Search

Background Treatment with fluoropyrimidines and concomitant long-course external radiotherapy (RTE) is the standard of care in locally advanced rectal cancer (LARC) preoperative chemoradiation. A randomized phase II study (RaP/STAR-03) was conducted that aimed to evaluate the activity and safety of the monoclonal antibody anti-epidermal growth factor receptor panitumumab as a single agent in combination with radiotherapy in low-risk LARC preoperative treatment. Materials and Methods Patients had adenocarcinoma of the mid-low rectum, cT3N− or cT2–T3N+, KRAS wild-type status, and negative circumferential radial margin. Panitumumab was administered concomitant to RTE. Rectal surgery was performed 6–8 weeks after the end of preoperative treatment. The adjuvant chemotherapy regimen was FOLFOX. The primary endpoint was the pathologic complete response (pCR) rate. The sample size was calculated using Simon's two-stage design. A pCR of 16% was considered to qualify the experimental treatment for further testing. Results Ninety-eight patients were enrolled in 13 Italian centers from October 2012 to October 2015. Three panitumumab infusions were administered in 92 (93.4%) patients. The RTE compliance was median dose 50.4 Gy; ≥28 fractions in 82 (83.7%) patients. Surgical treatment was performed in 92 (93.9%) patients, and no severe intraoperative complications were observed. A pCR was observed in 10 (10.9%) patients (95% confidence interval, 4.72%–17.07%). Pathological downstaging occurred in 45 (45.9%) patients. Grade 3 toxicities were observed in 22 (22.3%) patients, and the common adverse events were skin rash in 16 (16.3%) patients. No grade 4 toxicities were reported. Conclusion The pCR rate (our primary endpoint), at only 10.9%, did not reach the specified level considered suitable for further testing. However, the analysis showed a good toxicity profile and compliance to concomitant administration of panitumumab and RTE in preoperative treatment of LARC. The pCR evaluation in all wild-type RAS is ongoing. Implications for Practice The aim of the RaP/STAR-03 study was to evaluate the activity and safety of monoclonal antibody anti-epidermal growth factor receptor (EGFR) panitumumab as a single agent without chemotherapy in low-risk, locally advanced rectal cancer (LARC) preoperative treatment. Nevertheless, the use of panitumumab in combination with radiotherapy in preoperative treatment in patients with KRAS wild type and low-risk LARC did not reach the pathologic complete response primary endpoint. This study showed a good toxicity profile and compliance to combination treatment. Further analysis of NRAS and BRAF on tissue and circulating levels of the EGFR ligands and vascular factors (soluble vascular endothelial growth factor, E-selectin) may provide insight on the potential molecular pathways involved in the anti-EGFR response.

Purpose To investigate oxaliplatin combined with fluorouracil-based chemoradiotherapy as preoperative treatment for locally advanced rectal cancer. Patients and Methods Seven hundred forty-seven patients with resectable, locally advanced (cT3-4 and/or cN1-2) adenocarcinoma of the mid-low rectum were randomly assigned to receive pelvic radiation (50.4 Gy in 28 daily fractions) and concomitant infused fluorouracil (225 mg/m2/d) either alone (arm A, n = 379) or combined with oxaliplatin (60 mg/m2 weekly × 6; arm B, n = 368). Overall survival is the primary end point. A protocol-planned analysis of response to preoperative treatment is reported here. Results Grade 3 to 4 adverse events during preoperative treatment were more frequent with oxaliplatin plus fluorouracil and radiation than with radiation and fluorouracil alone (24% v 8% of treated patients; P < .001). In arm B, 83% of the patients treated with oxaliplatin had five or more weekly administrations. Ninety-one percent, compared with 97% in the control arm, received ≥ 45 Gy (P < .001). Ninety-six percent versus 95% of patients underwent surgery with similar rates of abdominoperineal resections (20% v 18%, arm A v arm B). The rate of pathologic complete responses was 16% in both arms (odds ratio = 0.98; 95% CI, 0.66 to 1.44; P = .904). Twenty-six percent versus 29% of patients had pathologically positive lymph nodes (arm A v arm B; P = .447), 46% versus 44% had tumor infiltration beyond the muscularis propria (P = .701), and 7% versus 4% had positive circumferential resection margins (P = .239). Intra-abdominal metastases were found at surgery in 2.9% versus 0.5% of patients (arm A v arm B; P = .014). Conclusion Adding oxaliplatin to fluorouracil-based preoperative chemoradiotherapy significantly increases toxicity without affecting primary tumor response. Longer follow-up is needed to assess the impact on efficacy end points.

Background: Oxaliplatin (OXA) significantly enhanced the antitumour activity of 5-fluorouracil (FUra) in patients with advanced colorectal cancer and displayed radiosensitising properties in preclinical studies. This study was thus performed to test the feasibility, identify the recommended doses (RDs) and explore preliminarily the clinical activity of weekly OXA and infused FUra combined with preoperative pelvic radiotherapy. Patients and methods: Forty-six patients with recurrent or locally advanced (cT3-4 and/or N+) adenocarcinomas of the mid-low rectum were treated with escalating doses of OXA (25, 35, 45, 60 mg/m2, weekly for 6 weeks) and FUra (200–225 mg/m2/day, 6-week infusion) concurrent to preoperative pelvic radiotherapy (50.4 Gy/28 fractions). The RDs for the phase II part of the study were immediately below the level resulting in dose-limiting toxicities in more than one third of the patients, or corresponded to the last planned dose level. Results: In the escalation phase, dose-limiting toxicities only occurred in one patient at the fourth level and one of six patients treated at the last planned dose level (grade III diarrhoea). OXA 60 mg/m2 and FUra 225 mg/m2/day are therefore the RDs for the regimen. Among 25 patients globally treated at these doses (phase II part), the incidence of grade III diarrhoea was 16% with no grade IV toxicity. Neurotoxicity did not exceed grade II (12%). All patients completed radiotherapy and were operated on as scheduled. Twenty-one of 25 patients had the tumour down-staged after chemoradiation with seven (28%) pathological complete responses and 12 (48%) residual tumours limited to ypT1-2N0. Conclusions: Weekly OXA, at doses potentially active systemically, can be combined with full-dose, infused FUra and radiotherapy. Given the low toxicity and promising activity, this regimen is being compared to standard FUra-based pelvic chemoradiation in a randomised study.

