[Prehepatectomy Chemotherapy Using Circadian Chrono-HAI for Liver Metastases from Colorectal Cancer].

We investigated the safety and efficacy of circadian chronotherapy via the hepatic artery(chrono-HAI)as a prehepatectomy chemotherapy for initially unresectable colorectal liver metastases. Five-day course of chrono-HAI using 5-FU, l-LV, and L-OHP plus systemic panitumumab with 9-day interval were administered to 24 patients with failure for previous chemotherapy. Response rate and Grade 3 adverse effect(AE) were 63% and 54%, respectively. Among 22 patients( excluding 2 CR patients), conversion surgery could be performed in 10(45%). Two-year overall survival of patients with surgery (58%)was longer in those without(20%, p=0.057). Although incidence of AE was a bit high, chrono-HAI plus systemic panitumumab is an effective prehepatectomy chemotherapy for patients with aggressive colorectal liver metastases.

Randomized phase II study of tegafur-uracil/leucovorin versus tegafur-uracil/leucovorin plus oxaliplatin after curative resection of high-risk stage II/III colorectal cancer (SOAC-1101 trial).

PURPOSE: This randomized phase II trial compared tegafur-uracil/leucovorin (UFT/LV) plus oxaliplatin (TEGAFOX) to UFT/LV as adjuvant chemotherapy for patients with high-risk stage II/III colorectal cancer. METHODS: From 2010 to April 2015, 159 patients who underwent curative resection were randomly assigned to receive TEGAFOX (85 mg/m2 oxaliplatin on days 1 and 15, 300 mg/m2/day UFT and 75 mg/day LV on days 1-28, every 35 days for five cycles) or UFT/LV. The primary study endpoint was disease-free survival. RESULTS: The 3-year disease-free survival rate was 84.2% in the TEGAFOX arm, versus 62.1% for UFT/LV. The stratified hazard ratio for disease-free survival for TEGAFOX compared to UFT/LV was 0.338 (P < 0.01). The incidence of any-grade adverse events was significantly higher in the TEGAFOX arm (96.1%) than in the UFT/LV arm (76.6%; P < 0.01). The rates of any-grade neutropenia, thrombocytopenia, aspartate aminotransferase/alanine aminotransferase elevation, and peripheral sensory neuropathy were higher in the TEGAFOX group, whereas the incidence of grade ≥ 3 adverse events did not differ between the groups. CONCLUSIONS: TEGAFOX is an additional adjuvant chemotherapy option for high-risk stage II/III colorectal cancer. TRIAL REGISTRATION: UMIN ID: 000007696, date of registration: April 10, 2012.

The updated five-year overall survival and long-term oxaliplatin-related neurotoxicity assessment of the FACOS study.

PURPOSE: We previously reported the first evidence of oncological benefits from a Japanese phase II trial of oxaliplatin-based adjuvant chemotherapy in patients with stage III colon cancer (the FACOS study). We herein report the long-term survival and persistent oxaliplatin-related peripheral sensory neuropathy (PSN) for patients enrolled in this trial. METHODS: Patients were scheduled to receive the mFOLFOX6 or CAPOX regimen in the adjuvant setting. The five-year overall survival (OS) rate and persistent PSN were evaluated. RESULTS: A total of 130 patients (mFOLFOX6, n = 73; CAPOX, n = 57) were eligible. The 5-year OS rate was 91.4%. No significant difference in the OS rate was observed between regimens (mFOLFOX6, 94.4%; CAPOX, 87.4%; P = 0.25). The incidence of PSN during adjuvant treatment was 55.4% in grade 1 (G1), 30.0% in G2, and 4.6% in G3. No patients showed G3 PSN at 12 months, but G1 or G2 residual PSN after 5 years was observed in 21.8% (G1, 20%; G2, 1.8%). CONCLUSIONS: Updated results from the FACOS study support the benefits of oxaliplatin-based adjuvant chemotherapy in terms of the long-term survival among Japanese patients with stage III colon cancer. However, long-term persistent PSN occurs in about 20% of survivors, counterbalancing the favorable OS.

Oxaliplatin-induced increase in splenic volume: experiences from multicenter study in Japan.

