1 Département d'Oncologie Médicale, Hôpital Pitié-Salpétrière, Paris; 2 Département d'Anatomopathologie, Hôpital Avicenne, Bobigny; 3 Département de Radiothérapie et Statistique, Hôpital Européen Georges Pompidou, Paris; 4 Département d'Oncopharmacologie, Centre Antoine Lacassagne, Nice; 5 Département de Gastro-Entérologie, Hôpital Avicenne, Bobigny; 6 Département de Chirurgie Viscérale, Hôpital Bobigny, Bobigny; 7 Département d'Oncologie Médicale, 8 Département d'Hématobiologie, 9 Département de Biochimie et de Biologie Moléculaire, Hôpital Avicenne, Bobigny; 10 Département d'Anatomopathologie, Centre Anticancéreux, Clermont-Ferrand, France
* Correspondence to: Dr J.-P. Spano, MD, Hôpital Pitié-Salpetrière, Département d'Oncologie Médicale, 47 Boulevard de l'Hôpital, 75013 Paris, France. Tel: +33-1-42-16-04-52; Fax: +33-1-42-16-04-65; Email: jean-philippe.spano{at}psl.ap-hop-paris.fr
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: A composite EGFR score, obtained by multiplying the grade (% positive cells) by the intensity of labeling (09) was used to define patients with low or high EGFR expression whose clinicopathological features were then compared. Univariate tests and multivariate Cox proportional hazards model were applied for data analysis.
Results: Tissue sections from 150 CRC patients with a median follow-up of 40 months were examined. Median patient age at diagnosis was 70 years (range 3889 years). EGFR reactivity was positive for 143 patients (97%) and high for 118 (80%). According to multivariate analysis, EGFR overexpression was significantly associated with tumor stage, with a higher percentage of EGFR overexpression in T3 than T4 (P=0.003) and not with overall survival.
Conclusions: EGFR was overexpressed in this CRC patient population and was significantly associated with TNM (tumornodemetastasis) stage T3. In the context of a new therapeutic strategy using EGFR-targeted therapies, although EGFR remains a controversial prognostic factor, this expressionstage association may play a crucial role in a decision to initiate an adjuvant treatment.
Key words: colorectal carcinoma, EGFR expression, TNM, therapeutic strategy
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
More recently, the epidermal growth factor receptor (EGFR) has been detected in a wide variety of cancers and, for some of them, its overexpression has been suggested as a factor of poor prognosis, associated with a more aggressive clinical progression, as in lung, breast, ovarian, bladder, oesophageal, cervical and head and neck cancers [410
]. Epidermal growth factor receptor (EGFR) is a 170-kDa transmembrane glycoprotein composed of an intracellular tyrosine-kinase domain, a transmembrane lipophilic segment and an extracellular ligand binding domain, whose main autocrine ligands are epidermal growth factor (EGF) and transforming growth factor-alpha, described 20 years ago [11
, 12
]. EGFR belongs to a family of receptors known as the EGFR family, or ErbB tyrosine-kinase receptors, which comprises four proteins encoded by the c-erb B proto-oncogene: EGFR itself or ErbB1, ErbB2 or HER2/neu, ErbB3 or HER3 and ErbB4 or HER4 [13
, 14
]. It has been shown already that EGFR mRNA in culture human colorectal carcinoma cells may play a significant role during tumor progression [15
] and EGF and EGFR levels have been shown to be higher in malignant zones of colorectal cancer specimens than in the surrounding mucosa [16
]. In a clinical study, EGFR expression has been demonstrated to be associated with poor outcome in CRC patients with stage IV [17
]. EGFR expression has also been reported to be associated with poor response rates in patients with locally advanced rectal cancer treated by radiation therapy [18
]. Nevertheless, the presence of EGFR in CRC may also justify the application of specific EGFR targeted therapy in this cancer localization [19
].
This context prompted us to evaluate EGFR expression in a standard population of CRCs and to analyze its relationship with the main clinicohistological prognostic factors and their respective impacts on patient prognosis and survival that could define new therapeutic approaches for patients with CRC.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Histology
All resected CRCs were received fresh, fixed in 10% pH-neutral formalin, embedded in paraffin. All patients had adenocarcinomas and were staged according to the International Union Against Cancer TNM staging (UICC/TNM) system [20]. Characteristics studied included age, sex, tumor site, tumor size, degree of histological differentiation (well/moderate/poor), number of invaded lymph nodes counted during the slide review and classified as N1 or N2, based on UICC/TNM staging system, perineural invasion and/or venous emboli (classified as present or absent), the presence of synchronous metastases and the occurrence of metastases during the follow-up period.
