Preoperative Chemoradiation Followed by Total Mesorectal Excision for Locally Advanced Rectal Cancer: Oncologic Outcomes According to Pathologic T and N Stage

PURPOSE: Tumor response of patients with locally advanced rectal cancer after chemoradiation showed 60~70% of tumor volume reduction and T and N downstaging. Curative resection with total mesorectal excision should be followed for good oncologic outcomes. This study was designed to analyze the oncologic outcomes in patients who received preoperative chemoradiation followed by total mesorectal excision for locally advanced rectal cancer according to pathologic T and N stage. METHODS: Total 108 patients with locally advanced rectal cancer treated between 1989 and 2000. All patients were analyzed retrospectively and staged as T3, 4 N (+) by transrectal ultrasonography and pelvic MRI. All patients received a 5, 040 cGy of radiation over 5 weeks and systemic intravenous bolus chemotherapy 5 FU 450 mg/m2 and leucovorin 20 mg/m2 for 5 days was given during first and fifth weeks of radiation treatment, followed four to six weeks later by radical surgery. RESULTS: Among 108 patients there were 74 males and 34 females. Mean age was 54.4 years in male and 52.3 years in female. Mean follow up periods was 41.3 months. Complete follow up was in 96.4% of patients. Curative resection was done in 90 patients (83.3%). The most common type of surgery was low anterior resection in 40 (44.4%) and unresectable patients in 10 (9.3%). Postoperative morbidities were wound infection (n=10, 9.2%), anastomostic leakage (n=2, 1.9%), and anastomotic stricture (n=1, 0.9%). After chemoradiation, tumor stage were as follows: pathologic complete remission was in 7 (6.5%), pT1, T2 N0 (stage I) was in 21 (19.4%), T3N0 (stage II) was in 28 (25.9%) and T3 N (+) (stage III) was in 34 (31.5%). The rate of local recurrence was 10.7% in stage II and 20.6% in stage III. Systemic recurrence was 21.4% in stage II and 47.1% in stage III. 5 year survival rate according to T stage was T0 (100%), T1 (100%), T2 (79.5%), T3 (43.7%), T4 (33.3%) (p=0.0088). According to N stage, N (-)(72.0%) and N (+) (35.7%)(p=0.002). Among T3 patients, 5 year survival rate was N (-)(58.2%) vs. N (+)(32.0%)(P=0.0228). CONCLUSION: Preoperative chemoradiation followed by total mesorectal excision downstaged locally advanced rectal cancer and showed high resectability. Clinical outcomes correlated with pathologic T and N downstaging. Patients who did not show pathologic T and N downstaging showed high local and systemic failure and poor prognosis.

Laparoscopic Surgery for benign Gastric Tumor

PURPOSE: Laparoscopic surgery of the abdomen has grown rapidly in popularity due to the benefits, including a low level of post operative pain, early recovery, short hospitalization and excellent cosmetic results. Concerning tumor resection, most benign gastric tumors are ideal for the use of the non invasive method of a laparoscopic procedure. To evaluate the feasibility of laparoscopic surgery for benign gastric tumors, we analysed the clinicopathological findings, post operative course and prognosis. METHODS: Between January 1995 and August 2001, laparoscopic surgery was performed on 18 patients with benign gastric tumors at the Department of Surgery at Yonsei University Hospital. To evaluate the feasibility of laparoscopic surgery for these lesions, the sex, age, pathologic diagnosis, operative methods, tumor location, tumor size, operative time, recurrence, gas passing time and diet recovery time were analyzed. RESULTS: The patients group consisted of 3 men (16.7%) and 15 women (83.3%), with a mean age of 51.9+/-15.0 years (range, 23~80). The histopathological examinations showed 12 mesenchymal tumors (5 leiomyomas, 4 stromal tumors, 3 Schwannomas), 2 mucosa origin tumors (1 retention polyp, 1 villous adenoma), 2 ectopic pancreata, 1 carcinoid tumor and 1 lipoma. The operation methods were 14 laparoscopic wedge resections of stomach, 1 laparoscopic assisted partial gastrectomy and 1 gastrotomy and polypectomy. In 2 patients, a laparotomy was required following the laparoscopy due to difficulties in detecting the tumors. In one of the 2 converted cases, preoperative endoscopic marking of the tumor site was performed, but the dye was spread very widely from the mid body to the prepylorus. The other tumor was located on the lesser curvature of the upper third, around the esophagogastric junction, but it was not exactly identified. The tumors were located in the upper, middle, and lower third of the stomach in 7, 9 and 2 cases, respectively. As a circumferential location, 7 tumors were mainly situated on the anterior wall, 6 on the posterior wall, 3 on the greater curvature and 2 on the lesser curvature. All tumors, even those on the lesser curvature and posterior wall, were able to be resected with laparoscopy. The size of the resected tumors averaged 2.2+/-0.9 cm (range, 0.8~4.3). The resection margins were all negative. The operation time averaged 145.8+/-57.0 min (range 70~280). The time to passing gas averaged 2.2+/-0.9 days (range 1~4). The recovery time to a soft diet averaged 5.9+/-1.9 days (range, 2~9). The postoperative course of all the patients was uneventful, and there were no postoperative complications with the exception of one case of diet intolerance. During the follow up there have been no recurrences to date. CONCLUSION: With its proper application in benign gastric tumors, laparoscopic surgery needs to identify the exact site of a tumor, can contribute significantly to an improved patient outcome because it is less complicated and safer compared to conventional gastrectomy methods.