BACKGROUND: Chemotherapy with oxaliplatin is known to induce sinusoidal obstruction syndrome (SOS). In a previous single-center study, we reported that oxaliplatin-induced increase in splenic volume (SV) is strongly indicative of SOS, and that this increase in SV persisted for > 1 year after completing chemotherapy. The aim of this study was to confirm the oxaliplatin-induced SV change in a multicenter study in patients with stage III colon cancer in Japan. METHODS: We enrolled 59 patients who underwent curative resection for stage III colon cancer in the FACOS study in a phase II multi-center clinical study. Participants received mFOLFOX6 or CAPOX as adjuvant chemotherapy. SV change was assessed three times by computed tomographic volumetry: before surgery, on completion of adjuvant chemotherapy, and 1 year after completing adjuvant chemotherapy. RESULTS: SV on completing and 1 year after chemotherapy was significantly higher than that before surgery (P < 0.001). Oxaliplatin-induced SOS persisted for > 1 year after the completion of adjuvant chemotherapy in half of the patients. There was no difference in 3-year disease-free survival with respect to the presence or absence of increased SV. An increase in SV was observed in 72% of patients treated with mFOLFOX6 and 94% of patients treated with CAPOX (P = 0.13). CONCLUSION: This study can be verified the findings observed in our previous single-center study, oxaliplatin-based adjuvant chemotherapy was associated with an increase in SV. Furthermore, this increase can persist for > 1 year. The continuous presence of SOS may have a negative impact on prognosis in patients that develop recurrent disease.

A Japanese multicenter phase II study of adjuvant chemotherapy with mFOLFOX6/CAPOX for stage III colon cancer treatment after D2/D3 lymphadenectomy.

PURPOSE: A phase II trial was conducted to investigate the benefit of oxaliplatin-based adjuvant chemotherapy in Japanese stage III colon cancer patients. METHODS: Eligible patients were scheduled to receive 12 cycles of mFOLFOX6 or 8 cycles of CAPOX in adjuvant settings. The primary endpoint was the 3-year disease-free survival (DFS). Cox proportional hazards regression was performed to identify risk factors for a worse DFS. RESULTS: A total of 130 patients, including 73 patients receiving mFOLFOX6 and 57 patients receiving CAPOX, were enrolled from 16 institutions between April 2010 and April 2014. The 3-year DFS was 82.2%, exceeding the expected primary endpoint of 81.7%. The 3-year DFS tended to be higher in patients receiving mFOLOFOX6 than in those receiving CAPOX (mFOLFOX6, 86.3%; CAPOX, 76.9%; P = 0.06). The 3-year DFS rates did not differ markedly based on the risk stratification (T1/T2/T3 N1 vs. T4 or N2) indicated by the IDEA COLLABORATION study (P = 0.22). In the multivariate analysis, stage IIIC (P = 0.046) and early discontinuation (P < 0.01) were identified as independent significant risk factors for a worse DFS. CONCLUSION: Our findings represent the first positive results in a Japanese phase II trial of adjuvant chemotherapy with mFOLFOX6/CAPOX. Early discontinuation within 2 months was an independent risk factor for a shorter DFS.

Safety of mFOLFOX6/XELOX as adjuvant chemotherapy after curative resection of stage III colon cancer: phase II clinical study (The FACOS study).

PURPOSE: Adjuvant chemotherapy with oxaliplatin combined with a fluoropyrimidine derivative is widely accepted as standard therapy for patients with stage III colon cancer, since few clinical data are available for Japanese patients. The FACOS trial investigated the tolerability of modified FOLFOX6 (mFOLFOX6) and XELOX regimens in Japanese colon cancer patients. METHODS: Twelve cycles of mFOLFOX6 or 8 cycles of XELOX were given to patients with eligibility: stage III curatively resected colon cancer, performance status of 0-1, age from 20 to 75 years, and adequate organ function. The primary endpoint was 3-year disease-free survival. Secondary endpoints were the incidence of adverse events (AEs) and the completion rate of study therapy. RESULTS: From April 2010 to April 2014, a total of 132 patients were enrolled. Safety was analyzed in 130 patients, with finalized data from 73 patients receiving mFOLFOX6 and 57 patients receiving XELOX. A total of 130 patients (100%) experienced AEs (any grade), and 52 patients (40.0%) experienced AEs of grade ≥ 3. No significant difference in the frequency of grade ≥ 3 AEs was observed between mFOLFOX6 and XELOX groups. Continuation of the planned cycle rate of protocol treatment was 69.9% in the mFOLFOX6 group and 68.4% in the XELOX group. Treatment was discontinued because of AEs in 14 patients (19.2%) in the mFOLFOX6 group and 8 (14.0%) in the XELOX group. Mean relative dose intensity for oxaliplatin was 78.0% in the mFOLFOX6 group and 82.8% in the XELOX group. CONCLUSION: As adjuvant chemotherapy for stage III colon cancer, mFOLFOX6/XELOX regimens are acceptable.

