Practical and molecular evaluation of colorectal cancer: new roles for the pathologist in the era of targeted therapy.

Colorectal cancer (CRC) is an important global health problem. In 2002, it was the third most commonly occurring cancer with about 1 million new cases (9.4% of the world total). It was the fourth most frequently diagnosed cancer in men (after lung, prostate, and stomach cancer) and the third most common in women (after breast and cervical cancer). Age-standardized incidence rates varied 25-fold, with the highest frequencies in developed countries. Mortality in 2002 was about half the incidence rate. Five-year survival estimates in men were 65% in North America and 54% in Western Europe, but less elsewhere. This overall relatively good prognosis means that the prevalence of the disease is high: 2.8 million patients are still alive within 5 years of diagnosis, making it the second most prevalent cancer after breast cancer. (1)

Recent medical and surgical advances have broadened the array of therapeutic options available for CRC, with a definite impact on survival. Management plans must be constructed for individual patients on the basis of all the available evidence and the expertise of the various specialists confronted with the case. This implicates a need for multidisciplinary team meetings with participation of gastroenterologists, surgeons, radiologists, pathologists, oncologists, and radiotherapists. The pathologists' role in such a team should not be underestimated, as tissue diagnosis and prognostication determine further treatment. Patients and clinicians deserve a consistent and informed approach to the reporting of CRC specimens. (2)

TRADITIONAL ROLE OF PATHOLOGISTS IN THE MULTIDISCIPLINARY TEAM

Pathologists may encounter several types of specimens as the patient progresses along the CRC care pathway. These are the diagnostic biopsies, the local excision specimens (polypectomies, endoscopic mucosal resections, and local rectal resections), and the surgical specimens (various colectomy specimens, anterior resections, and abdominoperineal resections). Each type of specimen has its own limitations with associated diagnostic problems and a particular way of reporting.

Diagnostic Biopsies

These can be obtained at colonoscopy, by punction (eg, lesions suggestive of liver metastases), or at laparoscopy. Tissue fragments obtained during endoscopy are taken from a polypoid, ulcerated, or flat lesion suggestive of malignancy. The aim is to confirm the diagnosis of malignancy, to classify its type (including primary versus secondary), and to assess tumor grade. In the colon and rectum, invasive adenocarcinoma can be diagnosed only when tumor cells are unequivocally present in the submucosa. This may be difficult, as biopsies from endoscopically malignant lesions may be superficial or consist of necrotic tissue. Also, not infrequently, only adenomatous tissue is present. A desmoplastic stromal reaction may then assist in the diagnosis of malignancy. Polypoid lesions should be properly sectioned before embedding to enable evaluation of the submucosa and section margin. In case of doubt, a repeated biopsy may be indicated. Biopsies taken from other sites than the colon should always be accompanied by sufficient clinical information. The histologic aspect may be suggestive of a metastasis of CRC. In most cases, however, assessment of classic morphology must be assisted by the evaluation of selected immunohistochemical stains.

Local Excision Specimens

In some circumstances, a biopsy diagnosis of adenocarcinoma may be followed by planned local excision. Lesions can be removed by endoscopic polypectomy, endoscopic mucosal resection, or, in the case of low rectal tumors, per anal excision after Parks procedure or transanal endoscopic microsurgery. Polypectomy specimens require the use of consistent terminology by pathologists and their clinical colleagues to avoid overtreatment or undertreatment. (3) Intraepithelial neoplasia (previously called dysplasia) in adenomas was classically graded according to a 3-tiered system (mild, moderate, and severe) based upon tissue architecture, nuclear changes, and cytoplasmic differentiation. The mild and moderate categories are currently contracted into low-grade intraepithelial neoplasia as opposed to the high-grade variant. The term intramucosal adenocarcinoma can be used when the architectural and cytologic features are essentially those of malignancy (eg, a cribriform growth pattern), but no invasion into the submucosa can be demonstrated. As this may be dependent on proper orientation of the specimen, carrying out multiple step sections and using selected immunohistochemical stains to evaluate the relation of the neoplastic glands with the muscularis mucosae may help to arrive at a correct diagnosis. When definite invasive adenocarcinoma has developed in an adenomatous polyp, the pathology report should at least include a measure of the proximity of the tumor to the endoscopic resection margin, an assessment of its degree of differentiation and the presence of tumor budding, and a statement concerning the presence or absence of lymphatic or blood vessel invasion. A tumor-free margin less than 1 mm, a high-grade cancer, and presence of lymphovascular invasion all increase the risk for nodal metastases and will likely lead to a full resection depending on the clinical status of the patient. (4) Local excisions of early rectal cancer may be an alternative to anterior and abdominoperineal resections for patients who are unfit for major abdominal surgery, but they are also increasingly used as a procedure with curative intention for those with early-stage rectal cancer. Each type of excision removes a full-thickness piece of rectal wall without regional lymph node sampling. The pathologist must ensure proper reporting of the depth of invasion, as this will determine subsequent case management. (5) The fresh specimen should be pinned on a cork board to prevent distortion during formalin fixation. After macroscopic inspection, it should be thoroughly sampled and investigated microscopically.

Surgical Specimens

Colorectal resection specimens require careful handling and assessment to determine the key prognostic data reliably and accurately. The role of the pathologist varies with respect to each of the other members of the multidisciplinary team. We can help the surgeon in determining prognosis by gross dissection and microscopy, but we must also identify the plane of surgery that has been achieved and give feedback on the completeness of the excision. We may provide valuable information to radiologists by assessing the accuracy of the preoperative radiologic staging and by controlling their prediction of completeness of excision, particularly in rectal cancer. With regard to oncologists and radiotherapists, our function is to describe the effectiveness of neoadjuvant therapy, to identify those patients who need adjuvant therapy and, if possible, to suggest the most appropriate therapy by using particular molecular tests.

We must provide complete, consistent, and intelligible reports. This goal can be reached by developing and updating reporting standards based on continuing medical education, regular audit, and, importantly, feedback from our clinical colleagues. It has been demonstrated that the adoption of a minimal data set, such as the one developed by the United Kingdom Royal College of Pathologists, (6) improves the completeness of pathology reports. (7) Of course, such forms can be subject to international consensus guidelines. Given the lack of evidence for some critical changes in TNM6 relative to TNM5 (6th edition and 5th edition, respectively, of the TNM staging system of The American Joint Committee on Cancer/International Union Against Cancer), the British position has been to remain, for the time being, with TNM5 as far as standardized reporting is concerned. (2)

A minimal data set must contain some key features focused on treatment and prognosis. It has been amply demonstrated that patients with node-positive CRC have an improved survival when treated with adjuvant chemotherapy. (8-11) Since the chance of detecting lymph node metastases depends partly on the number of retrieved nodes, pathologists must make an effort to find as many lymph nodes as possible. A failure of the pathologist to detect all lymph node metastases may come at a high cost for the patient. Some oncologists also offer adjuvant therapy to so-called high-risk stage II node-negative patients, because of the presence of pathologic features predictive of an increased risk of relapse (extensive extramural spread, extramural venous invasion, incomplete resection, peritoneal invasion, and tumor perforation). (12) Pathologists assist in the selection of such patients by mentioning the presence or absence of these features in their report.

Prognosis is furthermore very much determined by the completeness of the excision. While pathologists have always investigated for the presence of tumor at the longitudinal margins, as this is a recognized though rare (1%-2%) poor prognostic feature, only recently have other surgical margins acquired attention. For example, tumor frequently involves the circumferential resection margin (CRM) in the rectum (5%-36%) and is then significantly associated with higher rates of local recurrence, metastasis, and a poorer overall survival. (13,14) In contrast, in the rest of the colon, the surgeon usually creates a narrow mesocolic margin. Only in right hemicolectomies is there an area of variable size on the posterior surface of the cecum and ascending colon, which is involved by tumor in about 7% to 10% of cases. (15) The long-term consequences of a "positive margin" in this location are however still unclear.

