MINI-SYMPOSIUM: PATHOLOGY OF THE LARGE BOWEL DIAGNOSTIC HISTOPATHOLOGY 14:2 94 © 2007 Elsevier Ltd. All rights reserved. Epidermal growth factor receptor gene and immunohistochemical expression in colorectal carcinomas Torill Sauer Abstract Expression of epidermal growth factor receptor (EGFR) protein can be found in normal colon mucosa, adenomas, and colon carcinomas. Its impact on prognosis in colorectal carcinomas is moderate, but the avail- ability of a monoclonal antibody therapy designed to block EGFR has brought it into prominence in diagnostic histopathology. The literature reports a wide variance in expression of EGFR protein in primary tumours and metastases, ranging from <10% to almost 100%. Discordance in expression between the primary tumour and its metastases vary widely. EGFR gene mutations are rare. EGFR amplification and/or increased copy numbers have been reported and range from 4.5% to 30%. Most tu- mours seem to have a balanced EGFR copy number compared to chro- mosome 7 copy numbers (with a ratio of ∼1). Amplification of high copy numbers seems rare. The relationship between EGFR immunohistochemi- cal expression and gene amplification and/or copy numbers is unre- solved. Lack of standardization of investigation methods and procedures is probably a major cause of the divergent results. Keywords amplification; colorectal carcinoma; EGFR; in situ hybridization Basic concepts Epidermal growth factor receptor (EGFR) is a member of the HER family of tyrosine kinases. Expression of EGFR protein has been found in normal colon mucosa, adenomas and colon carcino- mas. 1 EGFR promotes tumour growth. Its impact on prognosis in colorectal carcinomas (CRC) is moderate. 2 Several anti-EGFR agents are being tested in clinical trials in patients with carcinomas from a number of locations. 3 The association between EGFR concentrations and response to EGFR- targeted therapies is unclear. 4 Not all tumours overexpressing EGFR protein respond to anti-EGFR agents; some EGFR-negative cases show objective response. Expression of EGFR protein is a poor predictor of response to anti-EGFR agents. 5 Recent studies have shown that an EGFR mutation in non- small-cell lung cancer correlates with clinical response to Gef- tinib (Iressa™) therapy. 6–8 Lynch et al. 6 and Paez et al. 7 found gene mutations in 8% and 14% of unselected cases or cases not exposed to Geftinib, respectively. Their results indicated Torill Sauer MD is Professor of Pathology at Ullevaal University Hospital, Oslo, Norway. that mutation/deletion of the EGFR gene might be an important mechanism for response to anti-EGFR agents. Objective response to anti-EGFR agents does not seem to correlate with survival. Pao et al. 8 found EGFR gene mutations in 7/10 Gefnitib-sensitive lung carcinomas, but none in Gefnitib-refractory tumours. They also found analogous mutations in 5/7 Erlotinib (Tarceva™)- sensitive tumours, and in none of 10 Erlotinib-refractory carcino- mas; most of these were adenocarcinomas. The most signifcant prognostic and independent prognostic factors in CRC are TNM stage and potential residual disease after initial surgery. 9 EGFR protein expression is reported to be associ- ated with poor response rates in patients with locally advanced rectal cancer treated by radiation therapy. 10,11 Recent results from studies of EGFR protein expression, EGFR gene mutations as well as gene copy numbers and/or amplifca- tion in CRC are discussed in this review. EGFR immunohistochemical expression in primary colorectal carcinomas (Figure 1) EGFR protein expression in CRC and their metastatic deposits has been investigated in a number of recent studies. 5,12–18 The rate of overexpression varies widely. Most studies report that overexpression in the primary tumour is a common fnding, ranging from ∼40% to nearly 100%. 13,15–19 Ooi et al. found only 8% overexpression in their series. 20 McKay et al. found that the frequency of high-level EGFR expression decreased from well dif- ferentiated, to moderate to poorly differentiated tumours. 15 There may be several reasons for the wide variance in protein expression. Case selection in each study may have an impact on the results, but the principal factor is probably the lack of a uniform and standardized method for immunohistochemical (IHC) staining and interpretation of the staining results. The dif- ferent studies present different IHC methods, antibody clones, antibody concentrations as well as different scoring and evalua- tion systems for interpretation of positivity and level of expres- sion (staining intensity, percentage of stained cells, cut-off point, membranous/cytoplasmic staining). IHC methods differ in sensitivity and antibodies directed against different epitopes Figure 1 EGFR immunohistochemical protein expression in colorectal carcinoma.