Efficacy and safety of neoadjuvant chemotherapy with oxaliplatin, 5-fluorouracil, and levofolinate for T3 or T4 stage II/III rectal cancer: the FACT trial.

PURPOSE: A multicenter phase II clinical study was performed in patients with T3 or T4 stage II/III rectal cancer to evaluate the efficacy and safety of neoadjuvant chemotherapy with 5-fluorouracil, levofolinate, and oxaliplatin (mFOLFOX6). METHODS: Patients received four 2-week cycles of mFOLFOX6 therapy (oxaliplatin at 85 mg/m2 + leucovorin at 200 mg/m2 + fluorouracil as a 400 mg/m2 bolus followed by infusion of 2400 mg/m2 over 46 h, all on Day 1). They were evaluated by computed tomography after completion of the fourth cycle. If there was no disease progression, two additional cycles were administered and then surgery was performed. Adjuvant chemotherapy was generally administered for 6 months using appropriate regimens at the discretion of the physician. RESULTS: mFOLFOX6 therapy was given to 52 patients with locally advanced rectal cancer. The preoperative response rate was 48.8% and the operation rate was 80.8%. Serious adverse events of Grade 3-4 were neutropenia (n = 5), leukopenia (n = 1), thrombocytopenia (n = 1), febrile neutropenia (n = 1), nausea (n = 1), vomiting (n = 1), and peripheral neuropathy (n = 2). The R0 resection rate, pathologic complete response rate, and sphincter preservation rate were 91.0, 11.9, and 73.8%, respectively. Postoperative complications were tolerable. CONCLUSIONS: The present results suggested that neoadjuvant therapy with mFOLFOX6 is safe and effective, representing a reasonable treatment option for locally advanced rectal cancer.

Tumor 5-FU-related mRNA Expression and Efficacy of Oral Fluoropyrimidines in Adjuvant Chemotherapy of Colorectal Cancer.

BACKGROUND: It has not been elucidated whether the clinical efficacy of oral fluoropyrimidines for adjuvant chemotherapy of colorectal cancer varies with tumor biological characteristics. PATIENTS AND METHODS: A multicenter randomized trial was performed comparing oral tegafur/gimeracil/oteracil (S-1) and uracil-tegafur/ leucovorin (UFT/LV) as adjuvant therapy for stage III colorectal cancer. Postoperative survival was compared based on the 5-FU-related mRNA levels in cancer tissues. RESULTS: Among patients with tumor expressing dihydropyrimidine dehydrogenase (DPD) mRNA within the 66.7th percentile (lower 2/3) of all cases, overall survival (OS) was significantly better in the S-1 than in the UFT/LV group. In the S-1 group, patients with low DPD-expressing tumors had significantly better OS than those with highly expressing tumors. Patients with low thymidine synthase (TS)-expressing tumors had significantly better OS than those with highly expressing tumors. CONCLUSION: The efficacy of oral fluoropyrimidines as adjuvant chemotherapy for colorectal cancer may be influenced by the level of 5-FU-related mRNA in cancer tissues.

Clinical pharmacology of daikenchuto assessed by transit analysis using radiopaque markers in patients with colon cancer undergoing open surgery: a multicenter double-blind randomized placebo-controlled study (JFMC39-0902 additional study).

BACKGROUND: This exploratory trial was conducted to investigate whether daikenchuto accelerates the recovery of gastrointestinal function in patients undergoing open surgery for sigmoid or rectosigmoid cancer. METHODS: Eighty-eight patients who underwent colectomy at one of the 11 clinical trial sites in Japan from January 2009 to June 2011 were registered in the study. Patients received either placebo or daikenchuto (15.0 g/day, 5 g three times a day) from postoperative day 2 to postoperative day 8. The study end points included the gastrointestinal tract transit time evaluated with radiopaque markers and the time to first flatus. The safety profile of daikenchuto was also evaluated until postoperative day 8. RESULTS: Seventy-one patients (daikenchuto, n = 38; placebo, n = 33) were statistically analyzed. Although the number of radiopaque markers in the anal side of the small intestine at 6 h was significantly greater in the daikenchuto group than in the placebo group (15.19 vs 10.06, p = 0.008), the total transit analysis results and the mean time to first flatus did not differ significantly between the two groups. CONCLUSIONS: Daikenchuto has a positive effect on the resolution of delayed gastric emptying, but has a limited effect on the resolution of postoperative paralytic ileus after open surgery in patients with sigmoid or rectosigmoid cancer. Daikenchuto may contribute to early oral intake in the postoperative course.