Neoadjuvant treatment may be applied to rectal cancer. When the pathologist receives the subsequent resection specimen, he or she should comment not only on the CRM as the most important predictor of outcome, but also on the grade of tumor regression (particularly in cases in which the CRM is not involved). In this regard, it is useful to adhere to a single classification scheme for regression. (16) The tumor site should be investigated according to a dissection guideline that combines extensiveness with practicality. (17)

TOTAL MESORECTAL EXCISION AND ITS IMPLICATIONS

Total mesorectal excision (TME) is a fine illustration of a surgical principle that should be applied to all CRCs, namely, precise anatomic dissection of affected bowel with its mesentery, enclosed within an intact fascial plane. Total mesorectal excision may be a technically demanding and time-consuming procedure, but it has led to a demonstrable improvement of local cancer control with a decreasing number of complications. Its systematic use by well-trained surgeons in some locations has led to reductions of cancer recurrence rates from 20% to about 10%. (18)

The mesorectum is a fatty tissue layer of about 2 to 3 cm in thickness, with in lying blood vessels, nerves, lymphatics, and lymph nodes. It surrounds the rectum and is enveloped by a fascia. Mesorectal excision is the surgical removal of this soft tissue envelope under direct vision, ideally by dissecting between the visceral and parietal pelvic fascia. A mesorectal excision can be total (TME) or partial. Total mesorectal excision refers to excision down to the pelvic floor and is indicated for carcinoma of the middle and lower third of the rectum. In partial mesorectal excision, the mesorectum is transsected perpendicular to the rectal wall at a distance of 5 cm beyond the gross distal border of the tumor. This is sufficient for surgical treatment of carcinomas of the upper third of the rectum. In general, a TME specimen with a smooth surface without incisions or tearing, and without coning, is an indicator of successful surgery. Coning is the result of a tendency of the surgeon to cut in the direction of the rectal wall rather than staying outside the visceral fascia, especially during distal dissection in a narrow pelvis. (19)

Mesorectal excision specimens offer an opportunity to study the CRM extensively. There is a nonperitonealized bare area both anteriorly and posteriorly to the rectum. While the anterior CRM is located only in the most distal part of the specimen, the posterior CRM has a triangular shape and extends to the mesosigmoid. A positive CRM is defined as continuous or discontinuous tumor extension or the presence of a positive lymph node within 1 mm of the radial, nonperitonealized soft tissue edge. (14) It is the foremost predictor for local tumor recurrence. (20) When the CRM is negative, the next most important factor for assessing recurrence risk is the quality of the mesorectal dissection. (21) It is therefore a good routine practice to include an assessment of this feature in all reports on surgical specimens for rectal cancer.

The assessment of the quality of mesorectal excision begins with an evaluation of the fresh specimen (complete, nearly complete, and incomplete mesorectum; Figure 1, A through C). Photographic documentation is recommended, especially when the mesorectum is incomplete. The specimen is then painted with ink, partially opened (leaving the bowel intact at a level just proximal and just distal to the tumor), and fixed for at least 48 hours. Afterwards, the unopened portion of the fixed specimen is sliced into 3- to 5-mm thick transverse sections. These are all laid out and inspected in detail to determine the relationship of the tumor to the CRM (Figure 2). A final grade can then be assigned to the completeness of the mesorectum. (22,23) An appropriate number of blocks must be taken to ensure a proper microscopic evaluation of the tumor and its locoregional lymph node metastases and their relation to the CRM. (22) Key factors to report are the location of the tumor within the mesorectum (anterior or posterior, proximal or distal), the effect of neoadjuvant radiotherapy on the lesion, the status of the distal margin and locoregional lymph nodes, and the distance of the tumor to the radial resection margin. (23)

A proper evaluation of the mesorectum in TME specimens by the pathologist is thus important for 3 main reasons: first, it provides feedback on technical issues to the surgeon; second, assessment of the CRM is the most significant predictor of local recurrence; third, the quality of the excised mesorectum is a key factor affecting the risk of local recurrence in patients with a negative CRM. Thus far, North American policy regarding appropriate handling of TME specimens has lagged behind that of European policy. American guidelines, such as the year 2000's Protocol for the Examination of Specimens from Patients with Carcinoma of the Colon and Rectum, do not yet include assessment of TME quality. (24) Many centers in the United States and Canada continue to examine rectal cancer specimens that have been opened in the region of the tumor, fixed for 24 hours or less, and make little or no attempt to assess the completeness of the mesorectum.

TARGETED THERAPIES AND TISSUE BIOMARKERS

Although the mainstay of CRC care is surgery to remove the primary tumor, there are many potential therapies depending on the tumor itself, the condition of the patient, and previous application of particular therapeutic modalities. Systemic chemotherapy is currently used as an adjunct to surgery to improve survival in stage III and certain "high-risk" patients with stage II CRC. Appropriate chemotherapy in the setting of distant metastasis will also prolong survival and improve quality of life. During the last 20 years, there have been significant advances in the treatment of patients with CRC by chemotherapy. Early improvements occurred because of an increased understanding of the pharmacology of 5- fluorouracil and the discovery of modulators of its activity, for example, leucovorin. In more recent years, new and potent cytotoxic drugs such as oxaliplatin and irinotecan have become available.

The classic cytotoxic chemotherapeutic agents came into use after a drug development process that was based on the selection of molecules displaying a particular desirable functional effect. The corollary is that there are often notable side effects, which may at times render the patient intolerant to the treatment. Therefore, investigators have tried early on to develop drugs on the basis of a different principle, namely that of (1) selecting a particular molecular target and (2) designing and synthesizing a small molecule or producing a monoclonal antibody to affect specifically the chosen target. It was initially hoped that such "targeted" therapies would display no systemic toxicity. However, because many targeted therapeutics entered clinical practice without an in-depth investigation of the anatomic distribution of the affected molecular pathways and their functional characteristics, they often have a much broader specificity than initially intended and may, at times, even lead to quite unexpected side effects.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

A second problem is that targeted therapies are not effective in all patients. Ideally, patients should be selected for a particular therapy by assessment of their tumors for tissue biomarkers predictive of a successful treatment. The problem is that such biomarkers are often only identified when large patient groups are already being treated with the drug. Biomarker identification currently faces major logistic and biologic challenges. New technologies in the areas of genomics, methylomics, transcriptomics, proteomics, and metabolomics may overcome these limitations. (25)

TARGETED THERAPIES AND COLORECTAL CANCER

Bevacizumab and CRC

Bevacizumab is a humanized murine immunoglobulin (Ig) G1 monoclonal antibody against human vascular endothelial growth factor (VEGF). It is directly antiangiogenic but may also alter tumor vasculature, thus decreasing tissue pressure in tumors and possibly enhancing the delivery of classic chemotherapeutic agents. It has been approved by the US Food and Drug Administration in combination with chemotherapy in first- and second-line treatment for metastatic CRC. Apart from anti-VEGF treatment, targeted therapies in CRC aim to inhibit the function of the epidermal growth factor receptor (EGFR).