Clinicopathological Characteristics of Sporadic Colorectal Cancer with DNA Microsatellite Instability

PURPOSE: Sporadic colorectal cancers, with DNA microsatellite instability (MSI), have been characterized by a predilection area of proximal colon, younger age onset, exophytic growth and larger tumor size. MSI colorectal cancers have recently been had a good survival rate. The aim of this study is to determine the MSI status in sporadic colorectal cancers, and compare their clinical and pathological characteristics with those of MSS (Microsatellite Stable) cancers. METHODS: Between March 1995 and December 1997, deep frozen fresh tissue of 107 eligible colorectal cancer patients, who underwent surgical resections, were used for analysis. Hereditary nonpolyposis colorectal cancer, and familial adenomatous polyposis, patients were excluded. All the patients were registered on a colorectal cancer database, and followed up completely with regular visits for a potential recurrence. Genomic DNA was prepared by the SDS-proteinase K and phenol chloroform extraction methods. The DNA was amplified by PCR at five microsatellite loci (BAT26, BAT25, D2S123, D5S346, and D17S250) to evaluate the MSI. The PCR products were separated in 6% polyacrylamide gels, containing 5.6 M urea, followed by autoradiography. The MSI was defined as being over 2 marker positive, and the MSS as 1 marker positive, all marker negatives were classed as MSS. The survival rates were calculated by the Kaplan- Meier methods. RESULTS: MSI was noted at 16/107 (15%), with mean ages for the patients of 51.8 vs. 58.6 years old for MSI and MSS, respectively. For the patients under 40 years old 5 (31.3%) vs. 6 (6.6%) had MSI and MSS, respectively (P<0.01). The cancer was located in the right colon in 12 of each of the MSI and MSS (P<0.01). There were no MSI rectal cancer tumors. The average tumor sizes were 7.6 3.6 cm vs. 5.3 2.2 cm (P<0.01) for MSI and MSS, respectively, but there were no correlations with the frequency of associated polyps, recurrence and distant metastasis between MSI and MSS. The cells were well differentiated (12.5% vs. 17.6%), moderately differentiated (68.8% vs. 76.9%), poorly differentiated (6.2% vs. 3.3%), and mucinous type (12.5% vs. 2.2%), with MSI and MSS, respectively. The overall survival rates were 93.8% vs. 73.8% for MSI and MSS (P=0.07), respectively. CONCLUSIONS: Sporadic colorectal cancer, with DNA microsatellite instability (MSI), was located predominantly in the proximal colon, in the younger age onset, and larger size of tumor. The survival rate of the patients with MSI tumors were good, but with no statistical significance.

Clinicopathologic Characteristics of Colorectal Mucinous Carcinoma

PURPOSE: The aim of this study was to assess the clinicopathological characteristics and prognosis of colorectal mucinous carcinomas. METHODS: 1, 809 patients who underwent curative resection, between January 1989 and December 1999, for colorectal carcinomas, were enrolled in this study. Of these, 128 patients had more than 50% mucin secreting acini that could be classified as the mucinous type. These mucinous carcinoma patients were compared with 1, 681 non-mucinous carcinoma patients, in order to evaluate differences in their clinicopathological characteristics, recurrence pattern and prognosis. RESULTS: The incidence of mucinous carcinomas was 7.1%. The mean age of the mucinous carcinoma patients was younger than the non-mucinous carcinoma patients (P=0.001). The mucinous carcinomas occurred in the right colon more frequently than non-mucinous carcinoma, and the size of the tumor was larger (P<0.001). There were no significant differences in the loco-regional and systemic recurrence rates, but the systemic recurrence pattern of the mucinous carcinomas was different from those of the non-mucinous carcinomas. With a mucinous carcinoma, the most common site for systemic recurrence was the peritoneal metastasis, whereas in a non-mucinous carcinoma, was the liver. There was no significant difference in the 5-year survival rate between the two groups. CONCLUSION: In a mucinous carcinoma, the most common pattern of treatment failure was a peritoneal recurrence. The hepatic recurrence rate of the mucinous carcinomas was less than that of the non-mucinous carcinoma. Therefore, more aggressive diagnostic and therapeutic approaches, under the consideration of peritoneal recurrence in mucinous carcinomas, will be required.