We have reported that an alternating regimen of bolus and continuous infusion 5-fluorouracil (FU) was superior to bolus FU in terms of response rate and progression-free survival in advanced colorectal cancer. Biochemical modulation was an essential part of this regimen and it was selective for the schedule of FU administration: bolus FU was in fact modulated by methotrexate (MTX) while continuous infusion FU was potentiated by 6-s-leucovorin (LV). Considering the low cost and the favourable report on the activity of mitomycin C (mito) added to CI FU, we have incorporated this agent in the infusional part of our treatment programme. 105 patients with untreated, advanced, measurable colorectal cancer were accrued from 13 Italian centres and treated with the following regimen. 2 biweekly cycles of FU bolus (600 mg/m2), modulated by MTX (24 h earlier, 200 mg/m2) were alternated with a 3-week continuous infusion of FU (200 mg/m2daily), modulated by LV (20 mg/m2weekly bolus). Mito, 7 mg/m2, was given on the first day of the infusional period. After a 1 week rest, the whole cycle (8 weeks) was repeated, if indicated. 5 complete and 34 partial responses were obtained (response rate, 37% on the intention to treat basis; 95% confidence limits, 28–46%). After a median follow-up time of 26 months, 37 patients are still alive. The median progression-free survival is 7.7 months with an overall survival of 18.8 months and a 2-year survival rate of 30%. The regimen was very well tolerated with fewer than 13% of patients experiencing WHO grade III–IV toxicity. These results are consistent with those obtained by our group in 3 previous trials of schedule specific biochemical modulation of FU. They also indicate a highly active, little toxic, inexpensive regimen of old drugs to be used (a) as an alternative to the more expensive combinations including CPT-11 or oxaliplatin or (b) as the basis for combination programmes with these agents. © 2001 Cancer Research Campaign http://www.bjcancer.com

We have recently reported high clinical activity against advanced colorectal cancer of a regimen-alternating bolus FUra, modulated by methotrexate (MTX), and continuous infusion FUra, modulated by 6-s-leucovorin (6-s-LV). Considering the low toxicity of the bolus part of this regimen and our recent in vitro finding of a strong synergism between bolus FUra and natural-beta-IFN (n-beta-IFN), this cytokine was incorporated in the bolus part of our treatment programme. Fifty-six patients with untreated, advanced, measurable colorectal cancer were treated with two biweekly cycles of FUra bolus (600 mg m(-2)), modulated by MTX (24 h earlier, 200 mg m(-2)), and n-beta-IFN (3 x 10(6) IU i.m. every 12 h, starting at the time of FUra administration for four doses), alternating with a 3-week continuous infusion of FUra (200 mg m(-2) daily), modulated by 6-s-LV (20 mg m(-2) weekly bolus). After a 1-week rest, the whole cycle (8 weeks) was repeated if indicated. A total of 5 complete and 17 partial responses were obtained (response rate, 41%; 95% confidence limits, 28-55%) in 54 assessable patients. After a median follow-up time of 36 months, five patients are still alive. Overall, the median time to treatment failure was 6.4 months. The median duration of survival was 15.0 months. There was one treatment-related death after a course of MTX --> bolus FUra/n-beta-IFN and grade III-IV toxicity occurred in 18% of the patients. As the addition of n-beta-IFN results in high toxicity, whereas the efficacy seems to be similar to that of the same regimen without the cytokine, our groups are currently randomizing the original regimen, without IFN, against standard modulated bolus FUra.

PURPOSE To determine if fluorouracil (FUra) has different mechanisms of action as a function of the dose schedule used. DESIGN The preclinical and clinical literature relating toxicity and antitumor effects of FUra as a function of its dose schedule, with and without modulating agents, was reviewed. RESULTS The data support the hypothesis that FUra may be considered to be two different drugs, depending on its dose schedule (bolus v continuous infusion [CI]). CONCLUSION These results suggest that additional therapeutic benefit may be obtained from FUra regimens by (1) appropriate schedule-dependent modulation, (2) the sequential or alternating use of cycles of bolus followed by cycles of CI FUra appropriately modulated, or (3) hybrid regimens, ie, those that contain both pulse and CI schedules.

A total of 20 patients with advanced pancreatic adenocarcinoma were enrolled in a phase II trial testing the activity of 5-fluorouracil given at 370 mg/m2 as a rapid i. v. bolus for 5 consecutive days, preceded by a rapid i. v. bolus of 200 mg/m2 5-methyltetrahydrofolic acid. The treatment was repeated every 4 weeks. The median age of the patients was 68 years and their median Eastern Cooperative Oncology Group (ECOG) performance status was 1. There were 7 patients with locally advanced disease and 13 with distant metastases (median, 2 sites). A median of 3 monthly cycles of treatment (range, 1–7) were given, with a corresponding dose intensity of 396 mg/m2 per week (86% of that planned). No complete response, 1 partial response, and 8 cases of disease stabilization were obtained. In general the regimen was well tolerated, with only 2 patients suffering from grade 3 stomatitis or diarrhea; the most common toxicity was nausea, which was experienced by almost 50% of the patients. The combination of 5-methyltetrahydrofolate plus 5-fluorouracil appears as little effective in this disease as 5-fluorouracil plus 5-formyltetrahydrofolate (leucovorin). It is suggested that bolus 5-fluorouracil is so inactive as an “effector agent” against pancreatic cancer that its biochemical modulation with exogenous high-dose reduced folates cannot improve the therapeutic outcome produced by the fluoropyrimidine in these patients.

e15149Background: Despite standard preoperative pelvic chemoradiation (CRT) and optimal surgery with total mesorectal excision (TME) leading to local recurrence rates below 5% in locally advanced r...

Oxaliplatin radiosensitizes human colon carcinoma cells in preclinical models, and it has been shown to improve the efficacy of fluoropyrimidines (FPs) in the treatment of metastatic and radically resected, early-stage colon cancer. Its combination with 5-fluorouracil (FU) and radiation in the preoperative treatment of rectal cancer may thus result in improved control of micrometastases at distant sites and increased local tumor shrinkage before surgery. High rates of tumor sterilization at surgery were indeed observed in multiple phase I–II studies testing the combination of oxaliplatin with FPs and preoperative radiation, particularly when a weekly schedule was used for oxaliplatin administration. Seven randomized trials have then been launched to investigate the addition of oxaliplatin to preoperative radiation and concomitant fluorouracil or capecitabine. Safety and activity data have now been reported. The results consistently indicate that adding oxaliplatin to a fluoropyrimidine does not improve primary tumor response to preoperative chemoradiation with only one study showing a statistically significant, but limited, improvement in the rate of complete tumor sterilization at surgery and other measures of response to neoadjuvant chemoradiation similarly distributed between the groups treated with or without oxaliplatin in all of these studies. Conversely, toxicity was significantly increased when oxaliplatin was added to standard, FP-based preoperative chemoradiation. Data on clinical outcome have been so far published for four out of seven studies and show a small improvement in 3- or 5-year DFS that reaches statistical significance in only one study. There was no impact on local control, while distant metastases were slightly reduced in the three studies reporting on this endpoint (with statistical significance in one of them). While further follow-up of these trials and clinical outcome data of STAR-01 and PETACC-6 are awaited to fully understand the role of oxaliplatin added to preoperative chemoradiation, alternative strategies are being developed to incorporate oxaliplatin in the preoperative treatment of locally advanced rectal cancer, including neoadjuvant chemotherapy before chemoradiation and consolidation chemotherapy in the interval between chemoradiation and surgery.

