Clinicopathological value of epidermal growth factor receptor (EGFR) and Ki-67 expression in colorectal adenoma and adenocarcinoma

Objective: To evaluate epidermal growth factor receptor and Ki-67 immunohistochemical expression in colorectal adenoma and carcinoma cases, and to relate their expression with the available clinicopathological data. Methods: The retrospective study was conducted at the Faculty of Medicine, Kafrelsheikh University, Egypt, from September 2019 to October 2020, and comprised formalin-fixed and paraffin-embedded specimens related to cases of colorectal adenoma and those of colorectal carcinoma, who had no previous radiation or chemotherapeutic treatment. Immunohistochemical staining of all thecases was done using anti-epidermal growth factor receptor and anti-Ki-67 antibodies. Data was analysed using SPSS 20. Results: Of the 70 cases, 20(28.5%) were of colorectal adenoma; mean age 52.95±13.47 years, and male-to-female ratio 1:1. The remaining 50(71.5%) cases had colorectal carcinoma; mean age 51.08±13.49 years, and male-to-female ratio 1.17:1. Epidermal growth factor receptorand Ki-67 overexpression related significantly to villous histopathological type and high-grade dysplasia in colorectal adenoma cases (p<0.05). In colorectal carcinoma cases, epidermal growth factor receptoroverexpression related significantly to tumour grade (p<0.05). Ki-67 overexpression related significantly to increased pathological stage (p<0.05). Conclusions: Overexpression of epidermal growth factor receptor and Ki-67 was found to be an ominous sign of colorectal adenoma aggressiveness, and the risk of progression to colorectal carcinoma.


Introduction
The third most prevalent cancer in the world is colorectal carcinoma (CRC). 1 It is the third leading cause of cancerrelated death in both men and women in the United States, after lung and breast cancer in women, and prostate and lung cancer in men. 2 CRC is the seventh most prevalent cancer in Egypt, accounting for 3.47% of male cancers, 3% of female cancers, and the top malignancy of the digestive tract. 3 Egypt has a high rate of young-aged CRC, as nearly 35% of Egyptian CRC patients are age <40 years and have a three-fold risk to die within 5 years compared to those who have CRC and are aged >50 years. 4 The existence of adenomas is considered an important risk factor of CRC which develops as slowly growing tumour in the luminal wall, acquiring an increasing degree of dysplasia before progressing to malignancy. 5 Although the most important and independent prognostic factor approved in CRC cases remains the tumour-node-metastasis (TNM) staging, 6 molecular biomarkers with prognostic or predictive value are also beneficial in selecting patients for preferred therapy and to prevent toxic side effects of the treatment. 7 Epidermal growth factor receptor (EGFR) is a glycoprotein transmembrane which is a part of the ErbB tyrosine kinase receptors, expressed by the proto-oncogene c-ErbB. Activation of EGFR stimulates carcinogenesis by increasing cell proliferation, migration, angiogenesis and apoptosis inhibition. It is also associated with metastasis, bad outcome and poor response to adjuvant postoperative therapy. 8 Currently, EGFR inhibitors are of value in metastatic CRC therapy and they have effective treatment options in combination with chemotherapy in EGFR-positive patients. 9 Ki-67 is a proliferation biomarker used in measuring the growth rate of cells in human tumours. Its expression is highly associated with tumour proliferation and is used in routine investigations as a prognostic and predictive biomarker in cancer. In CRC, Ki-67 expression has an undefined prognostic value. The high Ki-67 expression is related to bad prognosis and metastasis. 10 The current study was planned to evaluate EGFR and Ki-67 immunohistochemical (IHC) expression in colorectal adenoma and carcinoma cases, and to relate their expression with the available clinicopathological data.

