detail in a previous report, different Nordic groups enrolled 2,211 patients with CRC (Dukes' B and C) onto prospective clinical trials from September 1991 to December 1996.6 The patients were randomly assigned to either surgery alone or surgery plus adjuvant chemotherapy with a regimen containing FU according to the Moertel schedule with levamisole,28 the Mayo regimen with or without levamisole,29 or the Nordic schedule with or without levamisole.30 Only patients up to age 75 years and with no history of other malignancies within the last 5 years were included. The adjuvant chemotherapy was started within 10 weeks after surgery. In a previous study with 862 of the 2,211 CRC patients enrolled onto the Nordic trials, we reported the role of immunohistochemically detected TS expression in the primary tumor.6 Of the 862 trial participants, 49% had Dukes' C CRC, which was the same percentage as in the 2,211 patients in the whole Nordic trials cohort.
Among this group of 862 CRC patients, tissue samples from primary tumors and lymph node metastases were available from 348 (82%) of 423 patients with Dukes' C CRC. These samples were obtained from 28 hospitals in Sweden and two hospitals in Denmark. Patient characteristics are listed in Table 1. In the study population, 206 patients were male, and 142 were female. The median age of the patients was 64 years (range, 31 to 75 years). Parameters of clinical outcome, such as development of local recurrence, distant metastases, and date of death, were obtained from the Centers of Epidemiological Oncology in Sweden and Denmark.
Laboratory Methods
Paraffin-embedded, formalin-fixed samples from resected primary tumors and lymph node metastases were used for this study. The analysis of TS protein expression was carried out using the standard avidin-biotin peroxidase complex (Vectasin Elite ABC kit; Vector Laboratories Inc, Burlingame, CA), which is a method that has been described in detail before.24 The tumor sections were deparaffinized in xylene and then hydrated in graded ethanol. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in distilled water for 10 minutes. To reduce nonspecific background staining, the sections were exposed to 20% horse serum for 30 minutes. The tissue sections were incubated with the TS 106 monoclonal antibody31 at room temperature for 90 minutes. Thereafter, the tumor sections were rinsed and incubated with biotinylated horse antimouse secondary antibodies for 30 minutes, followed by further incubation with avidin-biotin complexes. Before counterstaining with a modified Harri's hematoxyline, the sites of bound peroxidase were visualized using 0.05% 3,3'-diaminobenzidine tetrahydrochloride.
Scoring
All tumor samples were examined under a light microscope and scored by two independent observers blinded to both clinical and pathologic data. TS expression was quantified using a grading system based on the intensity of the staining arbitrarily graded from 0 to 3. Each time a new set of tumor samples was stained, reference slices were included from tumors that had previously been classified into the four groups (0 to 3) according to the intensity of their TS staining. Low-intensity staining was defined as 0 and 1, and high-intensity staining was defined as 2 and 3.20 The highest staining intensity found in the primary tumor and the lymph node metastases was used for the classification, even if the high-staining area was small. If heterogeneous levels of TS expression were found in the tissue samples from the same patient, the levels of TS expression in the lymph node metastases and the primary tumors were defined according to the highest TS score that was recorded. The agreement in TS evaluation between the two observers was more than 90%. In case of disagreement, the final TS score was determined by consensus after re-examination.
Statistics
The Gehan-Wilcoxon univariate test32 was used to examine the possible correlation between DFS, overall survival (OS), tumor site, treatment, and TS expression. For DFS, an event was considered as recurrence of disease and death from any cause. Multivariate analyses were performed using Cox regression models.33 Survival distributions were estimated using the Kaplan-Meier method.34 The 2 test of association was used to calculate the differences in distribution between TS expression in the primary tumor and lymph node metastases.
RESULTS
Patient Characteristics
This TS study included 348 patients with Dukes C CRC who were all enrolled onto our previous report of 862 CRC patients.6 Two hundred thirty-two patients (67%) had a primary tumor located in the colon, and 116 (33%) had a primary tumor located in the rectum (Table 1). In each patient, a median of five lymph nodes (range, one to 20 lymph nodes) were resected, and the median number of lymph nodes containing tumor cells was two (range, one to 20 lymph nodes). TS expression was analyzed in the primary tumor and the lymph node metastases for all patients included in this report. Tumor recurrence was observed in 159 patients (45.7%). During a median follow-up time of 4 years (range, 0.2 to 8.1 years), 178 patients (51.1%) died, of whom 147 (82.6%) died with disseminated CRC.