EGFR in Normal Tissues and Cancers

The epidermal growth factor receptor is a 170-kDa transmembrane tyrosine kinase receptor that is present in most epithelia and that plays an important role in cell growth and function. Extensive research during the last few years has broadened our knowledge about this receptor's structural and functional characteristics. Epidermal growth factor receptor (also called HER1 or ErbB1) belongs to the HER or ErbB family, which also comprises HER2/neu, HER3, and HER4. All 4 receptors are topologically similar, comprising 3 domains: (1) an extracellular domain that binds specific ligands, such as epidermal growth factor (EGF), transforming growth factor a, and amphiregulin, which bind only to EGFR; (2) a hydrophobic transmembrane domain that is involved in interactions between cell surface receptors; and (3) an intracellular domain that shows tyrosine kinase activity. There are at least 2 exceptions to these general principles: HER2/neu binds no known growth factor ligand with high affinity, and HER3 shows virtually no tyrosine kinase activity. However, all receptors and their ligands interact to form an integrated system in which an initial signal can be amplified and diversified into multiple cellular responses. To activate the EGFR signaling system, 3 sequential steps are usually required. First, specific ligands bind to the extracellular domain of EGFR, resulting in a conformational change. Second, this structural change allows the receptor to form a dimer with another ligand-bound EGFR (homodimer) or with 1 of the other EGFR-related HER receptors (heterodimer). Finally, receptor dimerization causes autophosphorylation of tyrosine residues within the intracellular domain of the receptors, leading to the activation of particular downstream signal transduction pathways. (26) Several such signaling cascades have been identified with regard to EGFR-mediated biologic effects. The 2 most important are (1) the mitogen-activated protein kinase (MAPK) pathway, with participation of KRAS, BRAF, and the extracellular signal-regulated kinase (ERK) proteins; (2) the phosphoinositide 3-kinase (PI3K)/phosphatase protein homolog to tensin (PTEN)/AKT pathway (Figure 3). In normal tissues, ligand binding by EGFR results in the activation of particular transcription programs. The ultimate consequences of these processes are cell proliferation on the one hand and survival and motility on the other hand, through the MAPK and PI3K/PTEN/AKT pathways, respectively. The pathways also include an autocrine loop with production and secretion of HER receptor ligands, which may then act in an autocrine, juxtacrine, or paracrine manner.

[FIGURE 3 OMITTED]

Tumor cells can use 2 basic strategies to rely less on growth signals and ultimately become self-sufficient. First, they may modulate the level of expression and the functional characteristics of their growth signal receptors. Second, they may activate the downstream signaling pathways without actually involving or modulating the cell surface receptors. Both strategies have been observed in colorectal carcinoma. This ultimately leads to the classic features of cancer: uncontrolled cell proliferation and tumor growth, suppression of apoptosis, stimulation of angiogenesis, and facilitated metastatic spread. (27,28)

EGFR Expression as a Prognostic Marker in CRC

Since the EGFR axis is apparently involved in CRC tumorigenesis and progression, EGFR expression has been investigated as a possible prognostic marker in patients with CRC. To this end, several techniques at the DNA, RNA, and protein level can be applied. Commonly used methods include immunohistochemistry (IHC), Western blot, fluorescence in situ hybridization (FISH), and enzyme-linked immunoadsorption assays. Epidermal growth factor receptor protein expression detected by IHC in CRC has been reported to range between 16% in early cancer and 97% in advanced disease. (29-35) In most of these studies, any level of membranous EGFR expression above background level, irrespective of the percentage of positive cells or the completeness of the staining or its intensity, was considered positive. Actually, many tumors display a very heterogeneous staining pattern with usually between 1% and 10% positive cells. Staining is often accentuated along the invasive front of the tumor. (36) Completeness of staining and the presence of (irrelevant) cytoplasmic staining generally parallel increasing staining intensity. About 10 studies (30,32,33,37-42) have shown a correlation between EGFR expression and known prognostic indicators of CRC including tumor stage, grade, and lymphovascular invasion. In contrast, other evaluations could not demonstrate a relation between EGFR status and tumor type, grade, or stage. (27,43) In a review of 200 studies involving more than 20 000 patients and 10 cancer types, Nicholson et al (44) showed that increased EGFR expression was associated with reduced recurrence-free or overall survival rates in 52% of studies. The EGFR status was considered by these authors to provide only a "modest prognostic indicator" in CRC. Overall, there is thus a disparity in the reports of EGFR expression as a prognostic factor. This may be due to the variable composition of the studied patient populations, tumor heterogeneity, and certainly also to the lack of a consistent method for evaluating EGFR expression as detected by immunohistochemistry. (45) The lack of a clear relationship between EGFR expression and prognosis is perhaps to be expected given that its activity, and therefore, its influence on cancer cell survival can be amplified by a number of mechanisms other than increased receptor expression.

EGFR-Targeted Therapy in CRC

Given the evident role of EGFR in tumorigenesis and cancer progression, this receptor has been selected as a relevant and promising target for anticancer therapies. In vitro and in vivo preclinical studies have shown that blocking EGFR may lead to inhibition of cancer cell growth with good clinical effect. (46) Two types of medications have been developed that interfere with either the ligand-binding domain of EGFR or its intracytoplasmic signal-transmitting domain. (28) Thus, EGFR signaling can be targeted by (1) monoclonal antibodies that compete with the natural ligands for binding to the receptor; and (2) small-molecule tyrosine kinase inhibitors (TKIs) that compete with ATP for binding to the intracellular catalytic domain. Both approaches can suppress the downstream signaling pathways in tumors whose growth is still dependent on receptor modulation. Currently, monoclonal antibodies such as cetuximab and panitumumab (phase III trials) and low-molecular-weight TKIs such as gefitinib and erlotinib (phase II studies) are in the most advanced stages of clinical development in CRC. (47) The activity of cetuximab (Erbitux, chimeric IgG1) and panitumumab (ABX-EGF or Abgenix, fully human IgG2k), as single agents or in combination with chemotherapy for the treatment of recurrent and first-line metastatic CRC (mCRC), has been studied in phase III clinical trials. (48-50) Cetuximab in combination with irinotecan was approved in 2004 by the US Food and Drug Administration for the treatment of irinotecan-refractory mCRC. The drug was also approved as a single agent for the treatment of patients with EGFR-expressing mCRC who are intolerant to irinotecan-based chemotherapy. Panitumumab has recently been approved in the United States and in Europe as a third-line treatment for mCRC. (51,52) Compared with the anti-EGFR monoclonal antibodies, TKIs have not been extensively investigated in mCRC. It seems that gefitinib (Iressa, ZD1839) as a single agent is not active in chemorefractory CRC. (48) When used in combination with the FOLFOX (folinic acid [leucovorin], fluorouracil, oxaliplatin) regimen, it induced a high objective response rate (78% for chemotherapy-naive patients and 36% for recurrent patients), but toxicity was also present with grade 3/4 diarrhea for 49% of patients. (53) The combination of erlotinib (Tarceva, OSI-774) with capecitabine and oxaliplatin is more feasible but less active. (54) Finally, there are also less well-developed approaches including antisense nucleotides or ribozymes that target EGFR mRNA (messenger RNA) and block receptor translation or intracellular antibodies that prevent migration of the receptors to the cell surface. (55)

EGFR Protein Expression as a Predictive Marker for Success of EGFR-Targeted Therapy in CRC

Both cetuximab and panitumumab have been shown to reduce the risk of tumor progression and to improve quality of life, progression-free survival, and overall survival in patients with refractory mCRC. (56-61) Such were the results of clinical trials in which patients had been selected on the basis of a positive IHC staining for EGFR in the tumor. Because of prior experience with trastuzumab (a monoclonal antibody that blocks theHER2/neu receptor) in selected cases of breast cancer, it was expected that expression of EGFR in CRC tissues would be necessary for treatment efficacy. However, this does not seem to be the case. First, only a small proportion (8%-23%) of selected patients did achieve an objective response with either antibody. Second, no apparent correlation was observed between the extent of EGFR expression and the response of tumors to second-line cetuximab, panitumumab, or gefitinib therapy. (56,57,62-64) Third, some patients with tumors in which absolutely no EGFR could be detected by IHC have still achieved an objective response to cetuximab. (65,66)