Prognostic Significance of p53, nm23 and VEGF Expression in Primary and Hepatic Colorectal Cancer Metastases Following Surgical Resection

PURPOSE: The aim of this study is to analyze a correlation between related molecular markers and prognosis after curative resection for primary and hepatic metastasis for colorectal cancer. METHODS: A total 63 patients who have been resected curatively for primary and metastatic colorectal cancer between 1989 and 2000. All patients were completely followed up and recurrence and survival rates were analyzed. All paraffin embedded tumor tissues in primary and metastatic tumors were used for microtissue array and immunohistochemical staining of p53, nm23 and VEGF. RESULTS: Mean follow up period was 30.9 months. Recurrence was noted in 39/63 (61.9%) and 5 year survival rates was 27.7%. 5 year survival rates according to protein expression in primary tumor: p53+/-: 24.6% vs 27.3%, nm 23 +/-: 17.6% vs 38.9%, VEGF +/-: 38.8% vs 21.6% (P=0.16, 0.06, 0.9, respectively). 5 year survival rates according to protein expression in metastatic tumor, p53 +/-: 18% vs 59.2%, nm 23 +/-; 38.2% vs 15.8%, VEGF +/-: 38.8% vs 21.6% (P=0.03, 0.35, 0.96, respectively). A patients recurred within 1 year after surgery (group I, N=23) were compared with patients who recurred 1 year after (group II, N=16). nm23 expression in primary tumor in each group of patients: ; 15/23 (65.2%), : 4/16 (25 %), : 8/23 (34.8%), : 12/16 (75%), respectively (P= 0.013). But, p53, VEGF expression in primary tumor showed no statistical significance. nm23 expression in metastatic tumor revealed no statistical significance between two group of patients. CONCLUSIONS: p53 expression in metastatic tumor and nm 23 expression in primary tumor can predict poor prognosis after curative resection for primary and metastatic colorectal cancer. Molecular marker expression in primary and hepatic colorectal cancer can give us a reliable prognostic values.

Radiofrequency Ablation of Hepatic Metastasis from Colorectal Cancer; Early Experience

PURPOSE: The aim of this study is to describe the safety and efficacy of radiofrequency ablation (RFA) in the treatment of unresectable liver metastasis from colorectal cancer. METHODS: From January 1999 to December 2000, we undertook 48 RFA procedures in 36 patients with unresectable metastatic liver tumor from colorectal cancer. RFA was performed either via celiotomy (n=23) or using a percutaneous approach (n=25) under ultrasound guidance using a LeVeen needle electrode and a RF 2000 generator. Patients were followed with spiral computed tomographic scans at 1 week after RFA. RESULTS: RFA was performed in 15 patients with synchronous hepatic metastasis and 21 with metachronous hepatic metastasis. The mean tumor size was 3.1 cm (range; 0.5~7.2 cm). In 2 patients, complete ablation failed due to anatomic tumor location. Nine patients (25%) exhibited post RFA complications (9 cases of fever, 8 of abdominal pain, and 1 of hematoma) which showed spontaneous resolution and there was no treatment-related death. At a mean follow up of 10.1 months, 12 patients (33.3%) had recurred in the liver and 3 (8.6%) at the RFA site, while 16 (44.4%) remained clinically free of disease. CONCLUSION: Although RFA is a relatively safe procedure in patients with unresectable hepatic metastasis from colorectal cancer, the possible development of new metastatic disease after RFA is currently a limitation in this form of treatment. Further study on the efficacy of RFA versus other treatment modalities is needed.

Prognostic Factors Influencing the Recurrence Pattern and Survival Rates in Curatively Resected Colorectal Cancer