Material and Methods
The retrospective study was conducted at the Faculty of Medicine, Kafrelsheikh University, Egypt, from September 2019 to October 2020, and comprised formalin-fixed and paraffin-embedded (FFPE) specimens related to cases of colorectal adenoma and those of CRC, who had no previous radiation or chemotherapeutic treatment. The specimens were collected using comprehensive sampling technique from the Pathology Department and some private laboratories after approval from the institutional ethics review committee. Informed consent had been obtained from all the patients at the time of their treatment. Exclusion criteria comprised patients who had previously received radiotherapy or chemotherapy, insufficient tissue for immunostaining or poor quality of the blocks. Clinicopathological data of the patients regarding age, gender and clinical presentation was obtained from patient files, including surgery and oncology reports.
Sections were stained using haematoxylin and eosin (H&E), and optical microscope was used to examine them for the confirmation of the diagnosis and classification of histopathological types, tumour grade and other histological features, such as pT stage and lymphovascular invasion (LVI). The grade of dysplasia in adenoma cases was classified into high-grade (HG) or low-grade (LG). HG dysplasia included intra-mucosal carcinoma. 11 Colorectal carcinoma cases were graded and categorized according to American Joint Committee on Cancer (AJCC) staging system (2018) at the time of diagnosis. 12 A representative haematoxylin and eosin (H&E) stained section was selected for each case and sections 3 micronthick were cut from the corresponding block and mounted on positively-charged coated slides. The conventional manual IHC technique was performed using a method described in 2010. 13 The primary antibodies used were (Anti-EGFR, clone SP84, rabbit monoclonal antibody, SAB5500096-100UL [Cell Marque, Sigma-Aldrich Co., USA] and Anti-Ki-67, clone RM360, rabbit monoclonal antibody, SAB5600249-100UL [Cell Marque, Sigma-Aldrich Co., USA]).
Semi-quantitative analysis was used to determine the EGFR expression patterns. Following were the ratings for EGFR immunoreactivity intensity; 0 = no response; 1+ = poor intensity, weak membranous, cytoplasmic or both types of brown reactivity; 2+ = moderate intensity, brown stains in various shades and usually variegated reactive cell membranes; 3+ = high intensity for which a thick outline of the cell was created by a dark brown stain that covered the whole membrane pattern. 14 The mean proportion of EGFR-reactive cells per instance was used to calculate the degree of EGFR immunoreactivity, and the results were evaluated as: 1+ = 1-10%, 2+ = >10-50%)%, and 3+ = >50% of EGFR-reactive cells, respectively. The intensity and extent of staining for each tissue segment were multiplied to produce the EGFR IHC composite score, which varied from 0 to 9, and was then divided into composite scores of to differentiate between low and high EGFR expression. 15 Nuclear staining that is uniform or granular is a sign of a good response to Ki-67. The position of the area with the highest marking was considered when choosing the hot spot For which 500 cells were counted and at least 5 high-power fields were selected. The number of Ki-67positive nuclei divided by the total number of cells examined was used to construct the Ki-67 proliferative labelling index (LI), which was then represented as a percentage. 16 A cut-off value of 25% was chosen to categorise the samples into low (25% positively-stained tumour cells) or high nuclear Ki-67 expression (25% positively-stained tumour cells) based on the percentage of positively stained tumour cells in various representative visual fields. 17 Data was analysed using IBM SPSS 20. Qualitative data was expressed as frequencies and percentages. Quantitative data was reported and mean and standard deviation, or as median and interquartile range (IQR), as appropriate. When more than 20% of the cells had an estimated count <5, Fisher's Exact or Monte Carlo correction for chi-square tests was utilised. To compare results across several groups, categorical variables were tested using the chi-square test. P<0.05 was considered statistically significant.
In CRC cases, 29(58%) showed high EGFR composite score (≥6), and 21(42%) showed a low score (<6) (Figure 2). EGFR high composite score increased in tumours showing LVI 15(75%). The score had a significant relationship with the tumour grade and the presence of LVI (p<0.05), but it had no significant relationship with the histopathological type, the pT stage and the N status (p>0.05).
CRC cases showed a high score of Ki-67 LI 38(76%). The relation between Ki-67 LI score and tumour grade was significant (p<0.05). The relation between Ki-67 LI score and pT stage was significant (p<0.05). The Ki-67 LI score showed no significant relationship with the histopathological type, the N status and the presence of LVI (p>0.05) ( Table 2).
The expression of EGFR and Ki-67 LI in CRC was increased compared to adenoma tissue, but the difference was not significant (p>0.05). EGFR expression demonstrated a significant connection to the expression of Ki-67 in adenoma and CRC (p<0.05) (Figure 3).

