| Japanese Journal of Clinical Oncology | Pages |
Outcome of Local Excision for Locally Invasive Rectal Carcinomas with Special Reference to Histological Features at the Invasive Margin
Introduction
Patients' profiles and Methods
Results
Discussion
References
Outcome of Local Excision for Locally Invasive Rectal Carcinomas with Special Reference to Histological Features at the Invasive Margin
Methods: Demographic and pathological characteristics of eight patients with locally invasive tumors undergoing initially local excision were reviewed with reference to histological features at the invasive margin.
Results: All the tumors were well differentiated adenocarcinomas. In all but two tumors, the invasion was limited within the proper muscle layer. Radiation therapy was given preoperatively in one patient and postoperatively in two patients. Additional bowel resection was not attempted in these three cases. Among the remaining five patients, two received additional bowel resection with lymph node dissection. No lymph node metastasis was observed in these two patients. During the average follow-up period of 55 months, three patients had regional lymph node metastases at 7, 36 and 72 months, respectively. Another patient had regional lymph node and distant metastases at 5 months. Three out of five patients with moderate to severe grade of dedifferentiated histology at the invasive margin (H-inv) had regional lymph node metastases. On the other hand, one out of three patients with mild H-inv had lymph node metastases.
Conclusions: H-inv may be useful as a clinical predictor of lymph node metastasis. However, more experience is needed to confirm the usefulness of H-inv in selecting invasive rectal cancer patients in whom local excision is safe and appropriate.
INTRODUCTION
Local excision has been widely considered as a useful procedure for the treatment of early (intramucosal or submucosally invasive) carcinomas of the rectum (1-4). However, if cancer cells invade into or beyond the proper muscle layer, the risk of lymph node metastases is over 20% (5); therefore, local excision is usually regarded as a dangerous option for the treatment of locally invasive rectal carcinomas. Lower anterior resection or abdominoperineal excision with regional lymph node dissection is the standard procedure to eradicate malignant cells. However, the physical and psychological burden on patients with permanent colostomy after abdominoperineal excision and postoperative urinary and/or male sexual dysfunction have been serious problems that deteriorate patients' quality of life. In the last 10 years, several researchers have reported promising results that local excision with or without postoperative chemo-radiation therapy is an alternative approach for sphincter preservation in patients with locally invasive rectal carcinomas (6-12). Here we review our preliminary experience of local excision for locally invasive rectal carcinomas and discuss the implications of radiation therapy and histological features at the invasive margin.
PATIENTS' PROFILES AND METHODS
Figure 1. High-power view of the invasive margin of the tumor showing mild degree of dedifferentiated cancer cell nests (arrow). (Hematoxylin-eosin, ×25.) From 1967 through 1995, eight patients were treated initially with local excision in the Department of Surgical Oncology at the University of Tokyo. The indications of local excision for locally invasive rectal carcinomas were as follows: (i) localized polypoid or ulcerative lesion located in the lower rectum or anal canal; (ii) maximum diameter <30 mm; (iii) well differentiated histology; or (iv) poor medical condition or advanced age. Furthermore, from the mid-1980s radiation therapy has been combined with local excision for locally invasive lower rectal carcinomas in some leading western hospitals (6), so we started to adopt this modality in our patients. Four patients were men and four were women. The mean age was 63 years with the range 44-88 years. Tumor location, size, surgical procedure, Dukes' stage and adjuvant therapy (radiation, chemotherapy) were reviewed in the patients' clinical records. Patients' characteristics were described according to the General Rules of the Japanese Research Society for Cancer of the Colon and Rectum (JRSCCR) (13). Tumors were evaluated for predominant grade, lymphatic vessel invasion, blood vessel invasion, margins and histology at the invasive margin (H-inv). H-inv was graded quantitatively according to the classification proposed by Ono et al. (14). In brief, the number of `dedifferentiation' units as a single cancer cell or a solitary trabecular form was counted under a light microscope along the entire invasive margin and H-inv was graded into four levels: none (0 unit), mild (1-20 units) (Fig. 1), moderate (21-50 units) and severe (>50 units) (Fig. 2). Figure 2. High-power view of the invasive margin of the tumor showing severe degree of dedifferentiated cancer cell nests. (Hematoxylin-eosin, ×25.) Additional bowel resection was considered when pathological examination of resected specimens revealed unfavorable findings such as positive cut end or vascular invasion. Patients were followed by physical examination, carcinoembryonic antigen (CEA) monitoring, chest radiography, abdominal ultrasonography and pelvic computed tomography (CT).
