Controversies on the treatment strategy for rectal submucosal cancer: Case series and review of the literature


  • Shannon Melissa Chan,

    1. Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Philip Wai Yan Chiu,

    Corresponding author
    • Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Sok Fei Hon,

    1. Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
  • Anthony Wing Ip Lo,

    1. Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author
    • The laboratory of AWI Lo is partially supported by an Academic Equipment Grant from the Chinese University of Hong Kong (2011-12).
  • Simon Siu Man Ng

    1. Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong, China
    Search for more papers by this author

Corresponding: Philip Wai Yan Chiu, Department of Surgery, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, China. Email:


Endoscopic submucosal dissection (SD) has emerged as one of the treatment strategies for submucosal rectal cancers. The present study reviewed the clinical outcomes of patients with rectal submucosal cancer treated by ESD. This was a retrospective review of four patients who had rectal tumor treated by ESD from 2010 to 2012 with histopathology showing T1 submucosal adenocarcinoma. The mean age (SD) was 69.5 (7.33) and the male to female ratio was 3:1. There were no post-ESD complications. The mean (SD) size of the tumors was 2.93 (0.87) cm. One patient with deep resection margin involvement received laparoscopic low anterior resection. Another with deep margin involvement of <1 mm refused surgery and was treated by chemoradiotherapy.There was no recurrence in all the cases with a mean follow-up duration of 461.3 (209.0) days. ESD was one of the important treatment strategies for T1sm-s rectal cancer especially when the risk of nodal metastasis was low. ESD spared the patient from colostomy when the T1 cancer was located in the lower third of the rectum. The role of adjuvant and neoadjuvant chemoradiotherapy remains controversial.


The ideal management of early rectal cancer shall maximize preservation of the anal sphincter and reduce the rate of stoma without compromising oncological clearance. According to TNM classification, Tis rectal cancers are those involving the mucosa only, whereas T1 rectal cancers are defined as tumors invading the submucosa. These submucosal tumors can be further classified by quantifying the depth of submucosal invasion and can be divided into SM-slight (SM-s <1000 μm) and SM-massive (SM-m ≥1000 μm).[1, 2] These subclassifications are important, as early rectal cancers carry variable risk of lymph node metastasis and, thus, different treatment modalities may suffice. Advanced rectal cancers located in the middle to lower rectum often require neoadjuvant chemoradiation and surgical resection, which requires a temporary stoma after total mesorectal excision or a permanent stoma for abdominal perineal resection. Advances in endoscopic techniques have made endoscopic submucosal dissection possible for en bloc curative resection of these early rectal cancers. However, the optimal management for these tumors remains controversial. This is a retrospective review of a case series of T1 submucosal rectal cancers and a review of literature.


This was a retrospective review on all patients who had T1 submucosal rectal adenocarcinoma treated with endoscopic submucosal dissection (ESD) carried out at the Prince of Wales Hospital from 1 July 2010 to 31 October 2012. We excluded other rectal pathologies treated by ESD, including adenoma, hyperplastic polyp and carcinoid tumors. Baseline demographics, date and duration of ESD, complications, final pathology, follow-up date and any recurrences were recorded. These data were obtained from the computerized hospital records. Archival histological slides in hematoxylin and eosin stain were reviewed and converted to whole slide images with Leica (Leica Microsystem Ltd, Hong Kong) SCN400 at ×400 magnification. Depth of invasion was assessed by measuring the distance between the deepest smooth muscle cells of the muscularis mucosae and the deepest malignant cells.


A total of four cases with T1 submucosal rectal adenocarcinoma were included (Table 1). The mean age (SD) was 69.5 (7.33) with three male and one female patients. ESD was successfully carried out for treatment of all the lesions with no post-ESD complication. One patient sustained prominent intraoperative bleeding during ESD and was controlled with coagrasper and endoclips. The mean (SD) size of the tumors was 2.93 (0.87) cm. Upon histological examination, one tumor showed lymphovascular permeation, whereas another tumor was found to be poorly differentiated adenocarcinoma. Cases 1 and 2 had submucosal invasion of >1000 μm. Salvage surgery or adjuvant chemoradiotherapy was offered. Case 1 refused both treatments because of old age. Case 2 had concomitant T3N2 cancer of the cecum and transverse colon with extended right hemicolectomy done and was scheduled for chemoradiotherapy. He refused salvage surgery. Case 3 had deep margin involvement and received salvage laparoscopic low anterior resection and total mesorectal excision (TME). Final pathology after surgery confirmed a small remnant foci of dysplasia without invasive adenocarcinoma with 15 benign lymph nodes resected. Case 4 had a <1 mm deep resection margin. Salvage surgery was offered but the patient refused. He received post-procedure adjuvant chemoradiotherapy. There were no recurrences in all cases with a mean follow-up duration of 461.3 (209.0) days.

