Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 

 Table of Contents  
Year : 2019  |  Volume : 15  |  Issue : 2  |  Page : 437-441

Usefulness of large-section cytokeratin 20 in the detection of intestinal wall infiltration and mesangial metastasis in patients with middle and lower rectal cancer

1 Department of Gastrointestinal Surgery, Shandong Provincial Hospital, Jinan, China
2 Department of Abdominal Surgery, Cancer Center, Sun Yat-Sen University, Guangzhou, China

Date of Web Publication1-Apr-2019

Correspondence Address:
Prof. Chen-Sheng Li
Department of Gastrointestinal Surgery, Shandong Provincial Hospital Affiliated To Shandong University, 324 Jingwu Road, Jinan, Shandong 250012
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jcrt.JCRT_405_18

Rights and Permissions
 > Abstract 

Objective: The objective of the study was to evaluate the usefulness of large-section cytokeratin 20 (CK20) staining technique in the detection of infiltration on the distal wall and mesangial metastasis in patients with middle and lower rectal cancer.
Materials and Methods: A total of 62 patients with rectal cancer in the middle and lower segment were studied on large slices stained with CK20. Logistic regression was used to analyze the clinicopathologic factors related to distal low and middle rectal cancer metastasis to the mesorectum and rectal wall.
Results: Two types of distal metastasis of the tumor were observed in the rectal wall in 18% (11/62) of the patients: submucosal invasion and muscularis propria invasion. The extent of distal metastasis to the rectal wall was around 0.5–1.0 cm. Four types of distal metastasis occurred in the mesorectum: lymph node invasion, blood and lymphatic vessel invasion, perineural invasion, and isolated neoplastic microfoci. Distal metastasis to the mesorectum was observed in 24% (15/62) of the patients. The extent of metastasis to the mesorectum was around 0.5–4.0 cm. Another three patients with microcapillary invasion in the distal mesorectum were observed by immunohistochemistry, as it was difficult to determine the spread by conventional hematoxylin and eosin staining.
Conclusion: The large-section CK20 staining technique is useful for the detection of infiltration on the distal wall and mesangial metastasis in patients with middle and lower rectal cancer.

Keywords: Cytokeratin 20, distal metastasis, rectal cancer, total mesorectal excision

How to cite this article:
Shan KS, Zhang XP, Wang JS, Guo XB, Shang L, Tian F, Jing CQ, Li LP, Wan DS, Li CS. Usefulness of large-section cytokeratin 20 in the detection of intestinal wall infiltration and mesangial metastasis in patients with middle and lower rectal cancer. J Can Res Ther 2019;15:437-41

How to cite this URL:
Shan KS, Zhang XP, Wang JS, Guo XB, Shang L, Tian F, Jing CQ, Li LP, Wan DS, Li CS. Usefulness of large-section cytokeratin 20 in the detection of intestinal wall infiltration and mesangial metastasis in patients with middle and lower rectal cancer. J Can Res Ther [serial online] 2019 [cited 2022 May 21];15:437-41. Available from: https://www.cancerjournal.net/text.asp?2019/15/2/437/255095

 > Introduction Top

Heald first proposed the concept of total mesorectal excision (TME) in 1982. At present, some surgeons still perform conventional surgery to treat middle and lower rectal cancer. However, surgeries are not usually performed based on the TME (total mesorectal excision) principle as surgeons lack a deeper understanding of the significance and necessity of TME; hence, comprehensive and holistic research on distal rectal cancer metastasis is of great significance. This study aimed to evaluate the usefulness of large-section cytokeratin 20 (CK20) staining technique in the detection of intestinal wall infiltration and mesangial metastasis in patients with middle and lower rectal cancer.