Safety analysis of two different regimens of uracil-tegafur plus leucovorin as adjuvant chemotherapy for high-risk stage II and III colon cancer in a phase III trial comparing 6 with 18 months of treatment: JFMC33-0502 trial.

PURPOSE: The JFMC33-0502 trial is a phase III clinical study designed to determine the most appropriate duration of postoperative adjuvant chemotherapy with uracil-tegafur (UFT) plus leucovorin in patients with stage IIB or III colon cancer. We report the interim results of preplanned safety analyses. METHODS: Patients with stage IIB or III colon cancer who had undergone curative resection were randomly assigned to receive UFT (300 mg/m(2)) plus leucovorin (75 mg/day) for 6 months (control group, 4 weeks of treatment followed by a 1-week rest, five courses) or for 18 months (study group, 5 days of treatment followed by a 2-day rest, 15 courses). Treatment status and safety were evaluated. RESULTS: A total of 1,071 patients were enrolled, and 1,063 were included in safety analyses. Treatment completion rate at 6 months was 74.0 % in the control group and 76.7 % in the study group. Treatment completion rate in the study group at 18 months was 56.0 %. The overall incidence of adverse events (AEs) was 75.3 % in the control group and 77.6 % in the study group. The incidences of grade 3 or higher AEs were low in both groups. During the first 6 months, the incidences of the subjective AEs were significantly lower in the study group. CONCLUSIONS: Oral UFT plus leucovorin given by either dosage schedule is a very safe regimen for adjuvant chemotherapy. In particular, 5 days of treatment followed by a 2-day rest was a useful treatment option from the viewpoint of toxicity even when given for longer than 6 months.

Extrinsic surgical denervation ameliorates TNBS-induced colitis in rats.

BACKGROUND/AIMS: Neurogenic inflammation refers to an inflammatory reflex arc by sensory neurons which transmit nocious stimulus centrally and results in both pain perception and intense local inflammatory reaction. Specific neurons, receptors, and their respective neurotransmitters have been studied in numerous organ systems including the gastrointestinal tract. Neurogenic inflammation has been suggested to play a key role in the pathogenesis of inflammatory bowel disease. In this study, we studied the effect of surgical denervation of specific somatosensory neurons in a well-established animal model of colitis. METHODOLOGY: Adult male rats were underwent surgical denervation around the inferior mesenteric artery or sham operation. After ten days trinitrobenzene sulfonic acid (TNBS) or vehicle was administered by enema. Inflammation was assessed by, histological evaluation, macroscopic damage score, myeloperoxidase (MPO) activity, and substance P receptor immunoreactivity (SPRIR). RESULTS: Compared with sham operation with TNBS administration, surgical denervation with TNBS administration suppressed the score in all of the inflammatory indices and had almost no signs of inflammation in histological evaluation. CONCLUSIONS: Surgical denervation has a protective effect on TNBS-induced colitis in rats. Thus, sensory neurons play a key role in the pathogenesis of inflammation in this well-established model of acute colitis.

Randomized, controlled study of continuous 5-FU infusion starting immediately after curative surgery for advanced colorectal cancer.

BACKGROUND/AIMS: The usefulness of adjuvant chemotherapy started immediately following surgery was evaluated. METHODOLOGY: A randomized clinical trial was performed on 133 patients with advanced colorectal cancer using oral anticancer agents with one-week continuous 5-FU infusion starting immediately after curative surgery. In Group A and Group B, 300 mg/day of HCFU was orally administered for two years starting two weeks after surgery, and in addition, 333 mg/m2/day of 5-FU was drip infused from the central vein for seven days from the day of surgery in Group B. RESULTS: For 1919 days after surgery, there were no significant intergroup differences in overall survival (OS) and disease-free survival (DFS), but OS and DFS in Group B rectal cancer patients were significantly better when compared to Group A rectal cancer patients. On the other hand, OS tended to be better in Group A colon cancer patients, but no significant intergroup differences were seen, while intergroup differences tended to be smaller when corrected for disease stage. CONCLUSIONS: Continuous 5-FU infusion starting immediately after curative surgery for colorectal cancer was safe. While further investigation is necessary to elucidate the degree of improvement in postoperative prognosis, the results of the present study suggest that continuous 5-FU infusion improves prognosis in advanced rectal cancer.

