Difference In Overall Survival Curves Research Articles

Abstract 6519: Dendritic cell vaccine cell numbers and survival in patients with primary glioblastoma

Abstract Standard glioblastoma (GBM) treatment is associated with poor survival. A promising novel immunotherapy consists of autologous dendritic cells (DC) pulsed with autologous tumor antigens (ATA) from a lysate of irradiated self-renewing autologous cancer cells. An open-label, phase 2 trial was conducted in 57 patients with primary GBM. Here we report correlations between survival and DC-ATA doses. Two autologous intermediate products were manufactured for each patient: a tumor cell line established from resected GBM, and DC differentiated from cryopreserved peripheral blood monocytes (MC). Patients were enrolled prior to standard radiation therapy (RT) and temozolomide (TMZ) chemotherapy. DC-ATA were manufactured during RT/TMZ, divided into individual doses, then cryopreserved in liquid nitrogen. For each patient-specific vaccine batch, cell numbers were calculated using a hemocytometer; viability was determined before cryopreservation (cryo) using trypan blue and post-cryo using 7ADD per flow cytometry. After RT/TMZ, up to 8 doses were injected s.c. concurrent with TMZ over 6 months. Patients were grouped into tertiles based on DC-ATA cell numbers and survival. Prism 10.03 was used to generate survival curves and calculate correlation coefficients and log-rank and T-Test comparisons. As previously reported, cell line success rate was 71/73, sufficient MC numbers were collected by leukapheresis for 63/65; DC-ATA was successfully manufactured for 60/60; 57 patients were treated; median progression free survival (PFS) and overall survival (OS) were 10.4 and 16.0 months respectively. Correlations were strong between total and viable DC-ATA pre-cryo (0.97, p<0.0001) and post-cryo (0.95, p<0.0001), but weaker between DC-ATA pre- and post-cryo (0.30, p=0.02), and between viable DC-ATA pre- and post-cryo (0.37, p=0.004). When patients were grouped by OS tertiles, medians were 9.8, 18.1, and more than 36 months. There were no differences in the average millions (M) of DC-ATA pre-cryo (10.0, 9.7, 9.6), viable DC-ATA pre-cryo (8.5, 7.8, 7.9), DC-ATA post-cryo (9.0, 10.3, 10.8) or viable DC-ATA post-cryo (6.0, 7.3, 7.4). Similarly, when patients were grouped into tertiles by DC-ATA numbers, there were no differences in OS curves associated with DC-ATA pre-cryo, viable DC-ATA pre-cryo, DC-ATA post-cryo, or viable DC-ATA post-cryo. Pre-cryo viable DC-ATA numbers ranged from 0.27 to 27.0 M. An ineffective dose was not identified; pre-cryo viable DC-ATA doses of less than 2M were associated with OS of 10.4, 13.0, 13.3, and 36.4+ months. Also, there was no OS difference by number of MC used to manufacture DC or by number of DC manufactured. In conclusion, there was no evidence for a dose/response relationship between DC-ATA dose and OS for any of the parameters tested. This data supports that pre-cryo cell counts can be used to establish treatment doses. Pre-cryo viable cell counts of 2M DC-ATA per dose seem sufficient. Citation Format: Robert O. Dillman, Gabriel I. Nistor, Krystal Godding, Rockelle Robles, Hans S. Keirstead. Dendritic cell vaccine cell numbers and survival in patients with primary glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6519.

Prognostic Significance of Esophagogastric Junction Invasion in Patients with Adenocarcinoma of the Cardia or Subcardia.