Discussion
The EGFR IHC expression in colorectal adenoma case in the current study was in concordance with Williet et al. 18 . In cases of investigated adenoma, the grade and type of histopathological dysplasia were significantly correlated with the Ki-67 LI, which is close to earlier findings, 19 , but findings in contrast have also been reported. 20 In the current study, 58% CRC cases showed overexpression of EGFR compared to Spano et al. 15 who showed a higher percentage (80%). Most other studies showed variable ranges (25-82%) of EGFR overexpression in CRC. 21 Despite such issues with reproducibility and validation, IHC testing is still one of the most used ways to determine EGFR expression. In CRC cases, the EGFR composite score revealed a significant connection with tumour grade. High EGFR expression was mainly seen in moderately and poorly-differentiated HG cases. Similar 15 and conflicting 22,23 findings have been reported. High EGFR expression was more observed in tumours that showed LVI than tumours that did not show this feature, but Yun et al. 24 did not show such an association. The relation between EGFR expression and histopathological type of CRC in the present study was non-significant, which has also been reported by Brahim et al. 23 The EGFR expression showed a non-significant relationship with pT stage, which agreed with Azevedo et al. 22 and Brahim et al. 23 , but not with Goldstein et al. 14 and Spano et al. 15 . Concerning the N stage in the current study, no relation was observed between EGFR expression and lymph node metastasis. Kapogiannatos et al. 25 found a significant relationship. The current study agreed with Azevedo et al. 22 , who observed higher expression of EGFR in CRC compared to adenoma tissue. According to Liang et al. 26 , overexpression of EGFR is related to bad prognosis and poor survival in many malignant tumours. Additionally, EGFR-positive CRC is associated with an advanced stage of the illness and can indicate a risk of metastatic spread. 27 The Ki-67 LI in current CRC cases ranged from 2% to 93% with a mean of 25%. These results were close to those documented by Wang et al. 28 Petrowsky et al. 29 reported that the mean value of Ki-67 LI for CRC was 50%, and Terzi et al. 30 detected 84%. This variation may be caused by the difficulty and subjectivity involved in quantitatively assessing immunostaining for practical or diagnostic purposes. Ki-67 LI high score ≥25% showed a significant relationship with the tumour grade and the pT stage (p<0.05), and a high Ki-67 LI score was related to HG carcinoma. Likewise, Petrowsky et al. 29 found that HG tumours (G3) had a higher mean Ki-67 LI score. Also, Nayak, et al. 31 observed an increase in Ki-67 LI score in advanced grade. However, Wang et al. 28     current study, which agreed with the results of Wang et al. 28 Ki-67 LI score did not relate significantly with the histopathological type in the current cases. This was in agreement with Ihmann et al. 32 . Also, its expression showed no significant relationship with N stage or LVI in the studied carcinoma cases. This agreed with Wang et al. 28 The most reliable information for a high Ki-67's unfavourable prognostic value was reported by Ignatiadis et al. 33 which was for breast cancer and sarcomas. The relationship between cell proliferation and clinical prognosis in CRCs is ambiguous, if not contradictory. CRC tumours with high proliferative activity have greater responses to conventional radiation and chemotherapy and better disease-free survival. 34 In the current study, colorectal adenoma and CRC cases showed EGFR and Ki-67 expression having a significant association, as cases with high EGFR composite score showed a higher percentage of Ki-67 LI. In concordance, Rego et al. 35 reported that high EGFR expression was related to high cell proliferation and increased Ki-67 LI. Moreover, Chen et al. 36 stated that there was a positive correlation of advanced stage and bad prognosis with the two biomarkers.
In terms of limitations, the current study did not calculate the sample size which could have influenced the power of the study.
To find a potential meaningful cut-off value of Ki-67 expression between colorectal adenoma and carcinoma, more studies with larger sample sizes are require. The relevance of EGFR and Ki-67 biomarkers in the modification of therapy regimens requires further research.

Conclusions
The overexpression of EGFR and Ki-67 in adenomas, particularly those with villous components and HG dysplasia, is a risk factor for the development of CRC. Compared to colorectal adenomas, CRC cases had greater levels of EGFR and Ki-67 LI overexpression. The relationship between EGFR expression and Ki-67 LI in CRC was significant. Increased histological grade and tumour invasion in CRC were significantly correlated with EGFR and Ki-67 overexpression, which foretells a potential metastatic risk.

Conflict of Interest: None.
Source of Funding: None.

Abbreviation
l ErbB: family of proteins contains four receptor tyrosine kinases, structurally related to the epidermal growth factor receptor (EGFR) and ErbB is used as alternative name for (EGFR).
l pT: primary tumour size of pathologic stage according to the American Joint Committee on Cancer (AJCC) system. l SP84, SAB, RM: catalogue code number of the clone, which is used in research, named by manufacturing company.