The reasons why IHC detection of EGFR is a poor predictor of response to cetuximab may include a variety of biologic, technical, and analytic factors. A recent study by Francoual et al (67) indicates that EGFR is a mixture of high- and low-affinity EGF binding sites. Only the high-affinity sites, sometimes underrepresented as compared to the low-affinity sites, would carry a meaningful biologic activity. It has been suggested that IHC is not the optimal method for quantification of the high-affinity EGF binding sites. Furthermore, the EGFR signaling pathway is complex, and it is possible that the level of expression of the endogenous receptor ligands, the level of tyrosine phosphorylation of the receptor, and the expression of other downstream molecules are critically involved in the action of cetuximab and therefore are more predictive for treatment response than the total level of the receptor per se. From a technical point of view, tissue processing and handling (type and duration of fixation, type of antibody and antigen retrieval) and prolonged storage time of tissue samples may influence detectability of protein expression. 68 Finally, the positivity rate for EGFR staining in any study depends on the cutoff point applied, and this has varied in the literature. (30-35) The correct cutoff point may actually depend upon the antibody being used. (69)

Cetuximab and panitumumab therapy are both costly and may cause side effects. It was thus necessary to look for better predictive tissue biomarkers than EGFR protein expression. The currently identified possible markers are either related to EGFR itself or to its various downstream targets.

TISSUE BIOMARKERS RELATED TO EGFR ITSELF

EGFR Gene Mutations and Polymorphisms

In contrast to the situation in non-small cell lung cancer, only very infrequent mutations in the catalytic domain of EGFR, predictive for response to gefitinib, have been observed in CRC (1 of 293 cases). (70) Nagahara et al (71) detected no somatic mutations in 11 CRC cell lines, although 4 of 33 clinical tumors (12%) presented mis-sense mutations clustered in the EGFR kinase domain in exons 19 and 20. Ogino et al (64) confirmed the rarity of activating EGFR mutations in CRC. These studies suggest that for mCRC, the impact of treatment with a TKI is likely to be limited. (72) The type III mutated variant of EGFR (EGFR vIII), which is characterized by a deletion in the extracellular domain and has been identified in various human tumors, is rare in CRC. (73) On the other hand, a recent study reported on a variant of exon 13 (R521K; arginine-to-lysine substitution at residue 521) detected in 12 of 32 investigated tumors, 11 of which achieved objective response or stable disease with cetuximab treatment. (74) Exon 13 encodes a segment of the extracellular part of the receptor. The arginine-to-lysine amino acid substitution is located at the boundary between EGFR domain III, which represents the direct interaction site with cetuximab, and domain IV. While this genetic polymorphism is frequent in the general population (20% for the homozygous and 50% for the heterozygous variant), it has also been shown to significantly reduce transforming growth factor a binding and ligand-induced EGFR signalling. (75,76) It has been speculated that the polymorphism-associated, attenuated EGFR signaling could be more sensitive to targeted receptor inhibition.

Increased EGFR Gene Copy Number

The presence of an increased EGFR gene copy number (GCN) in CRC, as detected by FISH, has been associated with a better response to anti-EGFR therapies. Moroni et al (77) found that 8 of 9 patients with objective response after treatment with cetuximab or panitumumab had an increased EGFR GCN as opposed to only 1 of 21 nonresponders (P = < .001). In a study by Frattini et al (78) in which 27 patients with mCRC were treated with cetuximab, an increased EGFR GCN was observed in 6 of 10 responders compared with only 2 of 17 nonresponders (P + < .05). On the other hand, in a phase II trial of cetuximab in refractory mCRC, Lenz et al (65) showed that an increased EGFR GCN, assessed by polymerase chain reaction (N 5 34), did not correlate with response or progression-free survival (P = .03). This is perhaps not unexpected because, during DNA extraction for polymerase chain reaction, the abnormal signal derived from the tumor cells may be diluted by the normal signal present in stromal cells. (65) However, Italiano et al (79) also reported that increased EGFR GCN detected by FISH did not select patients with CRC who might benefit from cetuximab therapy. This lack of consistency among reports may be due to the existence of various subpopulations of EGFRs with different levels of low- and high-affinity sites, the lack of standardized EGFR FISH testing methods, tumor heterogeneity, and a less than perfect correlation between EGFR DNA and protein levels (low-level polysomy and gene amplification and their relation with immunohistochemistry; Figure 4, A through C). Personeni et al (80) concluded that for the time being, it may not be warranted to make clinical decisions on the basis of previously published cutoff points. It has also been suggested that the crucial finding in all these studies is that the non-increased, rather than the increased, EGFR GCN status is the most accurate predictive element for clinical outcome. Patients with a low EGFR GCN are indeed unlikely to have tumors that respond to treatment and have a worse time to progression and overall survival than patients with an increased GCN. In those tumors with an increased GCN that do not respond to treatment, resistance to therapy may be due to constitutive activation of downstream signalling pathways by mutations of oncogenes such as KRAS, BRAF, or PI3K or by loss of a tumor suppressor gene such as PTEN. (81)

Increased Expression of EGFR Ligands

Khambata-Ford et al (82) found that patients with high gene expression levels of the EGFR ligands epiregulin and amphiregulin in their metastatic tumor lesions were more likely to achieve disease control upon cetuximab treatment. Epiregulin binds more weakly to EGFR than EGF, but is much more potent and leads to a state of prolonged receptor activation. (83) Elevated expression of epiregulin and/or amphiregulin may play an important role in tumor growth and survival by stimulating an autocrine loop through EGFR. This may characterize a tumor that is dependent on EGFR for growth factor signaling and therefore particularly sensitive to the ability of cetuximab to block ligand-receptor interaction.

Activated EGFR

Among a subgroup of 23 patients with EGFR-positive mCRC refractory to irinotecan who were treated in the BOND (Bowel Oncology With Cetuximab Antibody) study, expression of the activated/phosphorylated EGFR (pEGFR) appeared to be a useful predictor of response to cetuximab-based therapy. Disease control rates differed between patients with a pEGFR immunohistochemical score of at least 7 (7/7, 100%) and those with a score of less than 7 (7/13, 53.8%), showing a trend toward higher disease control in patients with high levels of pEGFR who were treated with cetuximab, with or without irinotecan (P = 5 .05). (84,85)

Tissue Biomarkers Related to EGFR:

Conclusion

In contrast to HER2/neu expression and trastuzumab treatment in breast cancer, EGFR protein expression, as assessed by immunohistochemistry, has no predictive value for the efficacy of EGFR-targeted therapy in CRC. Other proposed tissue biomarkers related to EGFR are still under investigation. The most promising seem to be the evaluation by FISH of the EGFR gene copy number and the detection of an increased expression of EGFR ligands.