The aim of this phase II study was to assess the activity of panitumumab in combination with oxaliplatin, 5-fluorouracil, and external radiotherapy (RT) as preoperative treatment in locally advanced rectal cancer patients.Patients had rectal adenocarcinoma, cT3N+ or cT4N-/+ stage, located12 cm from the anal margin. Panitumumab was administered before the start of chemo-RT, and every 2 weeks in combination with 5-fluorouracil-oxaliplatin with concurrent RT. Rectal surgery was carried out 7-8 weeks after the end of neoadjuvant treatment. The primary end point was a pathological complete response rate of 25%.Sixty patients were enrolled from February 2007 to October 2009. Fifty-five (91.7%) patients underwent surgery. Rate of pathological complete response was 21.1% (95% confidence interval 10.4% to 31.6%). Pathological downstaging occurred in 33 of 57 (57.9%) patients. Grade 3-4 toxicity during neoadjuvant treatment was diarrhea (38.9%), cutaneous reactions (18.6%), nausea (5.1%), asthenia (3.4%), anorexia (3.4%), and neutropenia (1.7%). One toxic death was observed for diarrhea.In our study, the primary end point is not reached and panitumumab combination treatment was associated with high incidence of grade 3-4 diarrhea. The higher pathological complete response rate in comparison with the results of previous neoadjuvant rectal cancer trials with anti-epidermal growth factor receptor monoclonal antibodies supports further studies necessary to understand the possibility of optimal regimens and sequences with chemo-RT.

Background Our recent findings in vitro in the human colon adenocarcinoma cell line HCT-8 suggest that resistance to fluorouracil (5-FU) in patients with advanced colorectal cancer might be overcome by use of a different treatment schedule. Purpose We tested the hypothesis that HCT-8 cells resistant to short-term 5-FU exposure retain sensitivity to continuous exposure and studied interactions between the two schedules. Methods HCT-8 cell lines resistant to short-term (pulse) treatment with 5-FU or to continuous exposure were obtained by six exposures to different concentrations of 5-FU for 4 hours or 7 days. We used a monolayer clonogenic assay to determine 5-FU-induced cell kill in resistant HCT-8 cells and sensitive parent cells. Parent cells were exposed to different concentrations of 5-FU for 1, 4, or 24 hours (short term), for 7 days (continuous exposure), or in a combination of both types of schedules. In a study of the mechanism of interaction between short-term and continuous exposure in parent cells, we performed flow cytometric DNA analysis to determine the percentage of cells in S phase and assays of thymidylate synthase inhibition in intact cells and of incorporation of [6-3H)]5-FU nucleotides into nucleic acids. Results Sensitive HCT-8 cells became fully resistant to 5-FU within five or six treatments, and low-dose continuous exposure almost immediately produced resistant clones. HCT-8 cells resistant to 5-FU given every 4 hours retained full sensitivity to continuous exposure, suggesting lack of cross-resistance between the two schedules, but cells resistant to continuous exposure were cross-resistant to short-term treatment. Parent cells showed a statistically significant (synergistic) enhancement of the cytotoxic activity for 5-FU exposure for 1 hour (100, 300, or 500 microM) followed by continuous exposure (0.5, 1, or 2 microM) or 4 hours (10, 30, or 60 microM) followed by continuous exposure (1 or 2 microM). Short-term plus continuous exposure produced a marked increase in percentage of S-phase cells, compared with the percentage for each schedule alone. The combination of 1-hour exposure and continuous exposure (1000 and 2 microM, respectively) produced a marked accumulation of cells in S phase at 24 hours (59%), which lasted up to 96 hours (53%). The combination of the two schedules produced only additive enhancement of thymidylate synthase inhibition as well as incorporation of [6-3H]5-FU nucleotides into nucleic acids of HCT-8 cells. Conclusions Our findings provide a rationale for the use of bolus 5-FU and continuous infusion 5-FU in sequence. Implication We are conducting a clinical trial of bolus methotrexate followed by continuous-infusion 5-FU plus leucovorin.

Within three repeated 7-day incubation periods with either methotrexate (MTX) or trimetrexate (TMTX), human colon adenocarcinoma cells (HCT-8) developed high levels of resistance to these drugs, as evidenced by approximately 20- and 50-fold increases, respectively, in the median effective doses. Similarly, within six short-term exposures (4 hours) to the same drugs, a high degree of resistance developed in the cells. Alternating 4-hour treatment cycles with MTX and TMTX did not delay the onset of resistance to these antimetabolites in the HCT-8 cells. The same strategy produced no better results than giving either MTX or TMTX alone to (C57BL/6 x DBA/2)F1 mice bearing murine leukemia P388 cells. Furthermore, HCT-8 cells resistant to short-term (4-hour) exposure to MTX were cross-resistant to the same drug given for 7 days continuously, and cells resistant to MTX given continuously for 7 days were cross-resistant to the same drug given for 4 hours. Analogous results were obtained with TMTX, indicating that, under these circumstances, changing the schedule of administration of the same agent does not overcome resistance to it. The clinical relevance of these data to prolonged adjuvant chemotherapy, as well as loco-regional and continuous-infusion chemotherapy, is discussed.

e14502 Background: The treatment with fluoropyrimidines and concomitant RT is the standard care inLARC preoperative chemoradiation. The aim of RaP/STAR-03 study is to evaluate the activity and safe...

Purpose To determine whether deleted in colon cancer (DCC) protein expression in colorectal cancer (CRC) metastases could predict outcome to palliative fluorouracil (FU)-based chemotherapy and to assess whether it is similar to that observed in the corresponding primary tumors. Patients and Methods DCC protein expression was assessed immunohistochemically on archival specimens of CRC metastases from 42 patients homogeneously treated by methotrexate-modulated bolus FU alternated to 6-S-leucovorin–modulated infused FU and was retrospectively correlated with patient characteristics and clinical outcome. In a subset analysis, DCC immunoreactivity was compared between metastatic CRC and the corresponding primary tumors and regional lymph node metastases. Results Positive immunoreactivity for DCC was found in 45% of patients. Eighteen (78%) of 23 patients for whom multiple samples were available displayed a similar pattern of expression in distant metastases and primary tumors. The median survival time was 14.3 months in patients without DCC expression and 21.4 months in patients with DCC-positive tumors (log-rank test, P = .04); the 2-year survival rates were 8.5% and 42.5%, respectively. Response rates to chemotherapy were not significantly different between the two groups. By multivariate analysis, DCC protein expression maintained its prognostic value and showed to be the single best predictor of survival, with a relative risk of 2.16. Conclusion Our results indicate that expression of the DCC protein in CRC metastases is similar to that observed in the corresponding primary tumors and represents a dominant predictor of survival in patients with unresectable, advanced CRC who are undergoing palliative FU-based chemotherapy.