RESULTS
As shown in Table 1, all the tumors were located in the lower rectum or anal canal.
Table 1.
| Patient | Age (years) | Sex | Location | Shape | Size (mm) | Depth | Dukes | Initial treatment | Additional therapy |
| 1 | 81 | M | P | Type 1 | 10 | mp | A | TAR | - |
| 2 | 44 | M | Rb | Type 2 | 35 | mp | A | Y-M | - |
| 3 | 52 | F | Rb | Type 1 | 55 | mp | A | TSR | rad (post-op) |
| 4 | 54 | F | Rb | Type 2 | 25 | mp | A | TSR | rad (post-op) |
| 5 | 88 | F | RbP | Type 1 | 17 | mp | A | TAR | - |
| 6 | 64 | M | Rb | Type 2 | 16 | A1 -> mp* | B -> A* | TAR | rad (pre-op) |
| 7 | 60 | M | Rb | Type 1 | 16 | mp | A | TAR | LAR |
| 8 | 70 | F | P | Type 2 | 10 | a1 | B | TAR | APR |
Table 2.
| Patient | Grade | ly | v | Margin | H-inv | Recurrence | Prognosis |
| 1 | Well | + | + | - | Mild | 5 m; inguinal nodes, lung | 9 m DOD |
| 2 | Well | - | + | - | Mild | 8 y 8 m NED | |
| 3 | Well | - | - | - | Moderate | 7 y 5 m NED | |
| 4 | Well | - | - | - | Severe | 6 y; intra-pelvic | 8 y 2 m DOD |
| 5 | Well | - | + | + | Mild | 2 y 9 m DOC | |
| 6 | Well | - | + | - | Moderate | 3 y; intra-pelvic | 3 y 4 m LWD |
| 7 | Well | + | + | + | Moderate | 1 y 8 m NED | |
| 8 | Well | + | - | + | Moderate | 7 m; inguinal nodes | 3 y 10 m NED |
Operative procedures were trans-sphincteric in one, trans-sacral in two and trans-anal in the remaining five patients. In all but two tumors, the invasion was limited within the proper muscle layer. The mean size of the tumors was 24.9 mm (range 10-55 mm). All the tumors were well differentiated adenocarcinomas. Clinical stages were Dukes' A in six and Dukes' B in two patients. Radiation therapy was given preoperatively in one and postoperatively in two patients. The total radiation dose was 50 Gy in 25 fractions. Additional bowel resection was not attempted in these three cases. Among the remaining five patients, two received additional bowel resection (one low anterior resection and the other abdominoperineal excision) with regional lymph node dissection due to positive cut end or lymph vessel invasion. Lymph node metastasis was not observed in the resected specimens.
During the average follow-up period of 55 months (range 9-104 months), four patients had recurrence, two were dead with disease and another one was living with disease (Table 3). In patient 1, a 10 mm sized Type 1 tumor of the anal canal was resected peranally. Pathological examination revealed cancerous invasion into the proper muscle layer; however, no additional treatment was attempted owing to his advanced age. He died 9 months postoperatively of generalized metastases in inguinal lymph nodes, lung and skin. In patient 4, treated with transsacral resection (Fig. 3) and postoperative chemoradiation therapy, the serum CEA level began to increase at 72 months after the initial operation. Thereafter, the patient complained of lumbago and ischiatics. Magnetic resonance imaging performed 90 months postoperatively revealed unresectable intrapelvic lymph node metastasis. Chemotherapy was started using 5-Fu and cisplatin intravenously with no improvement and the patient died of disseminated intravascular coagulopathy. In patient 6, treated with preoperative radiation therapy and transanal resection for A1 lower rectal cancer, unresectable intrapelvic recurrence occurred 36 months postoperatively. In patient 8, treated with trans-anal resection and additional abdominoperineal excision for a1 cancer, inguinal lymph node metastases occurred 7 months postoperatively. Groin dissection and radiation therapy were given and the patient was alive and well 4 years postoperatively.