TS Expression in Lymph Node Metastases and Prognosis
Entire study group. TS expression in lymph node metastases had a prognostic value for the entire study group with regard to OS (P = .02; Fig 1A) and DFS (P = .04; Fig 1B). A longer DFS and OS were observed in one third of the patients who had a low level of TS in their lymph node metastases. Among patients whose lymph node metastases expressed high TS levels, the clinical outcome was worse. In 37% of the patients in whom more than one lymph node metastasis was analyzed, we found a heterogeneous TS expression with various TS levels in different lymph node metastases from the same patient. There was a statistically significant correlation between TS expression and the number of analyzed lymph node metastases (P = .03). Sixty-six percent of the patients with a high expression of TS in the lymph node metastases had two lymph node metastases analyzed compared with 46% of the patients with a low TS expression in the lymph node metastases. However, this difference was solely a result of the number of tissue samples that were obtainable from the treating hospitals.
Irrespective of the TS expression, more than three lymph node metastases correlated with a worse clinical outcome (OS, P = .04; DFS, P = .05). However, in multivariate analyses, the level of TS expression in lymph node metastases was the only factor that remained a prognostic marker (OS, P = .02; DFS, P = .04). Heterogeneity could be studied only among those patients in whom TS was high (staining score of 2 and 3) and analyzed in two different lymph node metastases (n = 148). Patients with TS heterogeneity in their lymph node metastases had a tendency for a better prognosis than patients with TS homogeneity with respect to DFS (P = .06) but not to OS (P = .1).
Subgroup treated with surgery alone. TS expression in lymph node metastases in the subgroup of patients treated with surgery alone had a prognostic value for OS (P = .04) and DFS (P = .03), where a low level of TS correlated with a better prognosis.
Subpopulation treated with adjuvant chemotherapy. In the subgroup of patients who were randomly assigned to receive adjuvant FU-based chemotherapy, TS expression in lymph node metastases did not give any prognostic information (OS, P = .5; DFS, P = .2).
TS expression and clinical benefit of adjuvant chemotherapy. There was no difference in the clinical outcome between the patients treated with adjuvant chemotherapy compared with the patients treated with surgery alone, irrespective of the TS expression in the lymph nodes metastases.
TS Expression in the Primary Tumor and Prognosis
Entire study group. In the entire study group, the level of TS expression in the primary tumors did not have a significant prognostic value with regard to OS (P = .2; Fig 2A) and DFS (P = .2; Fig 2B). The level of TS expression was correlated with the number of primary tumor samples available for analysis (P = .04). In the group of patients with a high TS expression in the primary tumor, there were two tumor samples analyzed in 91% of the patients compared with 82% of the patients who had low TS expression in the primary tumor. However, this result was simply a function of the tumor material that was submitted by the treating institutions and available for TS analysis. Among patients with high TS expression in their primary tumors, TS heterogeneity did not change the clinical outcome.
Subgroup treated with surgery alone. In the subgroup of 178 patients who were treated with surgery alone, TS expression in the primary tumor was a prognostic marker (OS, P = .03; DFS, P = .03).
Subpopulation treated with adjuvant chemotherapy. Among the 170 patients who were randomly assigned to adjuvant treatment with FU, the level of TS did not provide any prognostic information (OS, P = .8; DFS, P = .5).
Correlation Between TS Expression in the Primary Tumor and Lymph Node Metastases
There was a significant correlation between TS expression in the primary tumor and the expression of TS in the lymph node metastases (P = .001). However, in 28% of the patients, we found different levels of TS expression in the lymph node metastases compared with the primary tumor. Among the 92 patients with a low TS expression in the primary tumor, the TS level in the lymph node metastases was an additional significant prognostic marker for cancer-specific survival (P = .02, Fig 3). Patients who expressed a low level of TS both in the primary tumor and the lymph node metastases had a better cancer-specific survival, but the difference was not statistically significant with respect to OS and DFS.