PURPOSE: The treatment of colorecatal cancer depends primarily upon the stage, and whether or not the regimen of adjuvant therapy can also be decided through this staging. In fact, the clinicopathologic prognostic factors are well known. In addition to these prognostic factors, the importance of molecular biological prognostic factors has also come to light. METHODS: This study was devised to analyze the recurrence pattern and the survival rate and correlate them with the prognostic factors in a group of surgically treated colorectal cancer patients, who were recruited from 1989 to 1998 at the Severance Hospital, Yonsei Uinversity College of Medicine. The survival curves were analyzed according to the Kaplan-Meier method, and the Cox regression hazard model was used to analyze the prognostic factors influencing the survival rates. RESULTS: A total of 1973 patients were recruited, and among them 1848 (93.7%) had undergone a resection with 1643 (89.1%) having a curative resection. The overall follow-up time after surgery was 60.5+/-35.4 months and 1945 (98.6%) patients were confirmed to have either survived or died. The overall recurrence rate was 386/1643 (23.5%). Of these, the local recurrence rate was 86/386 (22.2%), the systemic recurrence rate was 236/386 (61.1%), and both local and systemic recurrence rates were 50/386 (12.9%). The overall 5 year survival rate was 61.5%, and according to the TNM stage, the survival rate was 84.3% for stage 1, 79.2% for stage 2, 61.5% for stage 3, and 31.5% for stage 4. Multivariate analysis indicated that the TNM stage, the number of lymph nodes involved, the gross appearance of the tumor, the positive lateral margin, vascular invasion, the preoperative CEA level (higher than 5ng/ml), and rectal cancer were significant prognostic factors for the 5 year survival rate. CONCLUSION: Patient with ulceroinfiltrative, poorly differentiated, rectal cancer and the positive lateral resection margin, vascular invasion etc. had a poor survival rate after a curative resection for colorectal cancer. In addition to these clinicopathologic prognostic factors, an investigation into the molecular biological prognostic factors is also needed.

Assessment of Sexual and Voiding Function after Total Mesorectal Excision with Pelvic Autonomic Nerve Preservation in Male Rectal Cancer Patients

PURPOSE: The aim of this study was to assess the safety of TME with pelvic autonomic nerve preservation in male rectal cancer patients in terms of voiding and sexual function. METHODS: We performed uroflowmetry using Urodyn (Dantec, Denmark) and a standard questionnaire employing the IIEF (International Index of Erectile Function) and the IPSS (International Prostate Symptom Score) pre- and postoperatively in 68 male rectal cancer patients. RESULTS: There were significant differences of mean maximal flow rate and voided volume before and after surgery (18.9+/-5.7 vs 13.7+/-7.0, 240+/-91.9 vs 143+/-78, P0.05). The total IPSS (International Prostate Symptom Score) was increased after surgery from 6.2+/-5.8 to 9.8+/-5.9 (P<0.05). There were no changes of score for one of each seven IPSS items in from 49 patients (73.5 percent) to 61 patients (89.7 percent). Five IIEF (International Index of Erectile Function) domain score (erectile function, intercourse satisfaction, orgasmic function, sexual desire and overall satisfaction) was statistically decreased after surgery (18.2+/-9.3 vs 13.5+/-9.0, 8.4+/-4.2 vs. 4.4+/-2.9, 5.8+/-2.9, vs. 4.4+/-2.9, 6.1+/-2.4 vs. 4.8+/-2.0, 6.1+/-2.2 vs. 4.5+/-2.3, P<0.05, respectively. Erection was possible in 55 patients (80.9 percent), but penetration ability was possible in 51 patients (75 percent). Complete inability for erection and intercourse was observed in 3 patients (5.5 percent). Retrograde ejaculation was noted in 9 patients (13.2 percent). IIEF domains such as sexual desire and overall satisfaction were markedly decreased in 39 patients (57.4 percent), 43 patients (63.2 percent), respectively. Multiple regression analysis of factors affecting postoperative sexual dysfunction showed that over 60 years (sexual desire: P=0.019), within 6 months (erectile function: P=0.04, intercourse satisfaction: P=0.011, orgasmic function: P=0.03), lower rectal cancer (erectile function: P=0.02, intercourse satisfaction: P=0.036, orgasmic function: P=0.027) were significant factors. CONCLUSIONS: TME with pelvic autonomic nerve preservation technique showed a safety and comparable data in preserving sexual and voiding function. The IPSS and IIEF questionnaire were useful and more investigative in assessing urinary and sexual function.