Figure 3. Low-power view of the resected specimen showing a 25 mm sized ulcerative tumor invading the proper muscle layer. Three out of five patients (60%) with moderate to severe grade of H-inv had local or distant metastases. In patient 4, H-inv was most prominent in our series (Fig. 2). On the other hand, one out of three patients with mild H-inv had recurrence and was dead of disease.
DISCUSSION
The indications for local excision of early colorectal cancers have been advocated by several researchers in the last 20 years and the following criteria have been reached so far (1-3,15). If locally excised specimens have well differentiated histology, no lymphatic or vessel invasion and negative cut end, local excision is appropriate enough for the treatment of early invasive carcinomas. However, if cancer cells invade into the proper muscle layer, the risk of lymph node metastases increases by [sim]20%. We have adopted the same criteria as those in early invasive carcinomas in the decision making for additional bowel resection after local excision, except for two cases with advanced age, and the remaining four cases were followed up without additional surgery. Although three of these four cases received additional radiation therapy, two had intrapelvic recurrence 3 and 6 years postoperatively. The local failure rate in these four cases was 50%. This figure seemed to be much higher than those reported from several other institutions (16). As shown in Table 3, the local failure rate after local excision alone for advanced rectal carcinomas ranged from 0 to 44%, average 27%. Over the last 10 years, several institutions have reported promising results of local excision combined with radiation therapy for advanced rectal carcinomas (6-11). As shown in Table 4, their local failure rates ranged from 0 to 28%, average 14%. However, the median follow-up period was still short in these studies (21-56 months) (7,17-23). In our series, although additional radiation therapy was given in only three patients pre- or postoperatively, two patients had unresectable intrapelvic recurrence 3 and 6 years after the initial treatment. The number of patients in our series was too small to draw any conclusions on the utility of adjuvant radiation therapy; however, this implies that long term follow-up of patients should be mandatory after local excision of locally invasive rectal carcinomas with radiation therapy.
Case 4 in our series had no risk factors such as poorly differentiated histology and/or lymph/venous vessel involvement; however, she had non-resectable intrapelvic recurrence 6 years after local excision with radiation therapy. Re-review of the histological sections from the surgical specimens revealed abundant dedifferentiated cancer cell nests at the invasive margin of this tumor (Fig. 2). This histological finding was almost the same as `budding' which Hase et al. stressed as one of the prognostic factors for lymph node metastases in colorectal cancers (24) or `focal dedifferentiation' which Ono et al. stressed as one of the markers for liver metastases (14). In our series, five out of eight patients had moderate to severe H-inv and three of them had lymph node metastases. On the other hand, one out of three patients with mild H-inv had lymph node metastases.
Recently, Minsky et al. (25) reported their results of local excision and postoperative radiation therapy, indicating that T stage, tumor size, blood vessel invasion, lymphatic vessel invasion and ulcerative morphology have some impact on local failure and disease-free survival. However, as Minsky et al. suggested, the number of patients was too limited to perform a multivariate analysis.
Table 3.
| Institution | Treatment | Size | Follow-up (months) |
Local failure | ||
| T2 | T3 | T2 + T3 | ||||
| Karlstad Hospital, Sweden (1977-83) | LE | <4 cm | 44 (mean) | 0% (0 of 5) | 0% (0 of 1) | 0% (0 of 6) |
| Hornsby, Australia (1969-84) | LE | 22 patients <3 cm 12 patients >3 cm |
>18 | 18% (5 of 28) | 33% (2 of 6) | 21% (7 of 34) |
| Royal Melbourne Hospital, Australia (1970-84) | LE | 23 patients <3 cm 5 patients >3 cm |
51 (mean) | 44% (4 of 9) | - | 44% (4 of 9) |
| Haukeland Univ. Hospital, Norway (1972-86) | LE | 2.5 cm (median) | 59 (median) | 43% (6 of 14) | - | 43% (6 of 14) |
| Univ. Erlangen, Germany (1978-88) | LE | NS | 77.5 (median) | 20% (4 of 20) | 30% (3 of 10) | 23% (7 of 30) |
| MGH (1962-87) | LE | NS | 36 (median) | 20% (1 of 5) | 75% (3 of 4) | 44% (4 of 9) |
| Average | 25% (20 of 81) | 38% (21 of 80) | 27% (28 of 102) | |||
Table 4.