Table 1. Characteristics of patients with T1 submucosal rectal adenocarcinoma treated with ESD
Case no.SexAge (years)Operative time (min)Post-ESD complicationSize (cm)Poorly differentiatedLymphovascular permeationDepth of submucosal invasion (μm)Deep margin involvementSalvage surgery/oncological treatmentsFU duration (days)Recurrence
  1. ESD, endoscopic submucosal dissection; FU, follow up; Lap TME, laparoscopic total mesorectal excision; NA, not available.
3M62NANone4.2NoNegative2990YesLap TME278No
4M66NANone2.5YesNegative1510Yes <1 mmChemoradiotherapy745No


Surgery for treatment of carcinoma of the rectum is associated with a 10–15% risk of anastomotic leakage. For T1 tumors where the risk of lymph node involvement is low, local resection has been widely accepted as curative treatment. Local resection includes endoscopic resection, transanal resection or transanal endoscopic microsurgery (TEMS). In a retrospective analysis by Kiriyama et al., ESD was shown to have a lower complication rate than laparoscopic-assisted colorectal surgery (LAC), with favorable en bloc and curative resection rates. Diversion stomas were carried out in 93% of patients who underwent LAC for rectal cancers below the peritoneal reflection. Thus, the authors concluded that ESD has advantages for the treatment of early colorectal cancers with minimal risk of nodal metastasis.[3]

Although endoscopic resection achieved better perioperative outcomes and avoided the necessity of colostomy especially for lower rectal tumors, there is still controversy in the adequacy of oncological clearance. Previously, only low-risk malignant polyps were recommended to receive endoscopic resection, and curative resection for these resected lesions was defined by the following criteria: complete resection of the polyp; well- or moderately differentiated histological grade; histological examination of the complete specimen (piecemeal resections were not accepted); and absence of lymphovascular invasion.[4, 5] Haggitt et al. described that polyps affecting levels 1, 2 and 3 present a low risk of developing lymph node metastases (<3%) and are amenable for definitive endoscopic treatment, whereas surgical resection is required for polyps affecting level 4.[6] The Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines in 2010 recommended that the indications for definitive endoscopic treatment include all intramucosal carcinoma or carcinoma with slight submucosal invasion and maximum diameter <2 cm of any macroscopic type.[7] Ueno et al. analyzed 252 patients with early invasive colorectal cancers and found that adverse outcome after local resection was associated with SM invasion of ≥1000 μm, lymphovascular permeation, poorly differentiated adenocarcinoma, signet-ring cell or mucinous carcinoma and grade 2/3 budding at the site of deepest invasion.[8] Kitajima et al. showed that the rate of lymph node metastasis for SM-s tumors was 0% and that for SM-m tumors was 12.5%.[9] With this information, indication for further treatment after local resection needs to be combined with other factors such as other risk factors for lymph node metastasis, the patient's physical status and wishes.[7] Depressed lesions are also at higher risk of nodal metastasis and thus endoscopic resection should be limited to these lesions <1 cm in diameter.[10, 11]

A recent retrospective review by Park et al. showed that ESD and transanal endoscopic microsurgery (TEM) achieved similar en bloc resection rates and complication rates, whereas ESD was associated with a shorter total procedural time and hospital stay.[12] Our center also published a retrospective review comparing ESD with transanal local excision for the management of early rectal neoplasms. The ESD group showed better short-term outcomes with no difference in complication and en bloc resection rates when compared with local transanal excision.[13] With precise localization of tumor margins through image-enhanced endoscopy, in theory, ESD shall achieve better dissection of early rectal lesions when compared with transanal excision. Moreover, TEM is limited by the size and length of the transanal endoscopic operation (TEO) device. Therefore, TEM for low rectal lesions is not possible.

The use of neoadjuvant chemoradiotherapy is well established for the treatment of advanced carcinoma of the rectum.[7] However, the adjunctive use of chemoradiotherapy for the management of submucosal rectal tumors with a high risk of nodal metastases after endoscopic resection remains controversial. As accurate staging of T1sm-s lesions before endoscopic resection is difficult, the role of neoadjuvant chemoradiotherapy is hard to determine. In our case series, all four cases were recognized to have submucosal invasion only upon histopathology of the ESD specimen despite the use of preoperative transrectal ultrasound. Currently, the use of magnifying colonoscopy to identify the different pit patterns (Types I–V)[14] and transrectal ultrasound have aided in predicting submucosal invasion. In a study by Meyenberger et al., although endoscopic ultrasound had 83% sensitivity for determining T-staging, distinguishing between mucosal and submucosal invasion remained difficult.[15] In a doubtful situation, we would recommend carrying out diagnostic ESD to determine deep submucosal invasion and other risk factors for nodal metastasis upon histopathology.

In conclusion, there is a definite role of ESD in managing T1sm-s rectal cancers because the risk of nodal metastasis is low, and ESD has been shown to achieve faster recovery and shorter hospital stay. The role of adjuvant and neoadjuvant chemoradiotherapy remains controversial. Larger scale studies should be done to investigate the outcomes of various management strategies for T1 submucosal rectal carcinoma.

Conflict of Interests

The laboratory of AWI Lo is partially supported by an Academic Equipment Grant from the Chinese University of Hong Kong (2011-12). The other authors declare no conflict of interests for this article.