DISCUSSION
TS expression in the primary tumors of CRC has a prognostic value according to previous studies in which low TS levels in the primary tumor correlated with a better clinical outcome.4-9,35 However, some studies have reported conflicting results that are not further discussed here.23,36
It has also been reported that TS expression in CRC metastases is a predictor of responsiveness to FU chemotherapy in patients with disseminated disease. Distant metastases with a low level of TS expression respond better to FU-based chemotherapy than distant metastases with high TS levels.10-18
Nevertheless, we still lack reliable predictive factors for CRC patients in the adjuvant setting. Because the primary tumor is often heterogeneous, the characteristics found in the primary tumor can be different from the characteristics found in lymph node metastases. It seems reasonable to believe that the grade of aggressiveness in lymph node metastases has a prognostic value.
In a study with 56 rectal cancer patients (Dukes' C), we have previously reported a correlation between the TS level in the primary tumor and the level of TS expression in the lymph node metastases.4 However, other studies have failed to show a correlation in TS expression between the primary tumor and the lymph node metastases.37,38
This present study with 348 patients has revealed that TS expression in lymph node metastases of CRC has a distinct prognostic value; patients whose lymph node metastases express a low level of TS have a longer DFS and OS. The prognostic value was independent of age, sex, site of tumor, and treatment. TS in lymph node metastases was a prognostic marker for patients who were treated with surgery alone and not for patients who received adjuvant chemotherapy. One reason for this could be a change in outcome as a result of the adjuvant chemotherapy, which might have an impact on the survival differences between patients with high compared with low TS levels in the lymph node metastases. If this is true, then adjuvant FU-based chemotherapy tends to reduce any differences in outcome between patients with high versus low TS expression. However, we did not consider risk factors such as impaired performance status, poor differentiation, mucinous pathology, perforation, and proximal and distal margins.
Another possible bias to consider is the number of tumor samples available for TS analyses from each patient because the possibility of detecting areas of higher expression would be expected to increase in proportion to the number of areas examined. In this study, there was a significant correlation between the number of analyzed tissue samples and the expression of TS in the lymph node metastases (P = .03) and the primary tumors (P = .04). The number of obtainable lymph node metastases and primary tumors is a result of the material submitted by the treating institutions. For every case included in the study, each available lymph node metastasis and primary tumor was analyzed for TS expression. The mean number of lymph node metastases was the same in both the patients with high TS expression in the primary tumor and the patients with low TS expression in the primary tumor.
A tendency towards a better prognosis with respect to DFS was observed among patients with a high TS expression and a TS heterogeneity in their lymph node metastases. However, patients with a high TS expression in their primary tumors and TS heterogeneity in the primary tumors did not have a better clinical outcome. Further studies are needed to evaluate whether the intratumoral heterogeneity of TS expression can be a marker for genomic instability and prognosis.
There was also no clinical benefit of adjuvant FU chemotherapy in this subgroup of patients, which is contrary to previously and recently published studies.28,39-45 However, the primary object of this retrospective study was not to evaluate the effect of FU-based adjuvant chemotherapy. The lack of a chemotherapy effect may represent inadequate statistical power in this study, which makes it hard to draw any conclusions. The effect of adjuvant chemotherapy in CRC is also a result of the accumulated dose of chemotherapy administered to each patient. Therefore, it is hard to evaluate the efficacy of the adjuvant chemotherapy because we do not know the number of patients who received a whole chemotherapy schedule without interruption or dose reduction. Furthermore, the patients in our study also received five different treatment schedules of adjuvant chemotherapy, which might have had an impact on the results. The anomalous results with chemotherapy may provide another window through which it may be possible to examine the biology of tumor metastasis and progression, as well as adjuvant chemotherapy.
We recently published a study of 862 CRC patients in whom we analyzed the predictive value of TS in the primary tumor.6 Unexpectedly, we found that the group of patients with a low level of TS expression in the primary tumor had a worse outcome with adjuvant chemotherapy (P = .01), but this observation did not persist in our present study of 348 CRC patients who were all included in this previous report. No other published studies have found any association between low TS expression and a deleterious effect of adjuvant chemotherapy.