 > Materials and Methods Top


A total of 62 patients with rectal cancer in the middle and lower segment were included. Thirty-four specimens from patients with low and middle rectal cancer were collected between August 2004 and December 2005 in Sun Yat-Sen University Cancer Center. Twenty-eight specimens from patients with low and middle rectal cancer were collected between October 2006 and October 2007 in Shandong Provincial Hospital of Shandong University. Patients (1) whose diagnosis was confirmed by pathology, (2) who underwent removal of a 5.0-cm distal tumor of the intestinal wall or mesentery, (3) who did not receive preoperative radiotherapy and chemotherapy, and (4) who developed a tumor within 10 cm from the anal verge were included. A total of 62 patients were analyzed, including 30 men and 32 women. Two primary types of tumors were noted: mass type (22 patients) and ulcer type (40 patients). With regard to pathological type, 54 patients had adenocarcinoma and 8 patients had signet-ring cell carcinoma. With regard to pathological grade, 2 patients had Grade I, 46 had Grade II, and 14 had Grade III. About 32 patients had lymph node metastasis, whereas 4 had distant liver metastasis. Blood carcinoembryonic antigen (CEA) increased in 22 patients, whereas blood carbohydrate antigen 19-9 (CA 19-9) increased in 8 patients. A total of 15 patients had tumor node metastasis (TNM) I stage, 16 had TNM II stage, 27 had TNM III stage, and 4 had TNM IV stage.


Specimen production

All patients underwent surgery according to the principles of TME. A total of 26 patients underwent transabdominal perineal resection, whereas 36 underwent low anterior resection. After the specimen was isolated, the unfolded intestine and mesangial membrane were nailed on a wooden board based on the shape of the bowel, and the wooden board was immersed in a 10% formalin solution for 1 week or more.

Hematoxylin and eosin staining

Tissue pieces about 0.5-cm wide were cut horizontally on fixed specimens. This tissue block contains the intestinal wall and mesangial tissue. Dehydration, transparency, and embedding were routinely performed. The slices were sectioned on a microtome, and the tissue thickness was about 8 μm. After hematoxylin and eosin (HE) staining, large sections were transparently sealed.

Cytokeratin 20 staining based on the EnVision two-step method

Mouse antihuman CK20 monoclonal antibody was purchased from Dako Company (USA). A two-step universal immunohistochemical detection kit (ChemMate™ EnVision) was purchased from Gene Tech Corp (Shanghai). The working concentration was 1:20, and the method was performed in accordance with the manufacturer's instructions.

Analysis of the results

Under the ordinary optical microscope, two experienced pathologists read the film, and the CK20 staining result was judged by observing the brownish yellow precipitate formed by the oxidized diaminobenzidine. The CK20 staining was located in the cytoplasm of the rectal mucosal epithelial cells. According to the principles discussed in the literature,[1] if the circumferential margin of the mesorectum is 1 mm away from the surgical margin, it was judged as circumferential margin carcinoma infiltration.


Postoperatively, the patients were treated according to the guidelines of the American National Comprehensive Cancer Network (NCCN) rectal cancer, and follow-ups were conducted using outpatient services, questionnaires, letters, and telephones. Blood tumor markers (CEA and CA19-9) were monitored every 3 months, abdominal ultrasound computed tomography (CT) and chest radiograph were performed every 6 months, and enteroscopy was performed annually to detect local recurrence and distant metastasis.

Statistical processing

SPSS 22.0 (Statistical Product and Service Solutions, SPSS Inc., IBM, Current versions (2015)) was used to establish a database. Logistic regression was used to analyze the relationship between distal intestinal wall infiltration and mesangial metastasis and clinical and pathological parameters. Chi-square test was used to compare the HE results and CK20 staining results.

 > Results Top

Hematoxylin and eosin staining results

Of 62 patients, 11 had distal intestinal wall infiltration and 12 had distal mesangial metastasis. Of these, 8 patients had both distal intestinal wall infiltration and distal mesangial metastasis. The distal intestine wall infiltration distance was <0.5 cm in 9 patients and 1.0 cm in <2 patients. The distance of distal mesangial transfer was within 0.5 cm in 5 patients, 1.0 cm in 4 patients, 2.0 cm in 1 patient, 3.0 cm in 1 patient, and 4.0 cm in 1 patient. Intestinal wall infiltration is a direct infiltration of the tumor (formation of a cancerous nest in the submucosal or intermuscular). Distal mesangial metastasis takes the form of (1) lymph node metastasis (tumor cells located in the lymph nodes) (2), vascular metastasis (tumor cells forming a tumor embolus or infiltrating the lumen wall within the vessel) (3), peripheral nerve transfer (tumor cells infiltrating the nerve sheath or nerve bundles within the sheath), and (4) isolated foci (tumor cells forming a nonlymphoid-like cancer nest in the mesangium distant from the primary tumor) [Figure 1].
Figure 1: Mesangial lymphatic metastasis of rectal cancer EnVision ×400