Background Different 5-fluorouracil (5-FU) schedules and/or biochemical modulators may resultin different mechanisms of cytotoxicity, potentially affecting the correlation between thymidylate synthase (TS) expression and the clinical response to the fluoropyrimidine. Patients and methods TS levels were measured immunohistochemically on archival specimens of colorectal cancer metastases from 124 patients homogeneously treated in a series of clinical trials at our institutions with: (A) leucovorin (LV)-modulated infusional 5-FU (n = 48); (B) LV-modulated bolus5-FU (n = 41); (C) methotrexate (MTX)-modulated bolus 5-FU (n = 35). Results A statistically significant correlation between TS levels and the clinical response was observed with the regimens involving continuous infusion and/or LV modulation (response rate in patients with low and high TS: 66% versus 24%, P = 0.003, and 50% versus 0%, P = 0.0001, in group A and B, respectively). Conversely, TS levels failed to predict the clinical response within the group of patients treated with MTX-modulated bolus 5-FU (response rate 21% versus 13%, P = 0.50, with low and high TS, respectively). Consistently, the median time to progression/overall survival time in patients with low and high TS were 9 versus 6 months/19 versus 14 months (P = 0.009/0.035, group A), 8 versus 2 months/12 versus 6 months (P = 0.002/0.0006, group B) and 3 versus 2 months/12 versus13 months (P = 0.14/0.74, group C). Conclusions The correlation between intratumoral TS levels and the clinical response to 5-FU depends strongly on the schedule of administration/biochemical modulators that are used in different5-FU regimens. These data strengthen the notion that different 5-FU schedules have different mechanisms of cytotoxicity.

Thymidylate Synthase (TS) is a rate-limiting enzyme in the DNA synthetic pathway and represents the cellular target of the antimetabolite drug 5-fluorouracil (FUra). Both preclinical and clinical studies have shown that the level of expression of this enzyme and the ability to achieve its inhibition are the major determinants of sensitivity and resistance to fluoropyrimidines (FP). In particular, five recent studies have consistently demonstrated an inverse correlation between the level of TS gene or protein expression measured in colorectal cancer metastases and the clinical response to either FUra or 5-fluorodeoxyuridine (FUdR). Patients with low levels of TS expression in their metastases have indeed shown response rates that are three to ten times higher compared to those obtained in patients with high TS levels. The independent predictive value demonstrated in a logistic regression model, the longer survival shown by patients with low TS levels in three of five studies and the consistency of the results obtained by independent groups using different techniques to quantitate TS expression, strengthen the predictive role of TS. Targeted treatment of colorectal cancer based on TS quantitation has thus been hypothesized similar to the use of hormone receptor in breast cancer. In this review preclinical and clinical data supporting the use of TS quantitation to predict for the clinical response to FUra will be described and unresolved problems including assays standardization, response prediction based on TS levels measured in primary tumors, intrapatient variations in TS levels and biological/biochemical limitations of this strategy will be discussed.

Summary Background We assessed the hypothesis that a compromised p53 function could account for the non response of colon cancer patients with low thymidylate synthase (TS) expression receiving a bolus 5-fluorouracil (5-FU)-leucovorin (LV) combination. Patients and methods The study population consisted of 41 patients with unresectable metastatic colon cancer, homogeneosuly, treated with bolus 5-FU and LV. Results Twenty-seven patients (66%) showed high levels of TS expression. The difference in the proportion of objective responses between patients with low (CR + PR: 7 of 14, 50%) and high (CR + PR: 0 of 27) TS levels was statistically significant (P = 0.0001, chi-square test). p53 nuclear over-expression was found in 27 of 41 patients (66%). No differences were observed in p53 overexpression in patients with high (66%) or low (66%) TS expression. p53 status was not found to be associated with response even in patients with low TS expression. Conclusions p53 status measured by immunohistochemistry does not seem to be useful to identify unresponsive patients with low TS expression.

PURPOSE: To determine whether immunohistochemical thymidylate synthase (TS) quantitation predicts for clinical outcome in patients with advanced colorectal cancer treated by fluorouracil (FUra)-based chemotherapy. PATIENTS AND METHODS: TS levels were measured immunohistochemically on archival specimens of colorectal cancer metastases from 48 patients homogenously treated by bolus FUra plus methotrexate alternating with continuous-infusion FUra plus leucovorin. These measurements were retrospectively correlated with patient characteristics and clinical outcome. RESULTS: A significant correlation was found between intratumoral TS expression and all the parameters of clinical outcome analyzed. In patients whose tumors had low (n = 27) and high (n = 21) TS levels, the overall response rates were 67% and 24%, respectively (P = .003). The percentage of tumor shrinkage after chemotherapy was linearly related to TS immunoreactivity (r = .56, P = .00004), and its mean values were 65% and 14% with low and high TS levels, respectively (P = .0001). By logistic regression analysis, low TS expression was the single best predictor of response to chemotherapy (relative probability, 5.0). In patients with low and high TS expression, the median time to progression was 9.6 months v 6.2 months (P = .005) and the median survival time 18.4 months v 15.4 months (P = .02), respectively. Two- and 3-year survival rates were 41% v 15% and 19% v 0% (P = .02), respectively. CONCLUSION: In this cohort of homogenously treated patients, intratumor TS content was a major predictor of clinical outcome. Immunohistochemical TS quantitation provides a convenient, low-cost technique for identifying patients unresponsive to TS inhibitors who may be candidates for alternative chemotherapy regimens.

compared with 16% of patients experiencing grade 3-4 events. This might depend on differences in drug and RT scheduling. The interval between OXA administration and delivery of RT prescribed in CALGB 89901 might have resulted in more effective RT sensitization and, therefore, toxicity. However, the similar antitumor activity observed between the two studies argues against this explanation. Similarly, although the shortened duration of OXA infusion in the US study (1 hour) might theoretically affect toxicity, a selective effect on gastrointestinal toxicity is unlikely. Neurotoxicity rates were in fact slightly higher in our study (grade 1-2, 80% v66%). This difference may have other explanations. While patients in our study were accrued and treated at a single institution, patientsintheUSstudywererecruitedatseveralcenters.Different toxicity grading systems may have also been used. While in our study toxicity was scored according to the Nation Cancer Institute Common Toxicity Criteria version 2, the scoring system used in theUStrialwasnotspecified.Despitethehigherincidenceofgrade