TISSUE BIOMARKERS RELATED TO THE DOWNSTREAM SIGNALING PATHWAYS

KRAS Mutations

The KRAS oncogene is mutated in more than 30% of CRCs. (86) More than 3000 point mutations have been reported in CRC samples thus far. Somatic missense mutations in the KRAS gene lead to single amino acid substitutions and occur independently of EGFR mutations. (87) The most frequent alterations are detected in codon 12 (Gly to Asp/Val/Ala/Cys/Ser/Arg, together about 82% of all reported KRAS mutations) and 13 (mainly Gly to Asp, about 17%) in exon 2 of the gene. Mutations in other positions, such as codons 61 and 146, have also been reported but are rare and of unknown significance. (88) The KRAS gene encodes a G protein that functions downstream of EGFR in the MAPK pathway. RAS proteins normally cycle between active GTP-bound and inactive GDP-bound conformations. They are activated by guanine nucleotide exchange factors (GEFs) which are recruited to protein complexes at the intracellular domain of activated receptors. Signaling is terminated when GTP is hydrolyzed to GDP by the intrinsic hydrolase activity of the RAS protein, which is enhanced by GTPase-activating proteins (GAPs). Under physiologic conditions, levels of activated RAS are tightly controlled by the counterbalancing functions of GEFs and GAPs. Mutations in genes that encode RAS proteins disrupt this balance. Particular KRAS mutations result in RAS proteins that remain permanently in the active state because of a defective intrinsic GTPase activity and resistance to GAPs. Unlike wild-type RAS proteins, which are inactivated after a short time, the aberrant proteins continuously activate signaling pathways, even in the absence of any upstream stimulation of HER receptors. In tumors carrying KRAS mutations, blocking EGF signaling at the receptor level will not ablate the transmission of a signal in the MAPK pathway. Tumor KRAS mutation status is thus expected to have a large impact on therapeutic decisions for patients with CRC. Indeed, several clinical studies have illustrated that KRAS mutations in codon 12 and 13 lead to resistance to anti-EGFR monoclonal antibodies. (51,89-93) These mutations are associated with poor response to therapy, reduced progression-free survival, and shorter overall survival in patients with CRC who are treated with cetuximab, either alone or in combination with chemotherapy. Similarly, an analysis of KRAS mutations in tumor samples from 92% of the patients in a registrational clinical trial of panitumumab for the treatment of mCRC predicted a lack of efficacy of panitumumab on progression-free and overall survival in patients with KRAS-mutant tumors. (51) Taken together, these results indicate that KRAS mutation status is an important parameter for selecting patients for therapy: patients with mutant tumors will not benefit from EGFR-targeted therapies. On the basis of these data, the European Medicines Agency has approved the use of cetuximab and panitumumab for the treatment of mCRC in patients who carry a normal, wild-type KRAS gene. (52) However, as only a fraction of patients with wild-type KRAS in colorectal tumors can achieve a clinical response with EGFR-targeted therapies, the search for additional predictive parameters must go on. The tissue chosen for KRAS mutation testing is usually the primary tumor as KRAS mutations develop early on in the carcinogenic sequence. (94) The representativity of endoscopically obtained tumor samples versus biopsies taken from surgical specimens is still being investigated. Various user-developed and commercial techniques are currently applied for the detection of KRAS mutations in CRC. A European Quality Assurance Program has been proposed. This will be organized by the European Society of Pathology in close collaboration with existing regional and national quality assurance programs. (95)

BRAF Mutations

BRAF is the principal effector of KRAS. It has thus been hypothesized that in patients with wild-type KRAS, BRAF mutations could have a predictive value for response to cetuximab or panitumumab. Di Nicolantonio et al (96) retrospectively analyzed objective tumor responses, time to progression, overall survival, and the mutational status of KRAS and BRAF in 113 tumors from patients with mCRC treated with cetuximab or panitumumab. The effect of the BRAF Val600Glu mutation (exon 15) on response was also assessed by using cellular models of CRC. KRAS mutations were present in 30% of the patients and were associated with resistance to cetuximab or panitumumab (P = .01). None of the KRAS-mutated samples carried a BRAF mutation, in accordance with previous findings on the mutual exclusivity of KRAS and BRAF mutations in CRC. The BRAF mutation was detected in 11 of 79 patients who had wild-type KRAS. None of the patients with this BRAF mutation responded to treatment, whereas none of the responders carried BRAF mutations (P = .03). Patients with BRAF mutation had significantly shorter progression-free survival (P = 5 .01) and overall survival (P = < .001) than patients with the wild-type gene. The introduction of a BRAF allele resulting in Val600Glu substitution in CRC cell lines impaired the therapeutic effect of cetuximab or panitumumab. On the other hand, treatment of mutant cell lines with the BRAF inhibitor sorafenib restored sensitivity to the EGFR inhibitors. A BRAF wild-type status thus seems to be required for tumor response and can potentially be used to select patients eligible for treatment. The authors (96) also suggested to test double-hit therapies with simultaneous inhibition of EGFR and BRAF in patients with CRC who carry the Val600Glu mutation.

Malfunctions of the PI3K/PTEN/AKT Pathway

Preclinical studies on tumor cell lines (A431, MDA-468, and H157) have suggested that PTEN inactivation is a predictor for resistance to EGFR-family antagonists. (97-99) PTEN is a lipid phosphatase and a tumor suppressor protein that regulates the PI3K/AKT signaling pathway. When PTEN function is lost, its major substrate, phosphatidylinositol 3,4,5-trisphosphate, which is a second messenger of PI3K, accumulates in the cell membrane, where it binds and activates AKT. Thus, the loss of PTEN function results in constitutive overactivation of the PI3K/ AKT pathway, increasing its antiapoptotic functions. PTEN loss has been associated with resistance to trastuzumab, which binds to the HER2/neu receptor. (100) As HER2/neu and EGFR form heterodimers upon ligand binding and use identical downstream signaling pathways, it seemed reasonable to test PTEN as a potential predictive factor for cetuximab resistance. PTEN expression has been found to be decreased in approximately 40% of colorectal cancers, often with associated mutations, deletions, or promoter hypermethylation. (101) Frattini et al (78) showed that PTEN tumor expression was correlated with response to cetuximab for 27 patients with mCRC. All responder patients were found to express PTEN in their tumor (10/10) versus 35% of nonresponder patients (6/17) (P < .001). In this study, PTEN expression was detected by IHC staining. PTEN-negative tumors were defined as those with a dramatic reduction or absence of immunostaining in at least 50% of the tumor cells, as compared with the internal control (eg, nerves).

AKT was also identified in preclinical studies as a major resistance factor for anti-EGFR therapy. (99) Perez-Soler et al (102) showed that erlotinib-resistant clones of human squamous cancer cell lines had significantly higher levels of total and phosphorylated AKT, suggesting persistent downstream signaling even in the presence of EGFR inhibitors. These findings need to be confirmed in colorectal carcinomas.

Investigation of cetuximab response in colon cancer cell lines suggests that constitutive and simultaneous activation of the MAPK and PI3K/PTEN/AKT pathways confers maximal resistance to this agent. A priori screening of colon tumors for PTEN expression status and PI3K and KRAS/ BRAF mutation status could help in stratifying patients who are likely to benefit from this therapy. (103)

Tissue Biomarkers Related to Signaling Pathways Downstream From EGFR: Conclusion

The results on the investigation of KRAS mutation status as a marker for clinical benefit of anti-EGFR antibodies are so impressive that they have already had a major impact on treatment strategies. The European Medicines Agency has approved panitumumab for the treatment of patients with wild-type KRAS tumors only. This would be the first example of the approval of a drug therapy for solid tumors that is based on a genetic test. The 10 ongoing cetuximab studies sponsored by the National Cancer Institute in the United States are currently being amended to include KRAS testing. Further molecular analysis of tumor specimens banked for these and other studies may yield other markers that can identify patients who have a chance to benefit from EGFR-targeting antibodies.

TAKE-HOME MESSAGES

1. Management of CRC is now a team process. Pathologists play an important role in diagnosis, prognostication, and, increasingly, selection of patients likely to react to particular therapeutic regimens.

2. Tissue biomarkers indicative for response to targeted therapies are often identified only when the medication is already being used clinically. Pathologists can participate in this endeavor by their access to and knowledge of the structure of tumor tissues and the various techniques with which these can be investigated.

3. Several biomarkers have been investigated with regard to EGFR-related targeted therapies in CRC. The KRAS mutation status of the tumor has a proven predictive value and is currently determined for many patients in a metastatic setting. Pathologists can assist in this effort by ensuring investigation of the proper tissues with the optimal techniques.

References

(1.) Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin. 2005;55(2):74-108.

(2.) Quirke P, Morris E. Reporting colorectal cancer. Histopathology. 2007; 50(1):103-112.

(3.) Rex DK, Ulbright TM, Cummings OW. Coming to terms with pathologists over colon polyps with cancer or high-grade dysplasia. J Clin Gastroenterol. 2005;39(1):1-3.