There are conflicting results reported on the role of TS expression in primary CRC as a predictive factor for patients treated in the adjuvant setting. The use of different technologies for the assessment of intratumoral TS levels might be a reason for this discrepancy. To date, four studies have shown a predictive value of TS expression in the primary tumor.6,20-22 In these studies, only patients whose cancer had a high TS expression were found to benefit from adjuvant treatment. Another group has reported that CRC patients whose tumors had a low TS gene polymorphism, which is associated with low TS levels, derived greater benefit from adjuvant chemotherapy.46 However, these results are discordant with some other reports using immunohistochemistry to assess TS expression, which did not show any predictive value of TS.9,23 In one of these studies, TS was assessed using two separate scoring systems.23 These different methodologies for evaluating TS yielded different results. Newer techniques using microdissection and polymerase chain reaction may prove more enlightening in the future.
This study indicates that TS assessment in locoregional lymph node metastases of CRC improves the prognostic precision compared with TS expression in the primary tumor. The patients included in our study underwent surgery between 1991 and 1996, and the median number of lymph nodes found in the surgical specimens was only five (range, one to 20 lymph nodes), which might have had an impact on the categorization of the TS cohorts. At present, colorectal surgeons are more aware of the importance of resecting a high number of lymph nodes. Pathologists have also learned the great significance of finding as many lymph nodes as possible in resected tumor specimens.
A thorough tumor staging has a crucial impact on the prognostic and predictive value of the TS expression assessed in lymph node metastases. The use of techniques to detect sentinel nodes might be one way to improve future tumor staging.47,48 Further prospective randomized studies measuring the TS expression in lymph node metastases are needed to evaluate the role of TS as a predictor of benefit from adjuvant chemotherapy and a prognostic marker in patients with locally advanced CRC.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Professor Anders Jacobsen and Professor Bengt Glimelius for valuable assistance during this project and Bo Nilsson for contributing to the statistical analysis of the data. We also thank Associate Professor Carlos Rubio and Associate Professor Johan Lindholm at the Division of Pathology, Karolinska University Hospital, Stockholm, Sweden, for support.
NOTES
Supported by grants from the King Gustaf V:s Jubilee Foundation, the Cancerf?reningen, Stockholm, the Swedish Cancer Society, the Bengt Ihre Foundation of the Swedish Society of Medicine, and the Ruth and Rickard Juhlin Foundation.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer: Adopted on May 17, 1996 by the American Society of Clinical Oncology. J Clin Oncol 14:2843-2877, 1996
Johnston PG, Allegra CJ: Colorectal cancer biology: Clinical implications. Semin Oncol 22:418-432, 1995
McDermott U, Longley DB, Johnston PG, et al: Molecular and biochemical markers in colorectal cancer. Ann Oncol 13:235-245, 2002 (suppl 4)
Edler D, Hallstrom M, Johnston PG, et al: Thymidylate synthase expression: An independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 6:1378-1384, 2000
Edler D, Kressner U, Ragnhammar P, et al: Immunohistochemically detected thymidylate synthase in colorectal cancer: An independent prognostic factor of survival. Clin Cancer Res 6:488-492, 2000
Edler D, Glimelius B, Hallstrom M, et al: Thymidylate synthase expression in colorectal cancer: A prognostic and predictive marker of benefit from adjuvant fluorouracil-based chemotherapy. J Clin Oncol 20:1721-1728, 2002
Lenz HJ, Danenberg KD, Leichman CG, et al: p53 and thymidylate synthase expression in untreated stage II colon cancer: Associations with recurrence, survival, and site. Clin Cancer Res 4:1227-1234, 1998