Focusing our effort on the importance of FUra scheduling we have tested the hypothesis that pulse and continuous infusion (CI) of the fluoropyrimidine have different mechanisms of cytotoxicity. Our initial approach was to compare the mechanism of resistance of a cell line resistant to a short term exposure to FUra (HCT-8/FU4hR) to that of a cell line resistant to a prolonged exposure to the fluoropyrimidine (HCT-8/FU7dR). Cytotoxicity studies showed that HCT-8/FU4hR cells were still sensitive to FUra given as a 7-d exposure, suggesting different mechanisms of resistance. Indeed, rapid recovery of TS activity after drug removal was evident in the HTC-8/FU7dR cell line while HCT-8/FU4hR cells were similar to the parental cell line with regard to both the degree of in situ TS inhibition by FUra and duration of inhibition after FUra removal. In contrast, labelling studies with [3H-6]FUra(4 h exposure, 100 microM) showed that the incorporation of the fluoropyrimidine into RNA is significantly decreased in HCT-8/FU4hR cells as compared to parental HCT-8 cells. Given the lack of cross resistance between the two schedules in vitro, a pilot trial was done on patients with colorectal cancer refractory to bolus FUra. On 15 patients failing after FUra+LV or FUra alone 1 PR, 3 MR, 3 SD and 8 P were observed, confirming a certain degree of activity of CI FUra in patients clinically resistant to bolus FUra. Based on this rationale, a phase II trial of schedule-oriented biochemical modulation of FUra in advanced colorectal cancer patients was conducted, employing a hybrid regimen of 2 biweekly cycles of FUra bolus (600 mg/sqm), preceeded by (24 h interval) methotrexate, 200 mg/sqm (in order to maximize the RNA effect of the drug) alternating with Fura continuous infusion, 200 mg/sqm daily for 3 weeks, modulated by leucovorin, 20 mg/sqm weekly bolus (in order to maximize the DNA effect). Thirty-three consecutive patients (median ECOG PS 1) with advanced measurable colorectal cancer and no prior therapy for metastatic disease entered the study, from February 1992 to August 1993. Three complete and 13 partial responses were obtained among these 33 patients (RR = 48%, 95% confidence limits, 31-66%). After a median follow-up time of 23 months, 16 patients are still alive. The median progression free survival and overall survival were 9.6 and 20.8 months, respectively. No toxic deaths or grade 4 toxicity occurred. The incidence of grade 3 toxicity per patient in any cycle was: mucositis 6%, diarrhea 3% and vomiting 3% for the bolus part and 21%, 3% and 6% respectively, for the continuous infusion part of the regimen. Hand-foot syndrome occurred in 27% of the patients treated with the continuous infusion regimen. In conclusion, this experimental and clinical project has generated a novel regimen of schedule oriented biochemical modulation that is twice as active and half as toxic compared to bolus FU+LV given with either the daily x 5 or the weekly schedule. This high clinical activity is very encouraging, especially considering that 1) consecutive patients were entered, 2) the responses were independently reviewed, 3) the progression free survival and survival were much longer than those actually reported for this disease, 4) the toxicity of the program, in particular the bolus regimen, was relatively low allowing further intensification.

Chemotherapy of colorectal cancer has improved over the last several years thanks to both biochemical modulation of 5-Fluorouracil (FUra) [1] and attempts to optimize its schedule of administration [2]. However, resistance, either intrinsic or acquired, continues to be the major cause of treatment failure, and the use of multiple agents in combination or high-dose chemotherapy seems to offer no advantage over FUra alone. The outcome of FUra therapy may be improved if the biochemical basis for FUra resistance in tumors is identified and strategies are designed to overcome resistance to this agent.

Between February 1991 and July 1992, 79 previously untreated patients with metastatic colorectal carcinoma were enrolled in a phase II study of combined 5-fluorouracil (5-FU) and l-folinic acid (FA). 5-FU 370 mg/m2/day was administered for 5 consecutive days as an intravenous (i.v.) bolus injection preceded by l-FA 100 mg/m2/day with the same administration modality. Treatment was given every 4 weeks until progression. 79 patients were evaluable for toxicity and 64 for response. 2 patients (3%) achieved a complete remission and 8 (12.5%) a partial remission, 33 (52%) had stable disease and 21 patients (33%) had progressive disease. Median duration of remission was 32.5 weeks and median survival for all evaluable patients was 64.5 weeks. Substantial to severe side-effects occurred in 39% of patients. Dose-limiting toxicity (grade 3-4) was mainly diarrhoea (18%) and mucositis (15%). Nausea/vomiting, cutaneous toxicity, leucopenia, alopecia and conjunctivitis of grade 3-4 occurred respectively in 6, 4, 2.5, 1 and 1% of cases. Toxicity appeared to be substantially similar to that characteristic of combined 5-FU and the chiral mixture of d,l-FA. Efficacy was within the range of that observed with the 5-FU/d,l-FA combination, although at the lower level.

We have recently demonstrated that continuous-infusion (CI) 5-fluorouracil (FU) eradicates human colon carcinoma cells made resistant to bolus FU in vitro. In addition, in the same experimental system, the mechanisms of resistance to pulse and CI FU were found to be different. These observations led us to test the clinical activity of a standard regimen of CI FU (300 mg/m2 per day) in a cohort of 15 patients with advanced measurable colorectal cancer who were in progression after having failed to respond to bolus treatment with FU alone (3 patients) or FU combined with high-dose 6-S-leucovorin (LV) (12 patients). The median age of the patients was 68 years, and their median Eastern Cooperative Oncology Group performance status (ECOG PS) was 1. No myelotoxicity was observed. Mild diarrhea, mucositis, and vomiting occurred in 32%, 26%, and 19% of the patients, respectively, with no WHO grade 3 or 4 episodes being noted. In all, 6 of 15 patients complained of hand-foot syndrome, which was severe in 2 instances, lasting approximately 1 week. Overall, 1 partial response and 6 instances of disease stabilization, including 3 minor responses, were obtained both in patients who had been pretreated with pulse FU alone and in patients who had failed first-line treatment with FU + LV. Finally, 8 patients failed CI FU. In conclusion, these results, obtained in patients who were clearly progressing after having failed first-line treatment, support our experimental finding that resistance to bolus FU may be overcome by CI FU and extend this possibility to patients who are resistant to bolus treatment with FU + LV.