(4.) Saclarides T. Minimally invasive surgery--where are we? In: Schofield JH, Abcarian H, Grothey A, Maughan T, eds. Challenges in Colorectal Cancer. 2nd ed. Oxford, UK: Blackwell Publishing; 2006:73-89.

(5.) Kikuchi R, Takano M, Takagi K, et al. Management of early invasive colorectal cancer: risk of recurrence and clinical guidelines. Dis Colon Rectum. 1995;38(12):1286-1295.

(6.) Quirke P, Williams GT. Minimum dataset for colorectal cancer histopathology reports. http://www.bioinformatics.leeds.ac.uk/tmadb/htdocs_tmadb/ colorectalcancer.pdf. Published 1998. Accessed August 3, 2009.

(7.) Branston LK, Greening S, Newcombe RG, et al. The implementation of guidelines and computerised forms improves the completeness of cancer pathology reporting--The CROPS project: a randomised controlled trial in pathology. Eur J Cancer. 2002;38(6):764-772.

(8.) Moertel CG, Fleming TR, Macdonald JS, et al. Fluorouracil plus levamisole as effective adjuvant therapy after resection of stage III colon carcinoma: a final report. Ann Intern Med. 1995;122(5):321-326.

(9.) Andre T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med. 2004;350(23): 2343-2351.

(10.) Twelves C, Wong A, Nowacki MP, et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med. 2005;352(26):2696-2704.

(11.) Quasar Collaborative Group, Gray R, Barnwell J, McConkey C, Hills RK, Williams NS, Kerr DJ. Adjuvant chemotherapy versus observation in patients with colorectal cancer: a randomised study. Lancet. 2007;370(9604):2020-2029.

(12.) Moertel CG, Fleming TR, Macdonald JS, et al. Intergroup study of fluorouracil plus levamisole as adjuvant therapy for stage II/Dukes' B2 colon cancer. J Clin Oncol. 1995;13(12):2936-2943.

(13.) Wibe A, Rendedal PR, Svensson E, et al. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg. 2002;89(3):327-334.

(14.) Nagtegaal ID, Marijnen CA, Kranenbarg EK, van de Velde CJ, van Krieken JH; Pathology Review Committee; Cooperative Clinical Investigators. Circumferential margin involvement is still an important predictor of local recurrence in rectal carcinoma: not one millimeter but two millimeters is the limit. Am J Surg Pathol. 2002;26(3):350-357.

(15.) Bateman AC, Carr NJ, Warren BF. The retroperitoneal surface in distal caecal and proximal ascending colon carcinoma: the Cinderella surgical margin? J Clin Pathol. 2005;58(4):426-428.

(16.) Dworak O, Keilholz L, Hoffmann A. Pathological features of rectal cancer after preoperative radiochemotherapy. Int J Colorectal Dis. 1997;12(1):19-23.

(17.) Rutten HS-MD, Glynne-Jones R, Rullier E, et al. Capecitabine, oxaliplatin, radiotherapy, and excision (CORE) in patients with MRI-defined locally advanced rectal adenocarcinoma: results of an international multicenter phase II study [abstract]. J. Clin Oncol. 2006;24(18s):A3528.

(18.) Visser O, Bakx R, Zoetmulder FA, et al. The influence of total mesorectal excision on local recurrence and survival in rectal cancer patients: a population-based study in Greater Amsterdam. J Surg Oncol. 2007;95(6):447-454.

(19.) Hermanek P, Hermanek P, Hohenberger W, Klimpfinger M, Kockerling F, Papadopoulos T. The pathological assessment of mesorectal excision: implications for further treatment and quality management. Int J Colorectal Dis. 2003; 18(4):335-341.

(20.) Adam IJ, Mohamdee MO, Martin IG, et al. Role of circumferential margin involvement in the local recurrence of rectal cancer. Lancet. 1994;344(8924): 707-711.

(21.) Maslekar S, Sharma A, Macdonald A, Gunn J, Monson JR, Hartley JE. Mesorectal grades predict recurrences after curative resection for rectal cancer. Dis Colon Rectum. 2007;50(2):168-175.

(22.) Nagtegaal ID, van Krieken JH. The role of pathologists in the quality control of diagnosis and treatment of rectal cancer--an overview. Eur J Cancer. 2002;38(7):964-972.

(23.) Parfitt JR, Driman DK. The total mesorectal excision specimen for rectal cancer: a review of its pathological assessment. J Clin Pathol. 2007;60(8):849- 855.

(24.) Compton CC; for the Members of the Cancer Committee, College of American Pathologists. Updated protocol for the examination of specimens from patients with carcinomas of the colon and rectum, excluding carcinoid tumors, lymphomas, sarcomas, and tumors of the vermiform appendix: a basis for checklists. Arch Pathol Lab Med. 2000;124(7):1016-1025.

(25.) Hamilton SR. Targeted therapy of cancer: new roles for pathologists in colorectal cancer. Mod Pathol. 2008;21(suppl 2):S23-S30.

(26.) Earp HS III, Calvo BF, Sartor CI. The EGF receptor family--multiple roles in proliferation, differentiation, and neoplasia with an emphasis on HER4. Trans Am Clin Climatol Assoc. 2003;114:315-333.

(27.) Koretz K, Schlag P, Moller P. Expression of epidermal growth factor receptor in normal colorectal mucosa, adenoma, and carcinoma. Virchows Arch. 1990;416(4):343-349.

(28.) Tabernero J, Salazar R, Casado E, Martinelli E, Gomez P, Baselga J. Targeted therapy in advanced colon cancer: the role of new therapies. Ann Oncol. 2004;15(suppl 4):iv55-iv62.

(29.) Shirai H, Ueno E, Osaki M, Tatebe S, Ito H, Kaibara N. Expression of growth factors and their receptors in human early colorectal carcinomas: immunohistochemical study. Anticancer Res. 1995;15(6B):2889-2894.

(30.) Goldstein NS, Armin M. Epidermal growth factor receptor immunohistochemical reactivity in patients with American Joint Committee on Cancer Stage IV colon adenocarcinoma: implications for a standardized scoring system. Cancer. 2001;92(5):1331-1346.

(31.) Resnick MB, Routhier J, Konkin T, Sabo E, Pricolo VE. Epidermal growth factor receptor, c-MET, beta-catenin, and p53 expression as prognostic indicators in stage II colon cancer: a tissue microarray study. Clin Cancer Res. 2004;10(9): 3069-3075.

(32.) Scartozzi M, Bearzi I, Berardi R, Mandolesi A, Fabris G, Cascinu S. Epidermal growth factor receptor (EGFR) status in primary colorectal tumors does not correlate with EGFR expression in related metastatic sites: implications for treatment with EGFR-targeted monoclonal antibodies. J Clin Oncol. 2004;22(23): 4772-4778.

(33.) Spano JP, Lagorce C, Atlan D, et al. Impact of EGFR expression on colorectal cancer patient prognosis and survival. Ann Oncol. 2005;16(1):102-108.

(34.) Bralet MP, Paule B, Adam R, Guettier C. Loss of epidermal growth factor receptor expression in lymph node and liver metastases of colon carcinoma. J Clin Oncol. 2005;23(24):5844.

(35.) Galizia G, Lieto E, Ferraraccio F, et al. Prognostic significance of epidermal growth factor receptor expression in colon cancer patients undergoing curative surgery. Ann Surg Oncol. 2006;13(6):823-835.

(36.) Abd El All HS, Mishriky AM, Mohamed FA. Epidermal growth factor receptor in colorectal carcinoma: correlation with clinico-pathological prognostic factors. Colorectal Dis. 2008;10(2):170-178.