All histological slides were reviewed by two pathologists from our institution to confirm the diagnoses, and evaluate the patterns and intensity of EGFR reactivity.
Immunohistochemistry
Immunohistochemistry was performed using the avidinbiotin complex technique (Dako LSAB2 System, Peroxidase K675, Dako, Carpinteria, CA). Briefly, 5 µm-thick sections were cut from formalin-fixed, paraffin-embedded tissue, deparaffinized and rehydrated. After blocking endogenous peroxidase activity by immersion in 0.3% hydrogen peroxide in methanol for 30 min, antigen retrieval was performed by immersing the sections in 0.01 M sodium citrate buffer (citric acid and sodium citrate, pH 6.0) for 40 min at 99°C in a water bath. Primary mouse monoclonal antibody (EGFR/113, Novocastra Laboratories Ltd, Newcastle, UK), diluted 1:20, was deposited on the tissue sections for 60 min, at room temperature, followed by biotin-labeled affinity isolated goat anti-mouse immunoglobulins and streptavidin-coupled horseradish peroxidase. Complexes were visualized (brown) with 3,3'-diaminobenzidine (Dako) and the slides were counterstained with Mayer's hematoxylin, dehydrated and mounted with Peramount.
Each run included, for each patient, non-immune mouse IgG used as the primary antibody for the negative controls and normal epidermis known to express EGFR served as the positive control.
Evaluation of EGFR label
All the slides were scored by two pathologists blinded for all patient characteristics but TN stage. No nuclear or nucleolar reactivity was observed. Slides were assessed using a light microscope. In the case of differing opinions, the definitive assessment was obtained by consensus. The percentage of labeled cells was graded as follows: grade 0, no positive cells; grade 1, 125% labeled tumor cells; grade 2, 2550% labeled tumor cells; grade 3, >50% positive tumor cells. The intensity of peroxidase deposits, ranging from light beige to dark brown, was assessed visually as indicating the tumor cell membrane, cytoplasm or both and was scored as 0 (negative), 1 (weak), 2 (moderate) or 3 (strong). A composite score, potentially ranging from 0 to 9, was obtained by multiplying the grade by the intensity [17]. Patients were analyzed as a function of their EGFR expression: low, <6 and high,
6.
Statistical analyses
All statistical analyses were performed using Statistica, version 5.1. The chi-square test was used to compare the major clinicohistological prognostic factors for CRC as a function of the composite EGFR expression score (<6 or 6) obtained for these adenocarcinomas. Various typical prognostic factors were considered for univariate analysis (Table 1). Multivariate analysis, conducted with a backward stepwise application of Cox regression [21
], was used to evaluate the influence of the selected prognostic factors on survival. Survival was calculated from the date of surgery. The effect of EGFR expression on survival was assessed using the KaplanMeier [22
] method and compared using the log-rank test [23
]. The significance level was defined as P
0.05.
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
EGFR expression in CRCs
EGFR expression was detected in 143 (97%) of 148 assessable tumors. According to the labeling-intensity scores, 82 (55%), 47 (32%) and 14 (10%) adenocarcinomas were accorded scores of 3, 2 and 1, respectively. According to the EGFR-positivity grade, five (3%) adenocarcinomas had no reactivity, two (1%) had <10% positive cells, 20 (14%) had 1050% reactive cells and 121 (82%) had >50% labeled cells. Examples of the different patterns of EGFR labeling are shown in Figure 1(A. The composite scores, the product of positivity grade multiplied by labeling-intensity score for each patient, are shown in Table 2. EGFR expression was considered high (6) for 118 (80%) patients and low (<6) for 30 patients.
|
|
|
Univariate prognostic analyses
With a median follow-up of 40 months (range 1284 months), among 132 assessable patients with a follow-up >1 year, 84 patients were still alive and 48 patients were known to have died. The median overall survival for all patients was 61.7 months. Causes of death included disease progression (n=31), sepsis secondary to chemotherapy (n=7), secondary malignancy (n=1), cardiovascular disease (n=3), postoperative complications (n=2) and unknown cause (n=4). Nineteen patients suffered disease recurrence (five local and 14 distant metastases) and 10 developed a secondary malignancy during follow-up. The most common site of metastasis was the liver (n=9; 69%), followed by the lung (n=3; 23%).