Van Triest B, Loftus BM, Pinedo HM, et al: Thymidylate synthase expression in patients with colorectal carcinoma using a polyclonal thymidylate synthase antibody in comparison to the TS 106 monoclonal antibody. J Histochem Cytochem 48:755-760, 2000
Allegra CJ, Paik S, Colangelo LH, et al: Prognostic value of thymidylate synthase, Ki-67, and p53 in patients with Dukes' B and C colon cancer: A National Cancer Institute-National Surgical Adjuvant Breast and Bowel Project collaborative study. J Clin Oncol 21:241-250, 2003
Kornmann M, Link KH, Lenz HJ, et al: Thymidylate synthase is a predictor for response and resistance in hepatic artery infusion chemotherapy. Cancer Lett 118:29-35, 1997
Aschele C, Debernardis D, Casazza S, et al: Immunohistochemical quantitation of thymidylate synthase expression in colorectal cancer metastases predicts for clinical outcome to fluorouracil-based chemotherapy. J Clin Oncol 17:1760-1770, 1999
Johnston PG, Lenz HJ, Leichman CG, et al: Thymidylate synthase gene and protein expression correlate and are associated with response to 5-fluorouracil in human colorectal and gastric tumors. Cancer Res 55:1407-1412, 1995
Lenz HJ, Hayashi K, Salonga D, et al: p53 point mutations and thymidylate synthase messenger RNA levels in disseminated colorectal cancer: An analysis of response and survival. Clin Cancer Res 4:1243-1250, 1998
Leichman CG, Lenz HJ, Leichman L, et al: Quantitation of intratumoral thymidylate synthase expression predicts for disseminated colorectal cancer response and resistance to protracted-infusion fluorouracil and weekly leucovorin. J Clin Oncol 15:3223-3229, 1997
Gonen M, Hummer A, Zervoudakis A, et al: Thymidylate synthase expression in hepatic tumors is a predictor of survival and progression in patients with resectable metastatic colorectal cancer. J Clin Oncol 21:406-412, 2003
Cascinu S, Aschele C, Barni S, et al: Thymidylate synthase expression in advanced colon cancer: Correlation with the site of metastasis and the clinical response to leucovorin-modulated bolus 5-fluorouracil. Clin Cancer Res 5:1996-1999, 1999
Davies MM, Johnston PG, Kaur S, et al: Colorectal liver metastasis thymidylate synthase staining correlates with response to hepatic arterial floxuridine. Clin Cancer Res 5:325-328, 1999
Salonga D, Danenberg KD, Johnson M, et al: Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase. Clin Cancer Res 6:1322-1327, 2000
Paradiso A, Simone G, Petroni S, et al: Thymidylate synthase and p53 primary tumour expression as predictive factors for advanced colorectal cancer patients. Br J Cancer 82:560-567, 2000
Johnston PG, Fisher ER, Rockette HE, et al: The role of thymidylate synthase expression in prognosis and outcome of adjuvant chemotherapy in patients with rectal cancer. J Clin Oncol 12:2640-2647, 1994
Takenoue T, Nagawa H, Matsuda K, et al: Relation between thymidylate synthase expression and survival in colon carcinoma, and determination of appropriate application of 5-fluorouracil by immunohistochemical method. Ann Surg Oncol 7:193-198, 2000
Yamachika T, Nakanishi H, Inada K, et al: A new prognostic factor for colorectal carcinoma, thymidylate synthase, and its therapeutic significance. Cancer 82:70-77, 1998
Allegra CJ, Parr AL, Wold LE, et al: Investigation of the prognostic and predictive value of thymidylate synthase, p53, and Ki-67 in patients with locally advanced colon cancer. J Clin Oncol 20:1735-1743, 2002
Edler D, Blomgren H, Allegra CJ, et al: Immunohistochemical determination of thymidylate synthase in colorectal cancer-methodological studies. Eur J Cancer 33:2278-2281, 1997
Calaluce R, Miedema BW, Yesus YW, et al: Micrometastasis in colorectal carcinoma: A review. J Surg Oncol 67:194-202, 1998
Noura S, Yamamoto H, Ohnishi T, et al: Comparative detection of lymph node micrometastases of stage II colorectal cancer by reverse transcriptase polymerase chain reaction and immunohistochemistry. J Clin Oncol 20:4232-4241, 2002