BIOCHEMICAL MODULATION of Muorouracil by leucovorin, methotrexate, N-phosphonacetyl-r.-aspartate (PALA) or interferon [l], is a current area of interest in clinical research in advanced colorectal cancer. However, randomised studies have shown little, if any, survival advantage over S-fluorouracil given alone in most instances [2]. New drugs are clearly needed. Fotemustine (FOTE) is a new nitrosourea that has demonstrated some activity in phase 2 studies against colorectal cancer [3], melanoma, renal cell carcinoma and brain tumours [4,5]. Preclinical experiments demonstrate that the main mechanism of resistance to this class of alkylating agents is the efficient DNA repair mechanism mediated by the enzyme 06-alkylguanine-DNA alkyltransferase (ATase) [6]. In experimental tumour systems resistance to nitrosoureas may be temporarily reversed by inactivating ATase. This can be accomplished by the administration of a methylating agent such as dacarbazine (DTIC) [7]. We have transferred this concept to the clinic and tested the sequential combination of DTIC (500 mg/m2 intravenous bolus) followed 2 h later by FOTE (100 mg/m2 in 150 cc of 5% isotonic glucose solution) in patients with advanced measurable colorectal cancer. The cycle was repeated after 4 weeks if toxicity allowed and if no progression was observed. The study was designed according to Simon’s two-stage optimal design [8]. PO and Pi were set at 20% and 40%, respectively. Setting alpha error at 0.05 and beta error at 0.20, the combination had to be rejected for untreated patients if three or fewer responses were observed among the first 13 patients or if 12 or fewer responses were observed in 43 patients. Standard WHO criteria were used to assess response and toxicity. The patients’ characteristics are shown in Table 1. A total of 30 cycles were administered (median 2). Thrombocytopenia was the most prominent toxicity, with 26% of patients suffering from WHO grade 3-4, while 1 patient only had leukopenia grade 3. Other toxicities were very mild, including nausea and vomiting that was well controlled by ondansetron. Interestingly, the only patient who did not receive the serotonin antagonist suffered from grade 3 vomiting. No partial or complete responses were observed in the first 13 previously untreated patients and the study was terminated. 6 stable disease and 7 failures were observed among these patients. No response was also obtained in an additional 6 patients who failed prior chemotherapy with 5-fluorouracil (3 stable disease and 3 failures). These data indicate the lack of activity of this

Systemic chemotherapy has only a marginal role in the treatment of unresectable hepatocellular carcinomas or secondary liver metastases from gastrointestinal malignancies, producing a 20% response rate at best. In the attempt to improve the efficacy of the chemotherapeutic agents against these diseases, several drugs have been directly administered via the hepatic artery instead of intravenous injection.1–3 The rationale behind this strategy lies in the finding that hepatic malignancies receive their blood supply primarily from the hepatic artery, while normal hepatocytes mainly derive it from the portal vein.4 Thus, hepatic artery administration allows increased tumor exposure to the drug and lower peripheral blood levels are obtained for those agents that are metabolized by the liver, when compared to the systemic administration.2 Accordingly, tumor cell killing will be enhanced and drug toxicity reduced.

ERYTHROCYTES HAVE been proposed in animal models as carriers of cytotoxic drugs or as bioreactors converting prodrugs to active drugs [ 1,2]. Doxorubicin has been encapsulated in human and animal erythrocytes [3-71. Treatment of the doxorubicin-loaded erythrocytes with glutaraldehyde, a bifunctional reagent, prevented the fast efflux of doxorubicin from the cells and produced their specific targeting to the liver of mice and dogs [4-71. The present research aimed to extend these experimental studies to the clinic. A 51-year-old male had a colorectal carcinoma, massively metastatic to the liver (60% tumour involvement). The primary tumour was removed surgically and, at the time our study started, the disease was confined to the liver and slowly progressing. The patietit had an ECOG performance status of 2. He had previously failed three lines of chemotherapy (5fluorouracil, fotemustine, mitomycin), either systemic or locoregional, through a Port-a-cath implanted in the hepatic artery at the time of surgery. Informed consent was obtained from the patient. Freshly drawn erythrocytes were loaded with doxorubicin by simple diffusion [6]; the yield of encapsulation of the drug was 2 mg/ml of packed cells. Glutaraldehyde treatment (0.15% and 0.30% final concentration) was performed after encapsulation of doxorubicin [6]. This treatment substantially reduced the rate of doxorubicin efflux in vitro (more than 70% of the drug retained after 210 min at 37°C). For in viva distribution studies erythrocytes were labeled with 99mTc [8] prior to encapsulation. All manipulations were performed under sterile pyrogen-free conditions. Figure 1 shows the time-radioactivity curves over the regions of liver and heart, after administration through hepatic artery of 99mTc labeled native (a) and unloaded GA treated (b) erythrocytes. Unlike the native cells, showing a rapid transit from the liver to systemic circulation (a), the glutaraldehyde treated erythrocytes were almost completely retained by the liver (b), up to 24 hours after injection (not shown). No accumulation of

Preliminary to studies on biopsy specimens from colorectal carcinomas, we have investigated the mechanisms of resistance to 5-fluorouracil (FUra) and 5-fluoro-2'-deoxyuridine (FdUrd) on a human colon carcinoma cell line, HCT-8 in vitro. After obtaining a high level of resistance, extracts were made from sensitive and resistant HCT-8 cells and the enzymes responsible for the activation or catabolism of fluoropyrimidines were assayed using a thin-layer chromatographic method. The activity of thymidine-kinase, thymidine phosphorylase, uridine-kinase, uridine phosphorylase, orotate phosphoribosyl transferase as well as the activity of the target enzyme thymidylate synthase were not significantly different in sensitive, FUra resistant or FdUrd resistant cells. The activities of these enzymes in sensitive HCT-8 cells were respectively 15.6, 3.4, 40.8, 2.1, 6.9 and 30.2 nmol/mg protein/h. Direct evidence that impaired transport was the mechanism of resistance to FdUrd was obtained using a short term "oil stop" technique, whereas indirect evidence suggests that reduced inhibition of thymidylate synthase is likely to be the mechanism of resistance to FUra. Cells seven hundred-fold resistant to FdUrd were in fact still sensitive to FUra (ED50 value after continuous exposure 2.1 microM) while FUra resistant cells were fully cross-resistant to FdUrd (ED50 value after continuous exposure 0.003 microM). The clinical relevance of these data is discussed in the light of the extensive literature about the potentiation of fluoropyrimidines by leucovorin.

8234 Background: Cancer patients with moderate-severe pain often require opioid analgesics. Methods: From October 2002 to November 2003, 62 outpatients underwent transdermal treatment with Fentanyl (TDF). All pts had been previously treated with other drugs according to the WHO analgesic scale. Pts characteristics: median age 64 years (range 44–84), M/F 29/33, median ECOG PS 0 (range 0–2). Primary tumors: 24 breast, 8 lung, 9 colorectal, 7 pancreas, 7 prostate, 7 others. Somatic pain was reported in 35/62 pts (56%), visceral pain in 12/62 pts (19%), 15/62 pts (24%) showed both types, with neuropatic features in 10%. Breakthrough pain was present in 24/62 pts (38%) requiring rescue medication. 47/62 pts (75%) received palliative chemo-radiotherapy. All pts were evaluated weekly with VAS (Visual Analogic Scale) and TIQ (Therapy Impact Questionnaire). A Geriatric Assessment with CIRS (Comorbility Index), ADL and IADL was performed for pts aged over 65 years (23/62; 37%) at the beginning of treatment and afte...