Malignant Tumors of the Anus: Spectrum of Disease, Treatment and Outcomes

PURPOSE: This retrospective study was performed to evaluate clinicopathologic findings, outcomes according to the treatment modality, and prognostic factors in anal cancer. METHODS: Among the 64 patients who were diagnosed as anal cancer at our department from September 1986 to December 1999, 55 patients were analysed retrospectively. Nine patients who refused the treatment or whose medical record could not be retrieved were excluded. Concurrent chemoradiotherapy was performed for twenty-seven patients with squamous cell carcinoma. The chemotherapy with 5-FU and cisplatin and the radiotherapy were started at the same time. 750 mg/m2/day of 5-FU was infused intravenously for 5 days and 100 mg/m2 of cisplatin was started on the second day of therapy. The second cycle chemotherapy was given for 5 days before the radiotherapy was completed. A dose of 5,400cGy was given to the primary lesion and whole pelvis including inguinal area. Eight patients with squamous cell carcinoma were treated by surgery including abdominoperineal resection, local excision, or wide excision. Abdominoperineal resection was the primary treatment modality for melanoma of anus. RESULTS: Among 55 patients with anal cancer, the dominant histologic type was squamous carcinoma (n=35), followed by cloacogenic carcinoma (n=6) and melanoma (n=6). The clinical stages by AJCC were classified as stage I: 4 cases, stage II: 15 cases, stage III: 29 cases, stage IV: 7 cases. The overall 5-year survival rate of anal cancer was 60%. The 5-year survival rate in squamous carcinoma was 79.9% for the concurrent chemoradiotherapy group (n=27) and 54.7% for the surgical resection group (n=8), which was statistically insignificant. Variables affecting the survival rate with statistical significance were age, the initial tumor size, and the state of lymph node and distant metastasis. CONCLUSIONS: The concurrent chemoradiotherapy for patients with squamous cell carcinoma of the anus offered the same outcomes equivalent to surgical modality and preserved anal sphincter function. Melanoma of the anus exhibited poor prognosis and more systemic recurrence regardless of treatment modality. On univariate analysis for risk factors, age, tumor size, and lymph node and distant metastasis had statistical significance.

Clinicopathological Characteristics of Multiple Primary Colorectal Cancer

PURPOSE: Through studying the clinical manifestation and prognosis of multiple colorectal cancer in comparison to those of solitary primary colorectal cancer, we expect this study to help establish a strategy for the diagnosis and treatment of it. METHODS: 2,302 patients with colorectal cancer, operated at the department of surgery, Yonsei university college of medicine, between Jan. 1989 and Dec. 1999 were evaluated for multiple colorectal cancer. Synchronous cancer was defined as distinct lesions separated by a distance of greater than 4 cm with the invasion of the tumor below the muscularis mucosa at the time of diagnosis. Metachronous cancer was defined as the development of colon cancer more than 6 months after the initial treatment without evidence of recurrence or metastasis from the primary tumor. RESULTS: The incidence of multiple cancer was 2.1% (49 patients). Among them, 1.8% were synchronous and 0.3% were metachronous. The average age was 59.7 years old, similar to the solitary colorectal cancer group whose mean age was 57.1 years old. The median follow up period was 44 months and average follow up period was 53 months. The frequency of cancer increased as it got closer to the rectum. The most frequent location was the rectum (43%) and the sigmoid colon (24%). The average time interval until the development of metachronous cancer was 46 months after initial operation. Adenomatous polyps were identified in 11 patients (43%) out of 49 patients with multiple colorectal cancer, whereas 538 patients (24%) out of 2,253 patients with solitary colorectal cancer were discovered with ademomatous polyps. The survival rate of multiple colorectal cancer was 59%, lower than that of solitary colorectal cancer (64%). But the difference was not statistically significant (P>0.05). CONCLUSIONS: The diagnosis of multiple colorectal cancer is very important in view of curative radical surgery and prognosis. This emphasizes the need for detection of early stage cancer by developing genetic markers and using advanced radiological diagnostic tools.

Suppression of Peritoneal Metastases by Expression of Murine Endostatin cDNA / Journal of the Korean Cancer Association, 대한암학회지

Peritoneal seeding is one of problems to be solved in gastrointestinal and ovarian cancers. Angiogenesis is the critical step for a dormancy tumor cluster to be an overt metastatic nodule. However, whether an anti-angiogenesis strategy is effective in the control of peritoneal metastases is still obscure. In this study, we evaluated whether endostatin, an endogenous angiogenesis inhibitor, suppresses peritoneal metastases. MATERIALS AND METHODS: We transduced a human gastric cancer cell line, AGS and a murine renal cancer cell line, Renca, with the plasmid pEndoSTHB, which encodes a secretable form of murine endostatin. Endostatin expression was tested with western blotting, and the biological activity of the secreted endostatin was confirmed with in vitro endothelial cell growth inhibition. In the animal experiments, stable transfectants were injected intraperitoneally. RESULTS: We demonstrated secretion of endostatin from two cell lines transduced with the plasmid pEndoSTHB. Conditioned media secreted from pEndoSTSB-transduced mammalian cells were shown to potently inhibit endothelial cell growth in vitro. We selected stable transfectants with similar in vitro growth rates of their parental cell lines. Significant tumor growth inhibition was observed in the endostatin-expressing Renca cells intraperitoneal injection group at days of 28, compared to the null transfectants intraperitoneal injection control group. CONCLUSION: These results support that peritoneal seeding is angiogenesis-dependant and an anti-angiogenesis strategy is a good way to control peritoneal metastases.