(37.) Steele RJ, Kelly P, Ellul B, Eremin O. Epidermal growth factor receptor expression in colorectal cancer. Br J Surg. 1990;77(12):1352-1354.

(38.) Hemming AW, Davis NL, Kluftinger A, et al. Prognostic markers of colorectal cancer: an evaluation of DNA content, epidermal growth factor receptor, and Ki-67. J Surg Oncol. 1992;51(3):147-152.

(39.) Kluftinger AM, Robinson BW, Quenville NF, Finley RJ, Davis NL. Correlation of epidermal growth factor receptor and c-erbB2 oncogene product to known prognostic indicators of colorectal cancer. Surg Oncol. 1992;1(1):97- 105.

(40.) 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(2):133-137.

(41.) De Jong KP, Stellema R, Karrenbeld A, et al. Clinical relevance of transforming growth factor alpha, epidermal growth factor receptor, p53, and Ki67 in colorectal liver metastases and corresponding primary tumors. Hepatology. 1998;28(4):971-979.

(42.) McKay JA, Murray LJ, Curran S, et al. Evaluation of the epidermal growth factor receptor (EGFR) in colorectal tumours and lymph node metastases. Eur J Cancer. 2002;38(17):2258-2264.

(43.) Yasui W, Sumiyoshi H, Hata J, et al. Expression of epidermal growth factor receptor in human gastric and colonic carcinomas. Cancer Res. 1988;48(1):137- 141.

(44.) Nicholson RI, Gee JM, Harper ME. EGFR and cancer prognosis. Eur J Cancer. 2001;37(suppl 4):S9-S15.

(45.) Dei Tos AP, Ellis I. Assessing epidermal growth factor receptor expression in tumours: what is the value of current test methods? Eur J Cancer. 2005;41(10): 1383-1392.

(46.) Lockhart AC, Berlin JD. The epidermal growth factor receptor as a target for colorectal cancer therapy. Semin Oncol. 2005;32(1):52-60.

(47.) Mendelsohn J, Baselga J. Status of epidermal growth factor receptor antagonists in the biology and treatment of cancer. J Clin Oncol. 2003;21(14): 2787-2799.

(48.) Van Cutsem E. Challenges in the use of epidermal growth factor receptor inhibitors in colorectal cancer. Oncologist. 2006;11(9):1010-1017.

(49.) Venook A, Niedzwiecki D, Hollis D, et al. Phase III study of irinotecan/ 5FU/LV (FOLFIRI) or oxaliplatin/5FU/LV (FOLFOX) 6 cetuximab for patients with untreated metastatic adenocarcinoma of the colon or rectum: CALGB 80203 preliminary results [abstract 3509]. J Clin Oncol. 2006;24(18S):148s.

(50.) Peeters M, Van Cutsem E, Siena S, et al. A phase 3, multicenter, randomized controlled trial of panitumumab plus best supportive care (BSC) vs. BSC alone in patients with metastatic colorectal cancer. Proc Am Assoc Cancer Res. 2006;47:CP-1.

(51.) Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26(10):1626-1634.

(52.) Ciardiello F, Tortora G. EGFR antagonists in cancer treatment. N Engl J Med. 2008;358(11):1160-1174.

(53.) Fisher GA, Kuo T, Cho CD, et al. A phase II study of gefitinib in combination with FOLFOX-4 (IFOX) in patients with metastatic colorectal cancer [abstract 3514]. Proc Am Soc Clin Oncol. 2004;22:249.

(54.) Van Cutsem E, Beale P, Delord J, et al. Tarceva (erlotinib) in combination with Xeloda and oxaliplatin in patients with metastatic colorectal cancer: a phase I dose escalation trial. Clin Cancer Res. 2003;9:109a.

(55.) Ono M, Kuwano M. Molecular mechanisms of epidermal growth factor receptor (EGFR) activation and response to gefitinib and other EGFR-targeting drugs. Clin Cancer Res. 2006;12(24):7242-7251.

(56.) Saltz LB, Meropol NJ, Loehrer PJ Sr, Needle MN, Kopit J, Mayer RJ. Phase II trial of cetuximab in patients with refractory colorectal cancer that expresses the epidermal growth factor receptor. J Clin Oncol. 2004;22(7):1201-1208.

(57.) 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(4):337-345.

(58.) Gibson TB, Ranganathan A, Grothey A. Randomized phase III trial results of panitumumab, a fully human anti-epidermal growth factor receptor monoclonal antibody, in metastatic colorectal cancer. Clin Colorectal Cancer. 2006;6(1):29-31.

(59.) Siena S, Peeters M, Van Cutsem E, et al. Association of progression-free survival with patient-reported outcomes and survival: results from a randomised phase 3 trial of panitumumab. Br J Cancer. 2007;97(11):1469-1474.

(60.) Van Cutsem E, Peeters M, Siena S, et al. Open-label phase III trial of panitumumab plus best supportive care compared with best supportive care alone in patients with chemotherapy-refractory metastatic colorectal cancer. J Clin Oncol. 2007;25(13):1658-1664.

(61.) Jonker DJ, O'Callaghan CJ, Karapetis CS, et al. Cetuximab for the treatment of colorectal cancer. N Engl J Med. 2007;357(20):2040-2048.

(62.) Hecht J, Mitchell E, Baranda J, et al. Panitumumab antitumor activity in patients with metastatic colorectal cancer expressing low (1-9%) or negative (<1%) levels of epidermal growth factor receptor [abstract 3547]. J Clin Oncol. 2006;24(suppl):157s.

(63.) Berlin J, Neubauer M, Swanson P, et al. Panitumumab antitumor activity in patients with metastatic colorectal cancer expressing $10% epidermal growth factor receptor [abstract 3548]. J Clin Oncol. 2006;24(suppl):158s.

(64.) Ogino S, Meyerhardt JA, Cantor M, et al. Molecular alterations in tumors and response to combination chemotherapy with gefitinib for advanced colorectal cancer. Clin Cancer Res. 2005;11(18):6650-6656.

(65.) Lenz HJ, Van Cutsem E, Khambata-Ford S, et al. Multicenter phase II and translational study of cetuximab in metastatic colorectal carcinoma refractory to irinotecan, oxaliplatin, and fluoropyrimidines. J Clin Oncol. 2006;24(30):4914- 4921.

(66.) Chung KY, Shia J, Kemeny NE, et al. Cetuximab shows activity in colorectal cancer patients with tumors that do not express the epidermal growth factor receptor by immunohistochemistry. J Clin Oncol. 2005;23(9):1803-1810.

(67.) Francoual M, Etienne-Grimaldi MC, Formento JL, et al. EGFR in colorectal cancer: more than a simple receptor. Ann Oncol. 2006;17(6):962-967.

(68.) Atkins D, Reiffen KA, Tegtmeier CL, Winther H, Bonato MS, Storkel S. Immunohistochemical detection of EGFR in paraffin-embedded tumor tissues: variation in staining intensity due to choice of fixative and storage time of tissue sections. J Histochem Cytochem. 2004;52(7):893-901.

(69.) Buffet W, Geboes KP, De Hertogh G, Geboes K. EGFR-immunohistochemistry in colorectal cancer and non-small cell lung cancer: comparison of 3 commercially available EGFR-antibodies. Acta Gastroenterol Belg. 2008;71(2): 213-218.

(70.) Barber TD, Vogelstein B, Kinzler KW, Velculescu VE. Somatic mutations of EGFR in colorectal cancers and glioblastomas. N Engl J Med. 2004;351(27): 2883.

(71.) Nagahara H, Mimori K, Ohta M, et al. Somatic mutations of epidermal growth factor receptor in colorectal carcinoma. Clin Cancer Res. 2005;11(4): 1368-1371.

(72.) Moroni M, Sartore-Bianchi A, Benvenuti S, Artale S, Bardelli A, Siena S. Somatic mutation of EGFR catalytic domain and treatment with gefitinib in colorectal cancer. Ann Oncol. 2005;16(11):1848-1849.