The different variables analyzed by univariate analysis are shown in Table 4. Vascular emboli, perineural invasion, tumor stage, lymph-node involvement, tumor site, tumor differentiation status, synchronous metastases and CEA level were significantly associated with overall survival and thus appear to be prognostic for CRC outcome. Regardless of its composite score, EGFR expression was not an independent prognostic factor for specific and overall survival.
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
EGFR has been found to be elevated in CRCs, with expression rates ranging from 25 to 77% [16, 17
, 24
]. This heterogeneity of expression is attributed to the different detection techniques, although most are based on quantitative immunohistochemical labeling with monoclonal antibodies [17
]. Despite these known reproducibility and validation difficulties [14
, 25
, 26
], immunohistochemical testing remains one of the most common methods used to assess EGFR expression and has also largely been validated for the screening of HER2 receptor [27
].
To evaluate overexpression of EGFR in our study, we used the same EGFR composite score as that described by Goldstein and Armin [17], which seems to be one of the most promising and accurate scoring systems currently defined. Our observations confirm EGFR expression in colorectal adenocarcinomas [14
, 16
, 17
, 28
31
], consistent with high EGFR expression in 80% of the tumors examined.
Consistent with the results of the main studies [17, 29
], we found no differences in EGFR expression among the different tumor sites. Only one study has demonstrated higher EGFR expression in cancers of the distal colon than the rectum [32
].
Possible associations between expression (and overexpression) of EGFR and other clinicohistological parameters in CRC patients remain unclear. EGFR expression was reported to be correlated with more aggressive disease [33], increased risk of metastases [24
], advanced tumor stage [34
], significantly more reactivity in more deeply invaded regions compared with the superficial tumors [17
] and higher rates of mesenteric lymph-node involvement [31
].
According to the major studies published to date [14, 17
, 25
, 35
], we found no significant association between histological grade and EGFR expression. Few studies reported a relationship between histological grade and EGFR overexpression [35
37
]. However, in these studies, moderately and poorly differentiated tumors predominated, whereas our analysis was based on mostly (67%) well-differentiated adenocarcinomas.
Our most relevant findings demonstrate a significant association between high EGFR expression and TNM tumor stage at diagnosis, highlighting a relationship between EGFR overexpression and tumor invasion. This is the largest study to show that EGFR expression is significantly overexpressed in TNM tumor stage T3. This finding is in agreement with an earlier study examining fewer adenocarcinomas [31] and confirms the results of Goldstein and Armin [17
], who demonstrated that EGFR immunoreactivity was significantly higher in the deepest regions of the tumors, as compared to the superficial or luminal zones.
Previous studies did not demonstrate any influence of EGFR expression on patient survival and disease-free survival [28, 29
, 35
, 38
, 39
]. Nevertheless, patients with EGFR overexpression seem to have a higher risk of generating liver metastases [24
]. We did not find EGFR overexpression to be associated with a poorer prognosis or shorter overall survival.
Although EGFR in the overall population is not considered as a prognostic factor in CRC, it plays an important role in tumor cell proliferation [15], and consequently its overexpression in stage II disease (a situation in which chemotherapy is not a standard) represents a window of opportunity to test the efficacy of EGFR inhibitors in this specific patient population [7
, 40
]. Some preclinical and clinical studies [41
, 42
] have already demonstrated the efficacy of EGFR inhibitors in advanced colorectal carcinomas and their potential synergistic effects with chemotherapy and radiation therapy [43
, 44
]. This new finding should logically encourage researchers to continue their investigations in an adjuvant setting. In this context, EGFR expression has still not proven its predictive value; more effort needs to be targeted at improving simple EGFR detection techniques, such as examining the downstream signaling pathway and EGFR somatic mutations [45
].
![]() |
Notes |
---|
Received for publication February 27, 2004. Revision received July 23, 2004. Accepted for publication July 26, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Shepherd NA, Saraga EP, Love SB et al. Prognostic factors in colonic cancer. Histopathology 1989; 14: 613620.[ISI][Medline]
3. Skibber JM, Minsky BD, Hoff PM. Cancer of the colon. Cancers of the gastrointestinal tract. In DeVita VT, Hellman S, Rosenberg SA (eds): Cancer: Principles and Practice of Oncology, 6th edition. Philadelphia, PA: Lippincott Williams and Wilkins 2001; 12161271.