Prognostic Factors Affecting Survival Rate Following Hepatic Resection for Metastatic Colorectal Cancer

PURPOSE: Hepatic resection for metastatic colorectal cancer has recently become a widely acceptable treatment modality due to its low surgical mortality and the significant improvement of 5 year survival rates seen after resection. However the use of this treatment modality remains controversial. The aim of study was to assess the survival benefits in patients who had undergone a hepatic resection for metastatic colorectal cancer as well as to determine the prognostic factors. METHODS: A retrospective study was conducted of 94 patients who had undergone curative hepatic resection for synchronous or metachronous metastatic colorectal cancer at Department of Surgery, Yonsei University College of Medicine, between June 1989 and June 2000. Cases demonstrating extrahepatic metastasis at the time of initial surgery were excluded. The survival rate was calculated using the Kaplan-Meier and Cox regression hazard model. The mean follow up period was 35 months. RESULTS: There were 58 (61.7%) and 36 (38.3%) cases of synchronous and metachronous metastasis, respectively. The 5 year survival rate was shown to be significantly lower in patients with more than 3 metastases (P=0.05), 4 or more regional lymph node metastases in primary colorectal cancer (P=0.02), bilobar metastasis (P=0.002), extra hepatic recurrence (P=0.03) and recurrence within 1 year after hepatic resection (P=0.001). Bilobar metastasis (P=0.004) and recurrence within I year (P=0.001) has been demonstrated independent factor for 5 year survival. The overall 5 year survival rate was 30.4%. CONCLUSION: Patients with multiple, bilobar hepatic metastasis demonstrated a poor survival rate. Extrahepatic recurrence within 1 year following hepatic resection was also related with a poor outcome. Therefore, in patients with poor prognostic factors, curative surgical resection accompanied by a multimodality treatment is necessary for the improvement of survival.

Pattern of Recurrence after Curative Resection for Rectal Cancer

PURPOSE: The aim of this study was to evaluate the rate and pattern of recurrence of rectal cancer as well as analyze the risk factors affecting recurrence following resection with curative intent. METHODS: 460 patients underwent curative resection for adenocarcinoma of the rectum at our clinic from 1994 to 1998. Among these, 132 patients (29.1%) whose recurrence was confirmed by clinical and radiologic examination or reoperation were studied retrospectively. The risk factors that determined the recurrence patterns were analysed with univariate and multivariate analyses. RESULTS: The mean time to recurrence was 22.0 months. The locoregional recurrence rate was 5.7% (25/440). The systemic recurrence rate was 18.4% (81/440). 12 patients (2.7%) had two or more sites of recurrence at the time of diagnosis. The most common locoregional recurrence was a pelvic recurrence (2.3%; 10/440), followed by anastomosis (2.0%; 9/440) and presacral (0.9%; 4/440). The most common site of systemic recurrence was the liver (7.0%; 31/ 440), followed by the lung (5.9%; 26/440) and peritoneum (3.2%; 14/440). The mean time from recurrence to death was 16.0 months. Logistic regression analysis demonstrated that nodal metastasis (P=0.002), vascular invasion (P=0.027), elevated CEA level (P=0.011), and microscopic invasion to the lateral margin (P=0.008) were risk factors for postoperative recurrence. When the recurrence patterns were compared to stage, the systemic recurrence rate was 3.0% in stage I, 15.3% in stage II, and 28.9% in stage III. The locoregional recurrence rate was 3.0% in stage I, 6.0% in stage II, and 6.8% in stage III. CONCLUSION: Even though an excellent local control was obtained following curative resection of rectal cancer, the main cause of recurrence was a systemic failure in advanced rectal cancer. More effective systemic chemotherapy is required for the prevention of systemic recurrence.

Prospective Randomized Trials Comparing Intravenous 5-Fluorouracil and Oral Doxifluridine as a Postoperative Adjuvant Treatment for Advanced Rectal Cancer