Benefit of FOLFOX to unresectable liver metastases secondary from colorectal carcinoma in an oncologic emergency.

Liver metastasis is an important prognostic factor in colorectal cancer. The efficacy of resection of metastatic lesions in liver metastasis of colorectal cancer is also widely recognized. However, studies on treatment methods of unresectable cases have not been sufficient and obtaining complete remission (CR) for liver metastasis is rare with chemotherapy. Selection of reliable chemotherapy for unresectable liver metastasis is an urgent necessity. The usefulness of oxaliplatin, 5-flurouracil and leucovorin combination therapy (FOLFOX) has recently been reported, but CR of liver metastasis is rare. The current status and new therapeutic significance of FOLFOX therapy are discussed based on the literature of colorectal cancer chemotherapy to date, and the clinical experience in which we obtained CR for liver metastasis is reported. The patient had stage IV rectal cancer, perforative peritonitis, pelvic abscess and simultaneous multiple liver metastasis. The patient underwent an emergency operation using the Hartmann's procedure. Liver metastasis is considered to be a prognostic factor and FOLFOX was selected as the postoperative chemotherapy, CR of the liver metastasis was obtained. FOLFOX was suggested to have new clinical significance in oncologic emergencies against unresectable liver metastasis in colorectal cancer and should serve as adjuvant chemotherapy that will contribute to improvement of treatment results.

[Pathological evaluation of neoadjuvant chemotherapy with low-dose FP therapy for advanced gastric cancer].

In the present study, we evaluated the pathological effects of preoperative chemotherapy with low-dose CDDP and 5-FU (low-dose FP therapy) in patients with advanced gastric cancer. 50 patients diagnosed as advanced gastric cancer were administered continuous infusion of 5-FU (300 mg/m(2)/day, x 14 days) and intermittent infusion of CDDP (3 mg/m(2)/day, day 1-5 and 8-12) before surgery. The pathological effects were considered comparatively regarding the response rate of chemotherapy for gastric cancer between primary lesions and metastasis of lymph nodes and the rate of downstaging cases with low-dose FP therapy. The rates of effective cases were 26% (primary lesions) and 28% (lymph nodes). Furthermore,in the same patient,the results of low-dose FP therapy with primary tumor were more effective than those of lymph nodes (the rates of effective cases were 34% and 19%, respectively). The rate of downstaging cases with low-dose FP therapy for gastric cancer was 6%, histologically. These results indicate low-dose FP therapy for gastric cancer is promising for effective clinical management of advanced gastric cancer in preoperative treatment.

Thymidine phosphorylase expression and efficacy of adjuvant doxifluridine in advanced colorectal cancer patients.

To clarify the correlation between the expression level of thymidine phosphorylase (TP) and efficacy of doxifluridine (5'-DFUR) and 5-fluorouracil (5-FU), samples from 177 colorectal cancer patients who underwent curative resection were evaluated by immunohistochemical staining using a newly developed monoclonal antibody 1C6-203. Patients were randomly given either oral 5'-DFUR or 5-FU as postoperative adjuvant chemotherapy. In Dukes' C staged colon cancer patients treated with 5'-DFUR, better survival was observed in the high TP patients than the low TP patients (P=0.025 by the log-rank test). The observed 5-year survival rates were 91.2 and 74.8%, respectively. No correlation between TP expression and patient prognosis was detected in the 5-FU group. In Dukes' C stage colon patients with high TP expression, the 5'-DFUR group had slightly better survival than the 5-FU group. These findings suggest that TP may be a chemosensitive marker for 5'-DFUR as postoperative adjuvant chemotherapy for advanced colon cancer patients.

Thymidine phosphorylase expression correlates with malignant potential and anti-tumor effect of doxifluridine on gastric cancer: multivariate analysis for adjuvant chemotherapy doxifluridine vs. 5-fluorouracil.