Click here to view

Cytokeratin 20 staining results

The positive results observed by HE staining were clearly expressed in CK20 staining [Figure 2]. CK20 staining was localized in the cytoplasm of rectal mucosal epithelial cells. In addition, three microvessel metastases were also observed in this study. As it was extremely difficult to detect the lesions by routine HE staining alone, HE staining was used together with CK20 staining [Figure 3]. Hence, three patients with distal mesenteric cancer were detected with CK20 staining, whereas 15 had distal mesangial metastasis. Six patients had distal mesangial transfer within 0.5 cm, four within 1.0 cm, three within 2.0 cm, one within 3.0 cm, and one within 4.0 cm. Eight patients had lymph node metastases. A distal (intestinal wall infiltration or mesangial) metastasis occurred in one patient with TNM I, two with TNM II, eight with TNM III, and four with TNM IV. There was no significant difference between HE and CK20 staining results [Table 1] χ2 = 0.426, P = 0.514].
Figure 2: Mesangial nerve transfer of rectal cancer, EnVision ×400

Click here to view
Figure 3: Mesangial venous transfer of rectal cancer, EnVision ×400

Click here to view
Table 1: X2 =0.426, P=0.514, no statistically significant difference between hematoxylin and eosin and cytokeratin 20 staining results

Click here to view

Relationship between distal intestinal wall infiltration and mesangial metastasis and clinicopathologic factors

Logistic regression analysis showed that the clinicopathologic factors include sex, age, maximum diameter of tumor, pathological type, pathological grade, distal intestine wall infiltration and mesangial metastasis of rectal cancer, lymph node metastasis and TNM stage, blood CEA, and blood CA 19-9. The following factors were used in the univariate analysis: (1) blood CEA, (2) lymph node metastasis, and (3) TNM staging of three factors [Table 2]. The factor used in the multivariate analysis was TNM staging (Wald = 4.758, P = 0.029) [Table 3]. Univariate analysis showed that serum CEA, lymph node invasion, and TNM staging were associated with distal proliferation of rectal and rectal wall rectal cancer [Table 2]. Multivariate analysis showed that TNM staging was an independent influencing factor (Wald = 4.758, P = 0.029)[Table 3].
Table 2: Univariate analysis of factors related to intestinal infiltration and mesangial metastasis

Click here to view
Table 3: Multivariate analysis of factors related to intestinal infiltration and mesangial metastasis

Click here to view


Postoperative follow-up time ranged from 9 to 60 months, with a median time of 47 months. Three patients were lost to follow-up at 23, 45, and 58 months after surgery. The follow-up rate was 95.1%. The postoperative local recurrence rate was 11.2 (7/62), the time of recurrence was 3–42 months, and the median time was 17 months. The distant metastasis rate was 24.1% (15/62). The time of distant metastasis was 2–58 months after surgery. The median time was 27 months; the 5-year survival rate was 58% (36/62); the time of death was 10–59 months postoperatively, and the median time was 37 months.