| Institution | Treatment | Size (cm) | Follow-up (months) | Local failure | ||
| T2 | T3 | T2 + T3 | ||||
| Henri Mondor (1980-87) | LE + RT | 4 (median) | 40.5 (median) | 25% (2 of 8) | 33% (3 of 10) | 28% (5 of 18) |
| MGH (1962-87) | LE + RT | 4 (mean) | 26 (median) | 18% (2 of 11) | 33% (1 of 3) | 21% (3 of 14) |
| MGH (1985-?) | LE + RT | 2.4 (mean) | 47 (median) | 0% (0 of 11) | 0% (0 of 1) | 0% (0 of 12) |
| JCRT (1986-?) | LE + RT | 5.1 (mean) | 21 (median) | 0% (0 of 10) | - | 0% (0 of 10) |
| NEDH (1988-?) | LE + RT | 2-3 | 25 (median) | 0% (0 of 13) | 0% (0 of 1) | 0% (0 of 14) |
| MSKCC (1986-92) | LE + RT | 3 (median) | 37 (median) | 17% (2 of 12) | 33% (2 of 6) | 22% (4 of 18) |
| Fox Chase (1986-91) | LE + RT | 3 (median) | 56 (median) | 13% (2 of 15) | 25% (1 of 4) | 16% (3 of 19) |
| M.D. Anderson (1985-89) | LE + RT | 3 (median) | 36 (median) | 7% (1 of 15) | 20% (3 of 15) | 13% (4 of 30) |
| Thomas Jefferson (1984-91) | RT + LE | 40 (median) | 0% (0 of14) | 67% (2 of 3) | 12% (2 of 17) | |
| Average | 8% (9 of 109) | 28% (12 of 43) | 14% (21 of 152) | |||
In our series of 73 mp-colorectal cancer cases (invasion into the proper muscle layer) treated with bowel resection and regional lymph node dissection in the last 20 years, lymph node metastases were seen in 45 and 16% in lymphatic invasion positive and negative cases and in 19 and 21% in vascular invasion positive and negative cases, respectively. Furthermore, lymph node metastases were seen in 7 and 34% in none to mild H-inv cases and moderate to severe H-inv cases, respectively (unpublished data). Although these data suggest that lymphatic invasion is a useful marker for predicting lymph node metastases (5), histological diagnosis of lymphatic invasion is often subjective and obscure because of interobserver variation, sampling errors or artifacts. Tissue retraction around cancerous foci may mimic lymphatic invasion. On the other hand, dedifferentiated histology at the invasive margin (H-inv) can be more easily and objectively diagnosed in routine pathological practice. More experience will be needed to confirm the usefulness of H-inv in selecting the subgroup of locally invasive rectal cancer patients in whom local excision is safe and appropriate from the oncological point of view.
References
This article has been cited by other articles:
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 16 Oct 1998
Copyright©Japanese Journal of Clinical Oncology, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
K. Matsuda, T. Masaki, T. Watanabe, J. Kitayama, H. Nagawa, T. Muto, and Y. Ajioka
Clinical Significance of MUC1 and MUC2 Mucin and p53 Protein Expression in Colorectal Carcinoma
Jpn. J. Clin. Oncol.,
February 1, 2000;
30(2):
89 - 94.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Request Permissions
Google Scholar
Articles by Masaki, T
Articles by Muto, T
Search for Related Content
PubMed
PubMed Citation
Articles by Masaki, T
Articles by Muto, T
Social Bookmarking
What's this?