There has been no comparison of the prognoses of Korean patients who underwent curative surgery for cancer located at the cardia or subcardia of the stomach. We performed this comparison and further investigated the prognostic significance of esophagogastric junction (EGJ) invasion in patients. The medical records of patients (n = 511) who were diagnosed with cardia or subcardia cancer and underwent surgery between January 2010 and May 2019 were retrospectively reviewed. Patients were further categorized into four groups for analysis: subcardia gastric cancer (sGC; subcardia cancer without EGJ invasion; n = 97), AEG (adenocarcinoma of the esophagogastric junction) type III (subcardia cancer with EGJ invasion, n = 54), AEG type II without EGJ invasion (n = 158), and AEG type II with EGJ invasion (n = 202). We compared the overall survival of the four groups using a gastric cancer staging system and evaluated the prognostic significance of EGJ invasion with multivariate analysis. The median follow-up of patients was 46.0 months (range: 0-124 months). There was significant difference in overall survival curves among the four groups (p < 0.001). Subgroup analysis showed a significant difference in overall survival between the groups with and without EGJ invasion (p < 0.001). Cancers with EGJ invasion were more frequently in the cardia (p < 0.001), had a larger size (p < 0.001), and showed a more advanced pathologic stage (stages II and III; 67.6% versus 33.7%, p < 0.001) than those without EGJ invasion. EGJ invasion and the pathologic stage were significant independent prognostic factors of overall survival in cardia and subcardia cancer patients (hazard ratio 2.24, 95% confidence interval 1.32-3.81, p = 0.003). The overall survival between patients with cardia or subcardia cancer was significantly different according to EGJ invasion. EGJ invasion was an independent prognostic factor and should be considered for staging. Additional research is needed to apply this feature to gastric and esophageal cancer classification.

The C-Reactive Protein to Albumin Ratio Is an Independent Prognostic Factor in Patients with Hepatocellular Carcinoma Undergoing Transarterial Chemoembolization: A Large Cohort Study.

This study aimed to investigate the prognostic value of C-reactive protein to albumin ratio (CAR) in hepatocellular carcinoma (HCC) patients after transcatheter chemoembolization (TACE). Totally, 958 HCC patients with Barcelona Clinic Liver Cancer (BCLC) stage B were incorporated into the secondary analysis. X-Tile software was applied to determine the optimal cutoff point for CAR, and the total patients were divided into two groups. Cox proportional hazard regression models and Kaplan-Meier analyses were used to estimate the relationship between CAR and overall survival (OS). Stratified analyses were performed to evaluate the prognostic role of CAR in subgroups of major confounding factors, such as alpha-fetoprotein (AFP), diameter of the main tumor, Glasgow prognostic score (GPS) and modified GPS (mGPS). The optimal cutoff level for the CAR was 0.06. There was a direct correlation between an elevated CAR (≥ 0.06) and shorter OS after adjustment (HR:1.580; 95%CI:1.193-2.092). Kaplan-Meier analysis and log-rank test showed a significant difference in OS curves between the two groups (P < 0.001). CAR showed the distinct value of prognostic stratification in most subgroups, especially in the subgroup of GPS-0 (HR:1.966; 95%CI:1.453-2.660), mGPS-0 (HR:1.984; 95%CI:1.509-2.608) and AFP ≤ 400ng/ml (HR:1.925; 95%CI:1.393-2.659). The CAR was one of the prognostic factors for HCC patients undergoing TACE treatment. CAR could also provide further prognostic stratification for HCC patients who appear to have a good prognosis, such as patients with AFP-negative, GPS-0 or mGPS-0 to identify patients at a higher risk of death for closer follow-up or more aggressive treatment. Level 3, Cohort Study.

Clinical Significance of the C-Reactive Protein to Albumin Ratio for Survival After Surgery for Colorectal Cancer.

This study was designed to estimate the clinical significance of the C-reactive protein (CRP)/albumin ratio (CAR) for prediction of postoperative survival in patients with colorectal cancer (CRC). The Glasgow Prognostic Score (GPS), calculated from the serum levels of CRP and albumin, is well known to be a valuable inflammation-based prognostic system for several types of cancer. A recent study has demonstrated that the CAR is also useful for prediction of treatment outcome in patients with hepatocellular carcinoma. Uni- and multivariate analyses using the Cox proportional hazards model were performed to detect the clinical characteristics that were most closely associated with overall survival (OS). All recommended cutoff values were defined using receiver operating characteristic curve analyses. Kaplan-Meier analysis was used to compare OS curves between the two groups. A total of 627 patients who had undergone elective CRC surgery were enrolled. Multivariate analysis using the results of univariate analyses demonstrated that CAR (>0.038/≤0.038) was associated with OS (hazard ratio 2.596; 95% confidence interval 1.603-4.204; P < 0.001) along with pathological differentiation (others/well or moderately), carcinoembryonic antigen level (>8.7/≤8.7, ng/ml), stage (III, IV/0, I, II), neutrophil to lymphocyte ratio (NLR) (>2.9/≤2.9), and GPS (2/0, 1). Kaplan-Meier analysis and log rank test demonstrated a significant difference in OS curves between patients with low CAR (≤0.038) and those with high CAR (>0.038; P < 0.001). CAR is as useful for predicting the postoperative survival of patients with CRC as previously reported inflammation-based prognostic systems, such as GPS and NLR.