Yasuda K, Adachi Y, Shiraishi N, et al: Pattern of lymph node micrometastasis and prognosis of patients with colorectal cancer. Ann Surg Oncol 8:300-304, 2001
Laurie JA, Moertel CG, Fleming TR, et al: Surgical adjuvant therapy of large-bowel carcinoma: An evaluation of levamisole and the combination of levamisole and fluorouracil—The North Central Cancer Treatment Group and the Mayo Clinic. J Clin Oncol 7:1447-1456, 1989
Poon MA, O'Connell MJ, Moertel CG, et al: Biochemical modulation of fluorouracil: Evidence of significant improvement of survival and quality of life in patients with advanced colorectal carcinoma. J Clin Oncol 7:1407-1418, 1989
Glimelius B: Biochemical modulation of 5-fluorouracil: A randomized comparison of sequential methotrexate, 5-fluorouracil and leucovorin versus sequential 5-fluorouracil and leucovorin in patients with advanced symptomatic colorectal cancer: The Nordic Gastrointestinal Tumor Adjuvant Therapy Group. Ann Oncol 4:235-240, 1993
Johnston PG, Liang CM, Henry S, et al: Production and characterization of monoclonal antibodies that localize human thymidylate synthase in the cytoplasm of human cells and tissue. Cancer Res 51:6668-6676, 1991
Gehan EA: A generalized Wilcoxon test for comparing arbitrarily single-censored samples. Biometrika 52:202-223, 1965
Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972
Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Mini E, Biondi C, Morganti M, et al: Marked variation of thymidylate synthase and folylpolyglutamate synthase gene expression in human colorectal tumours. Oncol Res 11:437-445, 1999
Sanguedolce R, Vultaggio G, Sanguedolce F, et al: The role of thymidylate synthase levels in the prognosis and the treatment of patients with colorectal cancer. Anticancer Res 18:1515-1520, 1998
Yamada H, Ichikawa W, Uetake H, et al: Thymidylate synthase gene expression in primary colorectal cancer and metastatic sites. Clin Colorectal Cancer 1:169-174, 2001
Marsh S, McKay JA, Curran S, et al: Primary colorectal tumour is not an accurate predictor of thymidylate synthase in lymph node metastasis. Oncol Rep 9:231-234, 2002
Wolmark N, Fisher B, Rockette H, et al: Postoperative adjuvant chemotherapy or BCG for colon cancer: Results from NSABP protocol C-01. J Natl Cancer Inst 80:30-36, 1988
Moertel CG, Fleming TR, MacDonald JS, et al: Levamisole and fluorouracil for adjuvant therapy of resected colon carcinoma. N Engl J Med 322:352-358, 1990
Moertel C, Fleming T, MacDonald J, et al: Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: A final report. Ann Intern Med 122:321-326, 1995
NIH Consensus Conference: Adjuvant therapy for patients with colon and rectal cancer. JAMA 264:1444-1450, 1990
International Multicentre Pooled Analysis of Colon Cancer Trials Investigators: Efficacy of adjuvant fluorouracil and folinic acid in colon cancer. Lancet 345:939-944, 1995
International Multicentre Pooled Analysis of B2 Colon Cancer Trials Investigators: Efficacy of adjuvant fluorouracil and folinic acid in B2 colon cancer: International Multicentre Pooled Analysis of B2 Colon Cancer Trials (IMPACT B2) investigators. J Clin Oncol 17:1356-1363, 1999
The Colorectal Cancer Chemotherapy Study Group of Japan: Five-year results of a randomized controlled trial of adjuvant chemotherapy for curatively resected colorectal carcinoma. Jpn J Clin Oncol 25:91-103, 1995
Iacopetta B, Gieu F, Joseph D, et al: A polymorphism in the enhancer region of thymidylate synthase promoter influences the survival of colorectal patients treated with 5-fluorouracil. Br J Cancer 85:827-830, 2001
Muslow J, Winter DC, O'Keane JC, et al: Sentinel lymph node mapping in colorectal cancer. Br J Surg 90:659-667, 2003
Trocha SD, Saha SS, Wiese DW, et al: Differential expression of thymidylate synthase in colorectal tumors and matched lymph nodes: Impact on adjuvant treatment. Am Surg 69:918-922, 2003(Katarina ?hrling, David E)