Doxifluridine (5'-DFUR) is an anticancer drug converted to 5-fluorouracil (5-FU) by thymidine phosphorylase (TP). TP is an angiogenetic and platelet-derived endothelial cell growth factor. We evaluated the relation between TP expression and chemotherapeutic efficacy and prognosis for gastric cancer. Advanced gastric cancer patients given oral adjuvant chemotherapeutics either 5'-DFUR; 163 patients or 5-FU; 162 patients were examined. TP expression was assessed with immunohistochemical staining. Multivariate analysis for influencing survival was done, employing variables such as gender, age, procedure, tumor size, location, Borrmann type, histologic factors [type, depth of invasion, lymph node metastasis (n), lymphatic invasion (ly), and venous invasion (v)], drug administered, and TP expression. In the patients with serosal invasion, 5'-DFUR in TP positive was an independent prognostic factor (risk ratio, 4.450; 95% confidence limit, 2.099-9.436), indicating significantly improved prognosis over the 5-FU group. In TP negative, n and ly were independent prognostic factors, but the survival curves of the two chemotherapeutic groups were not significantly different. TP expression was not prognostic factor in 5'-DFUR group, while, in 5-FU group, TP expression was an independent prognostic factor (2.834, 1.467-5.476). In conclusion, it was suggested that TP positive gastric cancer with serosal invasion increased malignant potential of the tumor and 5'-DFUR efficacy.

A randomized controlled trial of postoperative adjuvant immunochemotherapy for colorectal cancer with oral medicines.

Postoperative adjuvant chemotherapy reportedly improves advanced colorectal cancer patients' survival, however, it is necessary to assess what regimens are useful. Doxifluridine (5'-DFUR) is an intermediate of capecitabine approved in Europe and USA to treat metastatic colorectal cancer. 5'-DFUR is metabolized to 5-fluorouracil (5-FU) by thymidine phosphorylase existing in tumor at high concentrations, suggesting high 5-FU levels in tumor tissues and lesser complications. Present study compared usefulness of 5'-DFUR to that of oral 5-FU. Patients were enrolled at 38 centers from April 1993 to September 1996. They had diagnosed colorectal cancer of TNM stages II and III, and underwent macroscopic curative resection. Patients were prestratified into colon or rectum cancer and allocated into either 5'-DFUR (5'-DFUR 460 mg/m(2)/day + PSK 3 g/day) or 5-FU (5-FU 115 mg/m(2)/day + PSK 3 g/day) group by dynamic randomization (stratification factors such as depth of tumor, degree of lymph node metastasis, and location of tumor). Drugs were orally administered daily from postoperative week 2 to 54, with 6 mg/m(2) mitomycin C at operation and following days. Subjects for analysis were 277 in 5'-DFUR and 281 in 5-FU groups. Median follow-up was 6.5 years. Although no differences in overall survival curves were detected, multivariate analysis showed that 5'-DFUR + PSK regimen was a significantly better prognostic factor in patients with Dukes B or C (risk ratio, 1.451; p=0.048); with tumor depth of pT3 or pT4 (risk ratio, 1.568; p=0.020). For patients with advanced colorectal cancer, 5'-DFUR + PSK therapy may possibly be more useful than 5-FU + PSK, but further study is required.

Postoperative adjuvant chemotherapy is effective in gastric cancer with serosal invasion: significance in patients chosen for multivariate analysis.

There are few reports on overall usefulness of adjuvant chemotherapy in gastric cancer patients. We tried to clarify, using multivariate analysis, usefulness of postoperative adjuvant oral chemotherapy in advanced gastric cancer patients after curative resection. Four hundred and eighty-two gastric cancer patients enrolled in a randomized controlled trial were classified into 2 groups based on postoperative chemotherapeutic regimen: oral doxifluridine (5'-DFUR, an intermediate metabolite of capecitabine) (n=245) or oral 5-fluorouracil (5-FU) (n=237). The significant prognostic factors in patients with serosal invasion were chemotherapeutics (5'-DFUR vs. 5-FU) (risk ratio 1.649; 95% CI, 1.112-2.437), lymph node metastasis (no vs. yes) (2.823; 1.422-5.604), and tumor differentiation (differentiated vs. undifferentiated) (1.727; 1.068-2.791). Significant factors influencing peritoneal recurrence time were chemotherapeutics (1.756; 1.063-2.902), serosal invasion (no vs. yes) (2.237; 1.264-3.961), lymph node metastasis (2.541; 1.267-5.095), tumor differentiation (2.656; 1.374-5.136), and tumor location (others vs. total) (3.595; 2.006-6.443). There were no differences in the overall survival between chemotherapy. However, 5'-DFUR produced a better survival time of patients with serosal invasion than 5-FU, that might be attributed to the prevention of peritoneal recurrence in this subset.