PURPOSE: Intravenous 5-Fluorouracil (5-FU) and oral doxifluridine were compared with respect to therapeutic efficacy, drug toxicity, and quality of life to clarify the efficiency of oral doxifluridine. METHODS: One hundred sixty-six (166) patients who underwent a curative resection for TNM stage II and III rectal cancer between Oct. 1997 and Feb. 1999 were randomized to receive intravenous 5-FU (450 mg/m2/day) or oral doxifluridine (700 mg/m2/day) in combination with leucovorin (20 mg/m2/day). 5-FU was infused intravenously 5 consecutive days per month for a total of 12 cycles (IV arm, N=74) in one group, and doxifluridine was given orally daily for 3 weeks with a rest of 1 week for a total of 12 cycles (Oral arm, N=92). Drug toxicity and quality of life were observed. Quality of life was scored according to twenty-two daily activity items (good,>71, fair,530.05). Mean number of chemotherapy cycles was 6.5+/-3.7 (IV arm) vs 7.2+/-4.3 (Oral arm). The recurrence rate was 9/74 (12.1%) in IV arm and 6/92 (6.5%) in oral arm (P=0.937). Local recurrence was 2/74 (stage III; 2.7%) in IV arm and 1/92 (stage II; 1.1%) in oral arm. Systemic recurrence was 7/74 (Stage III; 9.4%) in IV arm and 5/92 (Stage III; 5.4%) in oral arm. Toxicity pro-files are as follows: Leukopenia (30/74, 17/92) and alopecia (21/74, 13/92) were more common in IV arm than in oral arm, and the difference was statistically significant. Diarrhea was more common in oral arm. The quality of life score was better at 1 month (19.5%, 49%) and at 2 months (47%, 72%) in the oral arm group (<0.05). CONCLUSION: Oral Doxifluridine with leucovorin as a postoperative adjuvant therapy shows a therapeutic efficacy comparable to the intravenous 5-FU regimen and has a high quality of life. The oral regimen also can be safely given with an appropriate toxicity and tolerability.

Stent Insertion as a Bridge to One Stage Operation in Malignant Obstruction of Left Colon (6 cases)

Surgical treatment of acute malignant obstruction of the left colon is problematic due to the poor general state of patients, lack of bowel preparation, and the urgency of the procedure itself. A multistage approach including emergent diverting colostomy has traditionally been the method of surgical treatment for acute left sided colonic obstruction secondary to the tumor. However, these treatment efforts have been marred by an inability to complete the sequence of operations due to numerous medical co-morbidities in the patients. The use of a self expanding metallic stent to relieve colorectal malignant obstruction has been recently, described for the palliative purpose of relieving the obstruction and for converting emergency surgery into elective surgery following routine preoperative bowel preparation. Either a covered or uncovered expandable colorectal stent (Choo colorectal endoprosthesis; Sooho Medi-Tech, Seoul, Korea) was inserted by endoscopic and fluoroscopic guidance in 6 patients with acute malignant left-sided colonic obstruction. We present these 6 patients who underwent self expandable metallic stent insertion for the treatment of left sided malignant colonic obstruction as a bridge to a single operation.

The Prognosis of Patients with Stage IV Gastric Carcinoma without Distant Metastasis

PURPOSE: In the UICC staging system, stage IV contains not only those patients with distant metastasis but also patients with far advanced T and N status but without distant metastasis. We investigated the prognostic factors of stage IV gastric carcinoma patients without distant metastasis after curative resection. MATENRIALS AND METHODS: 190 stage IV gastric carcinoma patients without distant metastasis were reviewed after curative resection. RESULTS: Male sex, distal third location, Borrmann type III, IV and histologically undifferentiated type were common. 5 year survival rate of the 190 patients was 22.2%. Depth of invasion and lymph node metastasis did not influence survival. The lymph node ratio (positive lymph node / retrieved lymph node) and combined resection affected survival by univariate and multivariate analysis. CONCLUSION: Combined resection and positive lymph node ratio were the independent prognostic factors in the patients with stage IV gastric carcinoma who underwent curative resection.

Risk Factors for Recurrence after Curative Surgery for Early Gastric Cancer

PURPOSE: Even with excellent surgical outcome, recurrence of early gastric cancer (EGC) after a curative resection is not declining because the incidence of EGC is increasing. The aim of this study was to propose an appropriate treatment strategy by assessing the risk factors for recurrence of curatively resected early gastric cancer. MATENRIALS AND METHODS: Of 3662 patients who had undergone gastric resections for gastric cancer from 1987 to 1996, the cases of 1050 curatively resected EGC patients were reviewed retrospectively. Among those 1050 patients, 50 patients (4.8%) were diagnosed as having recurrent cancer, which was confirmed by clinico-radiological examination or re-operation. The risk factors that determined the recurrence patterns were investigated by using univariate and multivariate analyses. RESULTS: The mean time to recurrence was 30.9 months, and hematogenous recurrence was the most frequent type (32.0%). Among the 50 recurred patients, peritoneal recurrence showed the shortest mean time to recurrence (18.5+/-17.7 months). Between the recurred and the non-recurred patients, there was no statistically significant difference with respect to age, sex, operation type, tumor size, tumor location, gross appearance, or histological differentiation. However, depth of invasion (submucosal invasion) and nodal involvement were significantly different (P<0.001) between the two groups. Using logistic regression analyses, nodal involvement was the only significant risk factor for recurrence in early gastric cancer (P<0.001). The median survival after the recurrence had been diagnosed was 4 months. CONCLUSION: Although the prognosis for EGC patients is excellent and recurrence of EGC after a curative resection is rare, the time to recurrence and the patterns of recurrence in EGC patients were diverse and unpredictable, and the result after recurrence is dismal. Considering the impact of lymph node metastasis on recurrence of EGC, a systematic lymphadenectomy, rather than limited surgery, should be performed if lymph node involvement is confirmed pre- or intraoperatively. Also if the postoperative pathologic findings reveal lymph node involvement, adjuvant chemotherapy is recommended.