4. Hendler F, Shum Siu A, Nanu L et al. Overexpression of EGF receptors in squamous tumors is associated with poor survival. J Cell Biochem 1988; (Suppl 12A) 105 (Abstr).
5. Sainsbury JRC, Farndon JR, Needham GK et al. Epidermal growth factor receptor as predictor of early recurrence of and death from breast cancer. Lancet 1987; 1: 13981402.[ISI][Medline]
6. Bauchnecht T, Kohler M, Janz I et al. The occurrence of epidermal growth factor receptors and the characterization of EGF-like receptors and the characterization of EGF-like factors in human ovarian, endometrial, cervical and breast cancer. J Cancer Res Clin Oncol 1989; 115: 193199.[CrossRef][ISI][Medline]
7. Neal DE, Marsh C, Bennett MK et al. Epidermal growth factor receptors in human bladder cancer: comparisons of invasive and superficial tumors. Lancet 1985; 1: 366368.[ISI][Medline]
8. Ozawa S, Ueda M, Ando N et al. Prognostic significance of epidermal growth factor receptor in esophageal squamous cell carcinomas. Cancer 1989; 63: 21692173.[ISI][Medline]
9. Pfeiffer D, Stellwag B, Pfeiffer A et al. Clinical implications of the epidermal growth factor receptor in the squamous cell carcinoma of the uterine cervix. Gynecol Oncol 1989; 33: 146150.[CrossRef][ISI][Medline]
10. Salomon D, Brandt R, Ciardiello F et al. Epidermal growth factorrelated peptides and their receptors in human malignancies. Crit Rev Oncol Hematol 1995; 19: 183232.[CrossRef][ISI][Medline]
11. Coffey RJ Jr, Goustin AS, Soderquist AM et al. Transforming growth factor alpha and beta expression in human colon cancer lines: implications for an autocrine model. Cancer Res 1987; 47: 45904594.[Abstract]
12. Carpenter G, Cohen S. Epidermal growth factor. J Biol Chem 1990; 265: 77097712.
13. Yarden Y, Sliwkowski M. Untangling the ErbB signaling network. Nat Rev Mol Cell Biol 2001; 2: 127137.[CrossRef][ISI][Medline]
14. Baselga J. Targeting the epidermal growth factor receptor with tyrosine kinase inhibitors: small molecules, big hopes. J Clin Oncol 2002; 20: 22172219.
15. Tong WM, Ellinger A, Sheinin Y et al. Epidermal growth factor receptor expression in primary cultured human colorectal carcinomas cells. Br J Cancer 1998; 77: 17921798.[ISI][Medline]
16. Messa C, Russo F, Caruso MG et al. EGF, TGF and EGF-R in human colorectal adenocarcinoma. Acta Oncologica 1998; 37: 285289.[CrossRef][ISI][Medline]
17. Goldstein NS, Armin M. Epidermal growth factor receptor immunohistochemical reactivity in patients with American Joint Committee on cancer stage IV colon adenocarcinoma. Cancer 2001; 92: 13311346.[CrossRef][ISI][Medline]
18. Giralt J, Eraso A, Armengol M et al. Epidermal growth factor receptor is a predictor of tumor response in locally advanced rectal cancer patients treated with preoperative radiotherapy. Int J Radiat Oncol Biol Phys 2002; 54: 14601465.[CrossRef][ISI][Medline]
19. Kawamoto T, Sato JD, Le A et al. Growth stimulation of A431 cells by EGF: identification of high affinity receptors for epidermal growth factor by an anti-receptor monoclonal antibody. Proc Natl Acad Sci USA 1983; 80: 13371341.[Abstract]
20. International Union Against Cancer (UICC). In Hermaek P, Hutter RVP, Sobin LH et al. (eds): TNM Classification of Malignant Tumors, 4th edition. Berlin: Springer-Verlag 1998.