 > Discussion Top

The performance of large-section CK20 staining at home and abroad to detect distal intestinal wall infiltration and mesangial metastasis has not been reported yet. In this study, 0.5-cm tissue pieces were continuously cut from the distal tumor specimens, to completely observe the status of the distal intestinal wall and mesentery, and may indicate metastasis and development of distal lesions at a certain distance from the main tumor, compared with conventional small slice HE staining; this method is more representative, continuous, and integrated and prevents missed diagnosis of cancer foci. The NCCN Rectal Cancer Treatment Regulations (2006) suggest that the examination of enough number of lymph nodes can ensure the accuracy of pathological staging after surgery, laying the foundation for a reasonable choice of the next treatment plan and improving the curative effect. In this study, large-section CK20 staining technique was used. Lymph nodes with diameter <2 mm were observed and the number of lymph node biopsies was increased. The average number of lymph node seizures was 30.5, which could not be achieved by routine medical examination. Keratin is one of the components of the intermediate filaments of the cytoskeletal system. CK20 is widely expressed in rectal cancer foci and is not expressed in normal mesangial tissues. Many scholars believe that CK20 has good sensitivity and specificity; therefore, CK20 was used as an index of immunohistochemistry.[2] The difference between CK20 staining results and HE staining results was not considered significant. This finding may be related to the small sample size. Hence, further studies must be conducted using a larger sample size.

This study showed that the incidence of distal wall infiltration in rectal cancer was 18% (11/62). The distance between the first large slice and the lowest edge of the main tumor was 0.5 cm. It was impossible to detect 0.5 cm from the lower edge of the tumor to the distal end. However, observation of the intestine every 0.5 cm can clearly determine whether there is tumor infiltration at a distance away from the main tumor. Currently, there is less probability of intramural hopping metastasis in rectal cancer. Therefore, this method is reliable for judging the distance of infiltration in the distal wall of rectal cancer. The longest infiltration distance of rectal cancer was 1 cm. No infiltration in the distal wall of 1.5 cm or more was found. In this study, it was considered that the distal intestine resection was more than 1.5 cm. Rullier et al.[3] thought that the distal bowel resection can be 1–2 cm, which is consistent with the results of this study. This study showed that the incidence of distal mesangial metastasis in rectal cancer was 24% (15/62), and the longest metastatic distance was 4 cm, which further supported the view that the distal mesorectum must be completely resected.[4] Of the 15 patients with distal mesangial metastases, 7 had nonlymphatic metastases, indicating that they could not emphasize the role of lymph node metastasis. Mesangial metastasis has the forms of lymph node metastasis, vascular metastasis, perineural nerve metastasis, and isolated cancer foci. In patients with rectal cancer, there may be nests of cancer cells in the mesorectum even without lymph node metastasis. If the mesorectal excision does not completely result in its residual, it becomes an important source of postoperative recurrence. This study found three mesangial micrometastatic tumors with irregular morphology. It was defined as VI (vascular invasion) according to the NCCN rectal cancer treatment protocol (2006). Although these three patients did not have lymph node metastasis, they were considered at higher risk of vascular invasion, and adjuvant chemotherapy should be performed. Ratto[5] defined cancer nodules as nonlymphoid-like adenocarcinoma nests in the mesorectum distant from the primary tumor; approximately 44% (34/77) of colorectal cancers were found to have disseminated tumor nodules. The 5-year survival rates of micronodule dissemination and no-tumor micronodule dissemination were 43% and 63%, respectively (P = 0.016). Further studies have shown that lymphatic metastasis, peripheral nerve metastases, and isolated foci were independent factors that influence prognosis. The inclusion of various disseminated forms of cancer nodules into the TNM stage may be more helpful in guiding the order of rectal cancer treatment and predicting prognosis. Logistic regression univariate analysis showed that increased blood CEA levels, lymph node metastasis, and TNM stage were associated with distal rectal cancer proliferation, and these indicators were negatively correlated with the prognosis. This study suggests that follow-up of patients with elevated blood CEA levels and lymph node metastases should be enhanced. Arumugam PJ et al.[6] reported that circumferential marginal involvement (CMI) is an independent factor affecting local recurrence; in his study, the presence of CMI was confirmed in eight specimens, leaving hidden risks for residual local pelvic cancer and postoperative recurrence. In our study, surgery was performed in strict accordance with the principle of TME; no CMI-positive pathological specimens were observed, and the effect of circumferential marginal infiltration on local recurrence and prognosis was avoided [Figure 4]. After entering the logistic regression multivariate analysis, only the TNM staging was retained. This may be due to the fact that the TNM staging replaced other factors. In this study, four had TNM IV, had preoperative CT scan but failed to detect lesions, and had postoperative pathology, which confirmed the presence of liver micrometastases. All four patients with liver metastases exhibited distal proliferation, indicating that tumor dissemination in TNM IV patients was multidirectional. For rectal cancer with liver metastases, Stewart et al.[7] suggested that patients with hepatic metastases do not develop liver cirrhosis, liver regeneration reserves are functioning well, and multiple liver resections can extend lifespan.
Figure 4: Circumferential marginal involvement-negative pathological specimens