Brain Metastasis after a Gastrectomy for Gastric Cancer

PURPOSE: The common features of brain metastases from gastric cancer are unknown because brain metastasis is an uncommon pattern of metastasis. The purpose of this study was to investigate the clinical features of and the prognosis for patients with brain metastases after a curative resection for gastric cancer. MATERIALS AND METHODS: Twenty-one (21) cases of patients with brain metastases of gastric cancer, who had been treated at the Department of Surgery, Yonsei University College of Medicine, were assessed retrospectively. RESULTS: The mean age was 55.8+/-9.6 years (range: 34~70 years), and the male-to-female ratio was 2.5:1. The most common neurologic symptom was headache (38.5%), and no patient was free from the neurologic symptoms. The incidence of parenchymal metastasis (PM: 76.2%) was higher than that of leptomeningeal metastasis (LM: 19.0%). Patients with gastric cancer and brain metastasis showed high rates of blood and lymphatic vessel invasion (lymphatic vessel invasion: 85.7%; blood vessel invasion: 80.9%). According to Lauren's classification, the incidence of intestinal types was 14/21 (66.7%), that of diffuse types was 3/21 (14.3%) and that of mixed types was 4/21 (19.0%). The mean interval between the gastrectomy and the diagnosis of brain metastasis was 24.7+/-4.0 months (PM: 26.8 months; LM: 20.3 months). The median period of survival after diagnosis of brain metastasis was 2 months for paren chymal metastasis and 0 months for leptomeningeal metastasis. CONCLUSION: During a follow-up period, patients with neurologic symptoms should be suspected of having brain metastasis. Early diagnosis and treatment is the only hope to prolong survival in such patients.

Laparoscopic-assisted Colorectal Resection in Malignant Polyps and Benign Disease

PURPOSE: Laparoscopic colorectal procedures are widely used for benign disease but controversial for malignant disease. In early colorectal cancer, laparoscopic colectomy can be performed safely on the basis of oncologic principles. The purpose of this study is to evaluate the safety and effectiveness of laparoscopic-assisted colorectal resection for malignant polyps and benign disease. METHODS: Twenty five patients submitted to surgical treatment between Oct. 1996 to June 2000 were reviewed retrospectively. RESULTS: Malignant polyps comprized 7 cases whose resection margins were all positive for cancer cells after endoscopic polypectomy and benign diseases in 18 cases (benign polyp: 7, diverticular disease: 4, submucosal tumor: 4 etc.). The common sugical procedures were anterior or low anterior resection (7 cases) and segmental resection (6 cases). There was no conversion to an open surgery. In malignant polyps, pathologic results revealed early cancer with no lymph node metastasis. There was no operative mortality. Postoperative recovery was uneventful except 2 cases (9.0%) of complications, which were, prolonged ileus in one patient and subcutaneous emphysema in another patient. CONCLUSIONS: Laparoscopic-assisted resection can be recommended as a safe and effective procedure for treatment of colonic malignant polyps and benign disease.

Clinical Usefulness of PET Scan during Follow up after Curative Resection for Colorectal Cancer

PURPOSE:The aim of this study is to assess the potential role of PET in patients with unexplained rising CEA levels after curative resection of colorectal cancer or for differential diagnosis of the result by conventional radiologic study. METHODS:18 patients with abnormal CEA levels and normal results of conventional methods of tumor detection and 7 patients who needed differential diagnosis after conventional X-ray were studied with PET. The PET results were compared with pathologic findings or conventional radiologic and clinical follow-up for 6 months. RESULTS:PET showed increased uptake in 17 of 25 patients. Using PET in 18 patients with elevated carcinoembryonic antigen levels, 12 patients showed increased uptake of FDG. Among them, 11 positive findings were turned out to be a true recurrence and there was one false positive finding. Six patients with elevated carcinoembryonic antigen levels showed normal PET scan. But conventional X-ray or repetition of PET confirmed the recurrence in 2 patients among them. Using PET in 7 patients for differential diagnosis of conventional X-ray, 3 patients showed hot uptake of FDG and all of them were turned out to be a true recurrence. There were no abnormal findings in 4 patients but one patient among them showed a hepatic recurrence. It was confirmed by abdominal MRI. Overall, sensitivity was 82.4% and specificity was 87.5% CONCLUSIONS:When conventional examinations are normal except CEA elevation or equivocal, PET is a valuable imaging tool for the detection of recurrence.