3637 Background: We have recently demonstrated that OXA, given weekly at the dose of 60 mg/m2, may be safely combined with infusional FU and preoperative pelvic RT for LARC resulting in rates of pathological complete responses (pCR's) approaching 30 %. Replacing infusional FU with an oral fluoropyrimidine might reduce the inconvenience and some of the potential complications of this regimen. Methods: Since June 2002, 15 patients (10 males, 5 females; median age 63, range 33–75 years) with LARC (cT3-T4 and/or N+; one patient also had a potentially resectable liver metastasis) were thus accrued in a phase I study aimed at defining the maximum tolerated dose (MTD) of UFT (200–250-300–350 mg/m2/day, monday through friday for all the duration of RT) combined with weekly OXA (60 mg/m2) and standard pelvic RT (50.4 Gy - 28 fractions). Oral LV (90 mg/day, fixed dose) was administered with the same schedule as UFT. Results: No grade III-IV toxicity was observed at the first dose level (n=4; UFT 200 mg/m2/day). At ...

3577 Background: Sequential combinations of FU and IRI or OXA are currently used as 1st and 2nd line therapy for advanced CRC. Multiple trials are also investigating the concomitant use of the thre...

S: POSTER PRESENTATIONS P99 Complications, continence and quality of life after an intensive preoperative radio-chemotherapy regimen for locally advanced rectal cancer: analysis of a phase I-II study S. Pucciarelli, ~* E. Urso, ~ S. Serpentini, ~ E Toppan, I G.L. De Salvo, 2 G. Fabris, 3 M.L. Friso, 4 C. Asclaele, 5 A. Bruttocao, 6 M. Lise. 1 1. Clinica Chirurgica I[, University of Padova, Padova, Italy," 2. Centro Oncologico Regionale, Padova, Italy; 3. Divisione Chirurgica IL Padova, Italy; 4. Divisione di Radioterapia, Padova, Italy; 5. Divisione di Oncologia, Padova, Italy, 6. Chirurgia Geriatrica, Padova, Italy. Purpose: To make a prospective investigation of toxicity, functional and oncological outcome, and quality of life (QoL) in pts with locally advanced rectal cancer given intensive preoperative radio-chemotherapy (RTCT). Methods: From May 2000 to April 2002, 44 TNM stage II-III mid-low rectal cancer pts (mean age: 58 yrs, range: 35-74; M/F: 25/19) enrolled in a multicentric prospective phase I-II study were given RTCT (50.4 Gy/28E 5FU: 200-225mg/m2/day CI and weekly Oxaliplatin, 25-60 rag/m2) followed, 4 to 6 weeks later, by surgery. In 22 pts in a single surgical unit, anorectal function and QoL were investigated using the fecal incontinence scoring system (FISS) and EORTC QLQ-CR38 questionnaire, administered before RT (tO), preoperatively (tl), and one year after surgery (t2). Statistical analysis: KaplanMeier curves for overall (OS) and disease-free survival (DFS); repeated measures analysis o f variance for differences in scores over time. Results: Acute grade 3+ toxicity was observed in 8 pts; both sphincter-saving procedure and radical resection were performed in 39 cases. There were no operative deaths; major postoperative complications (anastomotic leak, n=5; hemoperitoneum, n = 1) occurred in 6 pts, of whom 4 required re-operation. Late complications requiring re-admission occurred in 13 patients, of whom 6 required surgery. Twelve (27%) pts had a pathological complete response (pCR). At a median follow-up of 20 months, 4 pts had recurrence and 3 died. A not significant worsening in FISS was observed: median score of 25, 25, and 62, for time tO, t l , and t2, respectively. Out of the parameters investigated using QLQ-CR38, only "future perspective" showed significant improvement over time (mean score: 33, 43 and 54). Actuarial OS and DFS at 30 months were 93 and 88%, respectively. Conclusion: Following this new aggressive RTCT preoperative regimen used by us, pCR, short-term oncological results, QoL and intestinal function observed appear to be promising, although drug toxicity and late post-surgical complications are not negligible P100 mTOR signaling is a possible new biochemical target for pancreatic cancer therapy D. Ito,* K. Fujimoto, R. Doi, E. Toyoda, T. Mori, K. Kami, Y. Shimada, M. Imamura. Surgery and Surgical Basic Science, Kyoto University, Kyoto, Kyoto, Japan. Introduction: The mammalian target of rapamycin (mTOR) plays a central role in the cell proliferation through the regulation of cell cycle. Dysregulation of mTOR signaling occurs in diverse hmnan tumors, and can confer higher susceptibility to inhibitors ofmTOR. In this study, we investigated whether the mTOR signaling is activated in 6 pancreatic cancer cell lines and tissue specimens of pancreatic ductal adenocarcinoma (PDA), and examined antiproliferative effects of the mTOR inhibitor, CCI-779. Methods: First, we examined the expression pattern of PTEN, phospho-Akt (AKT), p-mTOR (mTOR) and p-p70S6K (p70S6K) in 6 human pancreatic cancer cell lines and 20 tissue specimens of PDA by western blot and immunohistochemistry. Next, we examined the cytotoxicity of the mTOR inhibitor, CCI-779 (0200 nM) in the 6 cell lines in vitro as single agent and in combination with standard chemotherapeutic drug, gemcitabine (10 nM). Finally, the antitumor effects of CCI-779 were examined using in AsPC-1 and Suit-2 xenograft models. Results: As shown in the table below, on a panel of 6 pancreatic cancer cell lines, PTEN expression was detected in 4 cell lines except for KMP-3 and KMP-4 cells. Interestingly, Akt was strongly phosphorylated even in the 4 cell lines which showed the positive expression of PTEN. mTOR and p70S6K were activated in all of the 6 cell lines examined. In the tissue specimens of PDA, p-Akt expression was detected in 60% of the specimens, p-mTOR was detected 55%, and p-p70S6K was detected in 45%. With respect to the growth inhibitory effects of CCI-779, AsPC-1 and KMP-3 cells were highly sensitive to the treatment, and BxPC-3 and Suit-2 cells were slightly resistant. Intriguingly, CCI-779 had additive effects with gemcitabine in the 2 cell lines resistant to the treatment with CCI-779 alone such as BxPC-3 and Suit-2. Furthermore, CCI-779 induced antitumor activity in both of AsPC-1 (sensitive in vitro) and Suit-2 (resistant in vitro) xenografts. Conclusions: These results suggest that roTOR may be a good target for pancreatic cancer therapy and the roTOR inhibitor, CCI-779 is an important new agent to investigate in the treatment of pancreatic cancer. Expression of roTOR signaling and effect of CCI-779 on the growth of human pancreatic cancer eel)