Click here to view

This study conclude that as the current resection cannot ensure the removal of the potential cancer from the mesangial tissues, abdominal perineal resection should be performed, but not due to liver metastases and contempt for the treatment of the primary lesion.

This study found that patients with poor staging have a higher probability of distant metastases. In 6.7% (1/15) of TNM I patients, distant metastases occurred in 12.5% (2/16) of TNM II patients, 29.6% (8/27) of TNM III patients, and 100% of TNM IV patients; patients with distant metastases may have a poor prognosis. In our follow-up, 15 rectal cancer patients had distant metastasis (positive rate: 24%), and distant metastatic rectal cancer patients had a 5-year survival rate of 26.7% (4/15), which was significantly lower than 53.2% of the negative group (25/47). Kaplan–Meier survival analysis showed that distant rectal cancer metastasis was closely related to survival time [P = 0.013, [Figure 5]. The relationship between distal metastasis and prognosis in this group of patients was further confirmed by follow-up.
Figure 5: Kaplan–Meier estimates of overall survival among rectal cancer patients with intestinal infiltration and mesangial metastasis

Click here to view

 > Conclusion Top

The large-section CK20 staining technique is useful for the detection of infiltration on the distal wall and mesangial metastasis in patients with middle and lower rectal cancer

Financial support and sponsorship

This work was supported in part by grants from the Young and Middle-Aged Scientists Research Award Fund of Shandong Province (BS2010YY060), Technology Development Project of Shandong Province (2014GSF118134), Shandong Key RandD Program (2017GSF221018), National Natural Science Foundation of China (81672379) and Natural Science Foundation of Shandong Province of China (ZR2016HM16).

Conflicts of interest

There are no conflicts of interest.

 > References Top

Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet 1986;2:996-9.  Back to cited text no. 1
Wang Z, Zhou ZG, Wang C, Zheng XL, Wang R, Li FY, et al. Regional micrometastasis of low rectal cancer in mesorectum: A study utilizing HE stain on whole-mount section and ISH analyses on tissue microarray. Cancer Invest 2006;24:374-81.  Back to cited text no. 2
Rullier E, Laurent C, Bretagnol F, Rullier A, Vendrely V, Zerbib F, et al. Sphincter-saving resection for all rectal carcinomas: The end of the 2-cm distal rule. Ann Surg 2005;241:465-9.  Back to cited text no. 3
Zheng YC, Zhou ZG, Li L, Lei WZ, Deng YL, Chen DY, et al. Distribution and patterns of lymph nodes metastases and micrometastases in the mesorectum of rectal cancer. J Surg Oncol 2007;96:213-9.  Back to cited text no. 4
Ratto C, Ricci R, Rossi C, Morelli U, Vecchio FM, Doglietto GB, et al. Mesorectal microfoci adversely affect the prognosis of patients with rectal cancer. Dis Colon Rectum 2002;45:733-42.  Back to cited text no. 5
Arumugam PJ, Vivek V, Beynon J. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer (Br J Surg 2002; 89: 327-34). Br J Surg 2002;89:1067.  Back to cited text no. 6
Stewart GD, O'Súilleabháin CB, Madhavan KK, Wigmore SJ, Parks RW, Garden OJ, et al. The extent of resection influences outcome following hepatectomy for colorectal liver metastases. Eur J Surg Oncol 2004;30:370-6.  Back to cited text no. 7


  [Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]

  [Table 1], [Table 2], [Table 3]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  >Abstract>Introduction>Materials and Me...>Results>Discussion>Conclusion>Article Figures>Article Tables
  In this article

 Article Access Statistics
    PDF Downloaded48    
    Comments [Add]    

Recommend this journal