21. Cox DR. Regression models and life table. J R Stat Soc 1972; 34: 187220.[ISI]
22. Kaplan EL, Meier P. Non parametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457481.[ISI]
23. Mantel N, Haenszel W. Chi-square with one degree of freedom: Extensions of the MantelHaenszel procedure. J Am Stat Assoc 1963; 58: 690700.[ISI]
24. Radinsky R, Risin S, Fan Z et al. Level and function of epidermal growth factor receptor predict the metastatic potential of human colon carcinoma cells. Clin Cancer Res 1995; 1: 1931.[Abstract]
25. Waksal HW. Role of an anti-epidermal growth factor receptor in treating cancer. Cancer Metastasis Rev 1999; 18: 427436.[CrossRef][ISI][Medline]
26. Formento JL, Francoual M, Formento P et al. Epidermal growth factor receptor assay: validation of a single point method and application to breast cancer. Breast Cancer Res Treat 1991; 17: 211216.[ISI][Medline]
27. Allred DC, Swanson PE. Testing for erbB-2 by immunohistochemistry in breast cancer. Am J Clin Pathol 2000; 113: 171175.[ISI][Medline]
28. Saeki T, Salomon DS, Johnson GR et al. Association of epidermal growth factor-related peptides and type I receptor tyrosine kinase with prognosis of human colorectal carcinomas. Jpn J Clin Oncol 1995; 25: 240249.[Abstract]
29. Koenders PG, Peters WH, Wobbes T et al. Epidermal growth factor receptor levels are lower in carcinomatous than in normal colorectal tissue. Br J Cancer 1992; 65: 189192.[ISI][Medline]
30. Komuta K, Koji T, Izumi S et al. Expression of epidermal growth factor receptor messenger RNA in human colorectal carcinomas assessed by non-radioactive in-situ hybridization. Eur J Surg Oncol 1995; 21: 269275.[CrossRef][Medline]
31. Karameris A, Kanavaros P, Aninos D et al. Expression of epidermal growth factor (EGF) and epidermal growth factor receptor (EGFR) in gastric and colorectal carcinomas. An immunohistological study of 63 cases. Pathol Res Pract 1993; 189: 133137.[ISI][Medline]
32. Koretz K, Schlag P, Moller P. Expression of epidermal growth factor receptor in normal colorectal mucosa, adenoma, and carcinoma. Virchow Arch A Pathol Anat Histopathol 1990; 416: 343349.[ISI]
33. Prewett M, Hooper A, Bassi R et al. Growth of human colorectal carcinoma xenografts by anti-EGF receptor monoclonal antibody in combination with 5-fluorouracil or irinotecan. Proc Am Assoc Cancer Res 2001; 42: 287 (Abstr 1543).
34. Radinsky R. Modulation of tumor cell gene expression and phenotype by the organ-specific metastatic environment. Cancer Metastasis Rev 1995; 14: 323338.[ISI][Medline]
35. McKay JA, Murray LJ, Curran S et al. Evaluation of the epidermal growth factor receptor (EGFR) in colorectal tumors and lymph node metastases. Eur J Cancer 2002; 38: 22582264.[CrossRef][ISI][Medline]
36. Steele RJ, Kelly P, Ellul B et al. Epidermal growth factor receptor expression in colorectal cancer. Br J Surg 1990; 77: 13521354.[ISI][Medline]
37. Steele RJ, Kelly P, Ellul B et al. Immunohistochemical detection of epidermal growth factor receptors on human colonic carcinomas. Br J Cancer 1990; 61: 325326.[ISI][Medline]
38. Moorghen M, Ince P, Finney KJ et al. Epidermal growth factor receptors in colorectal carcinoma. Anticancer Res 1990; 10: 605611.[ISI][Medline]
39. Yasui W, Sumiyoshi H, Hata J et al. Expression of epidermal growth factor receptor in human gastric and colonic carcinomas. Cancer Res 1988; 48: 137141.[Abstract]
40. Mendelsohn J. Epidermal growth factor receptor as a target for therapy with antireceptor monoclonal antibodies. J Natl Cancer Inst Monogr 1992; 13: 125131.[Medline]
41. O'Dwyer PJ, Benson AB. Epidermal growth factor receptor-targeted therapy in colorectal cancer. Sem Oncol 2002; 29 (Suppl 14): 1017.
42. Mendelsohn J, Baselga J. Status of epidermal growth factor receptor antagonists in the biology and treatment of cancer. J Clin Oncol 2003; 21: 27872799.
43. Saltz LB, Meropol NJ, Loehrer PJ Sr et al. Phase II trial of cetuximab in patients with refractory colorectal cancer that expresses the epidermal growth factor receptor. J Clin Oncol 2004; 22: 12011208.
44. Cunningham D, Humblet Y, Siena S et al. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 2004; 351: 337345.
45. Lynch TJ, Bell DW, Sordella R et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004; 350: 21292139.