1. Treatment of Early Malignant Rectal Polyp
Dr KP Tsui
Department of Surgery
Tseung Kwan O Hospital

2. Malignant Rectal Polyp
Polyps with cancer cells invading the muscularis mucosa
Invasion limited to submucosa
T1 lesion

3. Incidence of malignant colorectal polyps as a proportion of all adenomas removed varies between 2.6 and 9.7%.
Average 4.7%
Sobin L, Wittekind C (eds). TNM classification of Malignant Tumours (6th Edition). Wiler-Liss: New York, 2002.

4. Size most important determinant factor determining risk of malignant transformation within a polyp
> 1 cm: 38.5%
> 42 mm: 78.9%
Tytherleigh et al. BJS 2008;95:

5. Villous adenomas have highest risk of malignancy at 29.8%
Tubular adenomas have lowest at 3.9%
Tytherleigh et al. BJS 2008;95:

6. Haggitt Classification

7. Kikuchi Classification of Adenocarcinoma in Sessile Polyps
Haggitt level 1,2,3 = Kikuchi Sm1
level = Sm1, Sm2 or Sm3

8. Treatment
Staging
Histological Assessment

9. Clinical Scenario 1
Colonoscopy: 2 cm rectal polyp (5 cm from anal verge)
Biopsy: adenocarcinoma

10. Endorectal ultrasound
Best method to differentiate between T1 and T2 lesion
T stage N stage
Accuracy: 90 % Accuracy: 80%
Sensitivity : 85% Sensitivity: 70%
Specificity: 95% Specificity: 80%
Bretagnol et al. Dis Colon Rectum 2007;50:

11. Can assess residual tumor after polypectomy
Follow up after local excision
Hernandez De Anda et al. Dis Colon Rectum 2004; 47: 818–824
Sensitivity/specificity/overall accuracy rates for detection of slight submucosal invasion, massive
submucosal invasion, and muscularis propria invasion were 99%/
74%/96%, 98%/88%/97%, and 97%/93%/96%, respectively
Limitations: peritumoral fibrosis and inflammatory tissue, operator dependent, tumor height, tumour stenosis, effect of pre op radiotherapy or haemorrhage in bowel wall after bx
-the substitution of the typical five-layer structure of the rectal wall by a mixed echoic pattern involving a segment of the rectal wall. The staples after an anastomosis
are usually visualized as a circumferential line of small, bright echoes without a shadow. Local recurrence often appears as a well-defined hypoechoic homogenous mass involving the rectal wall or as a hypoechoic extrarectal node, persisting in at least two consecutive exams (Figs. 1 and 2). Recurrence was confirmed by endoscopic biopsy, ultrasound-guided biopsy, or CT scan-guided biopsy.

12. Limitations Operator dependent Upper rectal lesions Tumor stenosis
Peritumoral fibrosis and inflammatory tissue
Effect of radiotherapy or hemorrhage after biopsy

13. Pelvic MRI Overall T stage accuracy 59-95% T1,2 lesion (vs ERUS)
- Similar sensitivities
- Lower specificity (69%)
N stage
- Comparable to EUS
Can evaluate entire pelvis
Bretagnol et al. Dis Colon Rectum 2007;50:
Tytherleigh et al. BJS 2008;95:
Endorectal and phased-array coils--> better resolution and accuracy for predicting tumour stage
New techniques accuracy ~ 90%
A meta analysis: T1, 2 lesion: ERUS, MRI smilar sensitivities, ERUS 86% specificity sig higher MRI69%

14. CT abdomen + pelvis Distant metastases
Low accuracy for T staging, 52 – 94% and N stage, 54-70%
Alexandre Jin Bok Audi Chang et al. Journal of Surgical Education; Vol 65: Number 1
Bretagnol et al. Dis Colon Rectum 2007;50:

15. PET Limited role for local and regional staging
Sensitivities for lymph node metastases 22-29%
Abdel-Nabi H, Doerr RJ, Lamonica DM, et al. Radiology. 1998;206:
Functional rather than anatomic information

16. Surgical Options
Local excision vs Radical Surgery Park’s per anal excision Abominoperineal resection TEM Total Mesorectal Excision Anterior resection

17. Local Excision Opportunity of cure with less detriment
Sphincter preservation
Less morbidity and mortality
Less sexual or urinary dysfunction
Anterior resection and APR: 30 day mortality rate less than 7%, morbidity 35%, and poor functional outcome

18. Park’s per anal excision
Aid of anal retractors
6-10 cm of anal margin
Full thickness excision
At least 1 cm margin
Defect usually closed with absorbable sutures
Snare polypectomy or endoscopic mucosal resection is used to treat polyps that are thought to be benign.
A polyp that does not ‘lift’ on submucosal infiltration should be regarded as malignant
???peritoneal reflection

19. Transanal endoscopic microsurgery
Rectoscope
Usually below peritoneal reflection
Full thickness excision
Excision margin of 1 cm
Difficult for lesions within 6 cm
Lesions within 6 cm of anal verge best dealt with by Park’s since difficult to maintain CO2 seal needed for TEMS in anal canal

21. Long-handled transanal endoscopic microsurgery instrument

22. Complications Overall rate 6-31% Postoperative hemorrhage 1-13%
Perforation 0-9%
Suture line dehiscence
Perirectal abscess
Rectal stenoses
Hiroko Kunitake, et al. Perm J 2012 Spring;16(2):45-50

23. Local Excision Vs Radical Surgery

24. Generally accepted that local excision, by either endoscopic polypectomy or transanal surgery is adequate treatment for low risk ERC
Tytherleigh et al. BJS 2008;95:

25. Histopathological Features
Low risk early rectal cancer
High risk early rectal cancer
Well or moderately differentiated
Poorly differentiated
No vascular or lymphatic invasion
Vascular or lymphatic invasion
Hagitt 1-3
Kikuchi Sm 1 and ?Sm2
Kikuchi Sm3 and ?Sm2
Positive resection margin
Relative factors
Absence of lymphoid infiltration
Tumor budding
Poor demarcation at invasive front
Cribiform type structural atypia
Position in distal third of rectum

26. Poorly differentiated carcinoma: 50% risk of lymph node metastasis
Coverlizza S, Risio M, Ferrari A, Fenoglio-Preiser CM, Rossini FP. Cancer 1989;64:
Lymphovascular invasion, sm3 invasion, undifferentiated carcinomas have significant risks of LN metastases.
Nascimbeni et al. Dis Colon Rectum 2002;45:

27. Rate of lymph node metastasis Sm1 1-3% Sm2 8% Sm3 23%
Des.
Depth of invasion was found to be best estimate of the probability of regional LN metastasis
Bretagnol et al. Dis Colon Rectum 2007;50:
Rate of lymph node metastasis
Sm %
Sm2 8%
Sm3 23%
Nascimbeni et al. Dis Colon Rectum 2002;45:
Nascimbeni et al. [15] studied histological specimens retrospectively from 353 patients undergoing colorectal resection for sessile T1 lesions.( )
The authors reported that the depth of invasion into the lower third of the submucosa (classified as ‘sm3’), the presence of lymphovascular invasion, and lesions in the
lower third of the rectum were significant predictors of lymph node metastasis
*(St Marks) S. Rasheed et al. Colorectal Disease, 10, 231–237. ( )total 313 T1 in 55 (18.2%) and T2 in 248 (81.2%). The node positive group did however contain a significantly higher number of patients with poorly differentiated tumours (P = 0.001) and patients with evidence of extramural vascular invasion (P = 0.002). Finally there was no significant difference in the number of patients with sm1, sm2, sm3(12-14%), or T2 (~19%)
tumour depths within the lymph node positive and negative groups
Limitation: small sample size of T1, retrospective, surgeon choose for radical excision rather than local excision

28. Optimal choice of surgery
The role of local excision as a curative procedure has been questioned due to inferior outcome in some long term follow up series.
Alexandre Jin Bok Audi, MD, et al. Journal of Surgical Education; Vol 65: Number 1 (2008)

29. NASCIMBENI ET AL Dis Colon Rectum 2004; 47: 1773–1779
We compared 70 patients who underwent local excision with 74 patients who underwent oncologic resection. Among patients with lesions in the middle or lower third of the rectum, 1) the five-year and ten-year outcomes were significantly better for overall survival and cancer-free survival in the oncologic resection group, but there were no significant differences in local recurrence or distant metastasis; 2) the multivariate risk factors for long-term, cancer-free survival were invasion into the lower third of the submucosa, local excision, and older than aged 68 years; and 3) for lesions with invasion into the lower third of the submucosa, the oncologic resection group had lower rates of distant metastasis and better survival. Among patients with lesions
in the lower third of the rectum, 1) the five-year and tenyear outcomes showed no significant differences in survival, local recurrence, or distant metastasis between the two groups; and 2) for lesions with invasion into the lower third of the submucosa, the oncologic resection group showed a trend of improved survival, which was not statistically significant, possibly because of low statistical power from the small sample size
Alexandre Jin Bok Audi, MD, et al. Journal of Surgical Education; Vol 65: Number 1 (2008)

30. Most literature data are based on case reports or small series with no standard criteria for patient selection

31. Adjuvant chemoradiotherapy
May be beneficial
Recommended for high risk T1 lesions, assuming further surgery is not an option
Tytherleigh et al. BJS 2008;95:
Neoadjuvant chemoradiotherapy: overtreatment

32. Bretagnol et al. Dis Colon Rectum 2007; 50:523-533

33. Limitations Most retrospective studies Lack of controlled data
No defined protocol for chemotherapy

34. Salvage surgery
Between 56 and 100% of recurrence suitable for salvage surgery
May not offer same outcomes as initial treatment
Should not be delayed in case of recurrence
Tytherleigh et al. BJS 2008;95:
Disease free survival rates 20-35%

35. Clinical Scenario 2 Colonoscopic polypectomy of rectal polyp
Pathology: adenocarcinoma

36. Histological assessment not adequate
Pathology
No High Risks Features
Haggitt level 1,2,3
Kikuchi Sm1
High Risks Features
Sm3 (Sm2)
Grade
lymphovascular
ERUS MRI CT
LN-
LN+
Margin involvement
Yes
Histological assessment not adequate No
Local Excision No
Yes
Follow up
Radical Surgery
High Risks Features

37. Follow up FU with ERUS: Frequency: subject to debate One study showed
Digital rectal exam + Endoscopy + CEA
First 3 years: every 3 months
Next 2 years: every 6 months
Then annually
Endorectal ultrasound should be performed at every outpatient session
Mellgren et al. Dis Colon Rectum 2000; 43: 1064–1071
NCCN guideline
National comprehensive Cancer Network guidelines recommended for T1 with local excision
Q3m x 2yr
Q6m x 3 yr Q1
FU with ERUS:
Frequency: subject to debate
One study showed
More isolated local recurrence in the follow-up ERUS group underwent Salvage Surgery (44% vs 23 %), but the differences were not significant
Hernandez De Anda et al. Dis Colon Rectum 2004; 47: 818–824

38. Summary Local excision Recommended for low risk T1 Sm1 lesion
Radical surgery
For high risk T1 lesion
Adjuvant therapy if further surgery is not an option

39. Recurrence
Diagnose early for salvage surgery
Follow up
Endoscopic surveillance of rectum and scar