We have recently demonstrated that bolus FUra treatments of human colon carcinoma cells, HCT-8, produce resistance via an RNA-related mechanism, while prolonged exposures to the fluoropyrimidine is responsible for a DNA-related mechanism of resistance. In addition, cells resistant to short term FUra exposure retain full sensitivity to the continuous exposure to the same agent. These data suggest that biochemical modulation of FUra should take into account the schedule of fluoropyrimidine administration. Based on this rationale, we completed a phase 2 trial of schedule-oriented biochemical modulation of FUra in advanced colorectal cancer patients, based upon a hybrid regimen of 2 biweekly-cycles of FUra bolus (600 mg/sqm), preceeded by (24 h interval) MTX, 200 mg/sqm (in order to maximize the RNA effect of the drug) alternating with FUra continuous infusion, 200 mg/sqm daily for 3 weeks, modulated by leucovorin, 20 mg/sqm weekly bolus (in order to maximize the DNA effect). Among the 33 consecutive patients accrued there were 3 CR and 13 PR (RR = 48%, 95% CL, 31–66%). Eleven patients had a minor response and 4 of them showed tumor shrinkage ranging between 46% and 49%. After a median follow-up time of 26 months, 10 patients are still alive. The median PFS and overall S were 9.6 and 20.2 months, respectively. The low toxicity of the bolus part of our regimen prompted us to pursue its intensification by adding a further modulator to MTX. Our recent finding of a strong synergism between bolus FUra and IFN, obtained in vitro on the same tumor model, generated a phase 2 trial employing the same regimen plus IFN (3,000,000 U im q12 h × 4, starting at the time of FUra bolus administration). Among 42 patients 4 CR, 15 PR (RR = 45%), 6 MR and 10 SD were obtained. Since the results of the two trials are similar showing twice as much activity and less toxicity than bolus FUra + LV or MTX → FUra, a randomized comparison is now ongoing between our original hybrid regimen without IFN and MTX → FUra.

Recently, we have demonstrated that thymidylate synthase (TS) protein expression predicts for the clinical response to a regimen of infusional 5-fluorouracil (5FU) in advanced colorectal cancer patients. Previous studies by other groups that showed a correlation between TS gene expression and response to the fluoropyrimidine also involved infusional regimens. Considering the putatively different mechanism of action of bolus compared with continuous infusion of 5FU, the aim of the present study was to test whether the correlation between TS expression and the clinical response to 5FU is valid for bolus regimens. A secondary aim was to compare TS levels between liver metastases and abdominal recurrences from colon cancer, because these sites have a distinctly different responsiveness to 5FU chemotherapy. The study population consisted of 41 patients (25 males and 16 females; median age, 60 years) with unresectable metastatic or recurrent colon cancer, homogeneously treated with 5FU (420 mg/m 2 i.v., days 1–5) and leucovorin (20 mg/m 2 i.v., days 1–5); cycles were repeated every 28 days. Twenty-seven patients (66%) showed high levels of TS expression as defined by TS scores equal to 3 and 4. The proportion of cases with high levels of TS expression was significantly higher in abdominal recurrences (18 of 22, 82%) compared with liver metastases (9 of 19, 47%; P = 0.02). Intratumoral TS protein expression was inversely correlated with response to chemotherapy (response rate: 7 of 14, 50%, versus 0 of 27 in patients with low and high levels of TS expression, respectively; P = 0.0001). These results confirm that the level of TS protein expression predicts for response to 5FU, even with a bolus schedule. The higher TS levels observed in abdominal compared with liver metastases may account for their different responsiveness to 5FU chemotherapy. Immunohistochemical quantitation of TS protein levels may thus allow us to change the therapeutic approach to advanced colorectal cancer from a general to an individual treatment strategy at a time when new non TS-targeted drugs have become available for this disease.

4040 Introduction: Oxaliplatin (OXA) significantly enhances the efficacy of 5-Fluorouracil (FU)-based chemotherapy (CT) in metastatic and radically resected, early stage colon cancer. Weekly OXA, at systemically active doses, can be added to preoperative FU-based pelvic chemoradiation (CRT) (Aschele, Ann Oncol 2005). Methods: An open-label, multicenter, randomized, phase III trial was launched to compare the efficacy (disease-free and overall survival) and activity (pathological response rate) of infused FU (225 mg/msq/day) concomitant to external-beam pelvic radiation (50.4 Gy in 28 daily fractions) (Arm A) to that of the same regimen plus weekly OXA (60 mg/msq weekly x 6) (Arm B). TME surgery is scheduled 6–8 weeks after completing CRT. FU-based adjuvant CT (4 cycles) is planned in both arms. Eligible patients (pts) have resectable tumors located within 12 cm from the anal verge and radiological evidence of perirectal fat or regional nodes involvement. Results: 410 pts (median age 63 years; males/females: 66/34 %; cT3/cT4: 66/18 %; cN+: 56 %; median distance from the anal verge: 6 cm) were randomized between 11/2003 and 10/2006 at 40 Italian centers. Toxicity data from the first 250 randomized pts are reported in the table . CRT was completed by 89 % and 74 % of pts in arm A and B, respectively. The median cumulative dose of delivered radiotherapy was 50.4 Gy in both arms. 120 of 123 pts in arm A and 109 of 112 pts in arm B were operated with a median interval from the end of treatment to surgery of 7.3 and 7.4 weeks, respectively. Conclusions: This analysis provides the first comparative toxicity data showing that OXA, at systemically active doses, can be safely added to standard FU-based CRT with an increased frequency and severity of acute toxicity but without major unexpected adverse events and without affecting the ability to deliver full RT doses and to perform surgery. [Table: see text] No significant financial relationships to disclose.

3770 Background: A combination of OXA and FU is frequently used as 1st or 2nd line therapy for advanced CRC. Anyway, the optimal schedule for this drugs has not been defined. Methods: To identify the MTD of weekly OXA (60–70-80–90 mg/m2 d1,8,15) in combination with a fixed dose of FU CI (200mg/m2/die d1–21), at least 3 pts were treated at each dose level. Once identified the OXA recommended dose level (RDL), Leucovorin (LV; 20 mg/m2 d1,8,15) was added to the combination. Results: Since April 2000, 39 pts with progressive advanced CRC previously treated with fluoropyrimidines and Irinotecan (IRI) were accrued (males/females: 29/10; median age 67; ECOG PS 0/1/2: 10/24/5). Considering the first 2 cycles of treatment (2 months) no grade III-IV toxicity was observed up to 90 mg/m2/week of OXA (dose level 4). 3/6 pts treated at dose level 4 required a major change in the treatment program due to a dose limiting toxicity (DLT; asthenia, weight loss and unbearable peripheral sensory neuropathy) and OXA 90 mg/m2 +...