(73.) Azuma M, Danenberg KD, Iqbal S, et al. Epidermal growth factor receptor and epidermal growth factor receptor variant III gene expression in metastatic colorectal cancer. Clin Colorectal Cancer. 2006;6(3):214-218.

(74.) Goncalves A, Esteyries S, Taylor-Smedra B, et al. A polymorphism of EGFR extracellular domain is associated with progression free-survival in metastatic colorectal cancer patients receiving cetuximab-based treatment. BMC Cancer. 2008;8:169.

(75.) Wang WS, Chen PM, Chiou TJ, et al. Epidermal growth factor receptor R497K polymorphism is a favorable prognostic factor for patients with colorectal carcinoma. Clin Cancer Res. 2007;13(12):3597-3604.

(76.) Moriai T, Kobrin MS, Hope C, Speck L, Korc M. A variant epidermal growth factor receptor exhibits altered type alpha transforming growth factor binding and transmembrane signaling. Proc Natl Acad Sci U S A. 1994;91(21): 10217-10221.

(77.) Moroni M, Veronese S, Benvenuti S, et al. Gene copy number for epidermal growth factor receptor (EGFR) and clinical response to antiEGFR treatment in colorectal cancer: a cohort study. Lancet Oncol. 2005;6(5):279-286.

(78.) Frattini M, Saletti P, Romagnani E, et al. PTEN loss of expression predicts cetuximab efficacy in metastatic colorectal cancer patients. Br J Cancer. 2007; 97(8):1139-1145.

(79.) Italiano A, Follana P, Caroli FX, et al. Cetuximab shows activity in colorectal cancer patients with tumors for which FISH analysis does not detect an increase in EGFR gene copy number. Ann Surg Oncol. 2008;15(2):649-654.

(80.) Personeni N, Fieuws S, Piessevaux H, et al. Clinical usefulness of EGFR gene copy number as a predictive marker in colorectal cancer patients treated with cetuximab: a fluorescent in situ hybridization study. Clin Cancer Res. 2008; 14(18):5869-5876.

(81.) Moroni M, Sartore-Bianchi A, Veronese S, Siena S. EGFR FISH in colorectal cancer: what is the current reality? Lancet Oncol. 2008;9(5):402-403.

(82.) Khambata-Ford S, Garrett CR, Meropol NJ, et al. Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab. J Clin Oncol. 2007;25(22): 3230-3237.

(83.) Shelly M, Pinkas-Kramarski R, Guarino BC, et al. Epiregulin is a potent pan-ErbB ligand that preferentially activates heterodimeric receptor complexes. J Biol Chem. 1998;273(17):10496-10505.

(84.) Personeni N, Hendlisz A, Gallez J, et al. Correlation between the response to cetuximab alone or in combination with irinotecan and the activated/ phosphorylated epidermal growth factor receptor in metastatic colorectal cancer. Semin Oncol. 2005;32(6 suppl 9):S59-S62.

(85.) Spano JP, Milano G, Vignot S, Khayat D. Potential predictive markers of response to EGFR-targeted therapies in colorectal cancer. Crit Rev Oncol Hematol. 2008;66(1):21-30.

(86.) Edkins S, O'Meara S, Parker A, et al. Recurrent KRAS codon 146 mutations in human colorectal cancer. Cancer Biol Ther. 2006;5(8):928-932.

(87.) Kosaka T, Yatabe Y, Endoh H, Kuwano H, Takahashi T, Mitsudomi T. Mutations of the epidermal growth factor receptor gene in lung cancer: biological and clinical implications. Cancer Res. 2004;64(24):8919-8923.

(88.) Oliveira C, Westra JL, Arango D, et al. Distinct patterns of KRAS mutations in colorectal carcinomas according to germline mismatch repair defects and hMLH1 methylation status. Hum Mol Genet. 2004;13(19):2303-2311.

(89.) Lievre A, Bachet JB, Le Corre D, et al. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Cancer Res. 2006;66(8): 3992-3995.

(90.) Benvenuti S, Sartore-Bianchi A, Di Nicolantonio F, et al. Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies. Cancer Res. 2007;67(6):2643-2648.

(91.) De Roock W, Piessevaux H, De Schutter J, et al. KRAS wild-type state predicts survival and is associated to early radiological response in metastatic colorectal cancer treated with cetuximab. Ann Oncol. 2008;19(3):508-515.

(92.) Di Fiore F, Blanchard F, Charbonnier F, et al. Clinical relevance of KRAS mutation detection in metastatic colorectal cancer treated by Cetuximab plus chemotherapy. Br J Cancer. 2007;96(8):1166-1169.

(93.) Lievre A, Bachet JB, Boige V, et al. KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab. J Clin Oncol. 2008;26(3):374-379.

(94.) Andreyev HJ, Norman AR, Cunningham D, et al. Kirsten ras mutations in patients with colorectal cancer: the 'RASCAL II' study. Br J Cancer. 2001;85(5): 692-696.

(95.) Van Krieken JH, Jung A, Kirchner T, et al. KRAS mutation testing for predicting response to anti-EGFR therapy for colorectal carcinoma: proposal for an European quality assurance program. Virchows Arch. 2008;453(5):417- 431.

(96.) Di Nicolantonio F, Martini M, Molinari F, et al. Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer. J Clin Oncol. 2008;26(35):5705-5712.

(97.) She QB, Solit D, Basso A, Moasser MM. Resistance to gefitinib in PTEN-null HER-overexpressing tumor cells can be overcome through restoration of PTEN function or pharmacologic modulation of constitutive phosphatidylinositol 39-kinase/Akt pathway signaling. Clin Cancer Res. 2003;9(12):4340-4346.

(98.) Bianco R, Shin I, Ritter CA, et al. Loss of PTEN/MMAC1/TEP in EGF receptor-expressing tumor cells counteracts the antitumor action of EGFR tyrosine kinase inhibitors. Oncogene. 2003;22(18):2812-2822.

(99.) Camp ER, Summy J, Bauer TW, Liu W, Gallick GE, Ellis LM. Molecular mechanisms of resistance to therapies targeting the epidermal growth factor receptor. Clin Cancer Res. 2005;11(1):397-405.

(100.) Mellinghoff IK, Cloughesy TF, Mischel PS. PTEN-mediated resistance to epidermal growth factor receptor kinase inhibitors. Clin Cancer Res. 2007;13(2, pt 1):378-381.

(101.) Goel A, Arnold CN, Niedzwiecki D, et al. Frequent inactivation of PTEN by promoter hypermethylation in microsatellite instability-high sporadic colorectal cancers. Cancer Res. 2004;64(9):3014-3021.

(102.) Perez-Soler R, Ling Y, Lia M. Molecular mechanisms of resistance to the HER/EGFR tyrosine kinase inhibitor erlotinib HCL in human cancer cell lines. Proc Am Soc Clin Oncol. 2003;22:190.

(103.) Jhawer M, Goel S, Wilson AJ, et al. PIK3CA mutation/PTEN expression status predicts response of colon cancer cells to the epidermal growth factor receptor inhibitor cetuximab. Cancer Res. 2008;68(6):1953-1961.

Gert De Hertogh, MD, PhD; Karen Paula Geboes, MD, PhD

Accepted for publication August 12, 2009.

From the Department of Pathology, University Hospitals KU Leuven, Leuven, Belgium (Dr De Hertogh); and the Department of Gastroenterology, Division of Digestive Oncology, University Hospital U Gent, Gent, Belgium (Dr Geboes).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Gert De Hertogh, MD, PhD, Department of Pathology, University Hospitals KULeuven, Minderbroedersstraat 12, 3000 Leuven, Belgium (e-mail: [e-mail]).