Prognostic Value of Lymphocyte Vascular Density and E-Cadherin in Inflammatory Breast Cancer

Background: We recently evaluated four laboratory assays, vascular endothelial growth factor D (VEGF-D), E-cadherin, lymphatic vessel density (LVD) measured by podoplanin, and intra-lymphatic tumor emboli (ILTE), which showed notable differences between inflammatory breast cancer (IBC) and non-inflammatory locally advanced breast cancer (LABC). In this study we investigated the potential of the three most quantitatively measured markers, E-cadherin, LVD and VEGF-D, to predict survival in the IBC patients. Materials and Methods: This study involved the 100 cases identified in the Inflammatory Breast Cancer Registry (IBCR) whose tumors were previously evaluated for the four assays noted above. Living patients were recontacted and survival data were available for up to 17 years. Overall survival (OS) was analyzed through the Kaplan-Meier method stratified by E-cadherin, LVD, VEGF-D, and response to chemotherapy. The differences in OS curves were compared using the log-rank test. Results: The median OS for patients with high LVD was 6.63 years (95% CI: 4.06 to 10.14), compared to median at 10 years not reached in those with low LVD (p = 0.03). There was a trend towards a longer median OS in patients with high E-cadherin (10.14, 95% CI: 6.63 to 11.67), compared with those with low E-cadherin (6.26, 95% CI: 3.42 to undeterminable). VEGF-D levels showed no correlation with survival. Conclusion: Low LVD significantly predicts better survival. High E-cadherin expression, as with non-IBC breast cancer and several other malignancies, tends to be associated with a better prognosis.

Dynamics and Prognostic Impact of Peripheral Blood Blast Clearance in Patients with Acute Myeloid Leukemia (AML) Receiving FLT3 Inhibitor Therapy in Combination with Induction Chemotherapy

Abstract 1417 Background:We have previously shown that for patients with acute myeloid leukemia (AML) treated with induction chemotherapy with cytarabine (ara-C) plus an anthracycline, the day of blast clearance from peripheral blood (PB) is a powerful prognostic marker. Earlier PB blast (but not white blood cell [WBC]) clearance portends improved overall survival (OS). We expanded our investigation to include patients with AML undergoing induction chemotherapy with ara-C (1.5g/m2 ×3 days) plus idarubicin (12mg/m2×3 days) that is, the AI regimen, plus sorafenib (400mg orally twice daily). Patients and Methods:We reviewed the clearance of PB blast and WBC (PB blasts = 0%, WBC≤0.1×109/dL), for patients with AML (except APL) undergoing AI alone (n=168) or AI+sorafenib (n=75). Patient characteristics for the AI alone group (n=168) were as follows: median age 55 years (range, 19–72), diploid cytogenetics (n=57, 34%), poor cytogenetics (n=61, 36%), FLT3-ITD positive (n=15, 9%), FLT3-negative (n=122, 73%), median WBC 5.0×109/dL (range, 0.3–132.3), median platelets 37×109/dL (range, 1–581), median hemoglobin 9.1g/dL (range, 4.0–13.2), median PB blasts 15% (range, 1–96), median BM blasts 42% (range, 1–96). Patient characteristics for the AI+sorafenib group (n=75) were as follows: median age 52 years (range, 18–66), diploid cytogenetics (n=38, 51%), poor cytogenetics (n=10, 13%), FLT3-ITD (n=25, 33%), FLT3-negative (n=48, 64%) median WBC 5.9×109/dL (range, 0.6–228.5), median platelets 51×109/dL (range 7–306), median hemoglobin 9.2g/dL (range, 7.4–12.6), median PB blasts 21% (range 0–98), median BM blasts 56% (range 6–98). Results:The overall response rate (ORR=CR+CRp) in the AI group was 63% (58%+5%) and 50 patients (30%) were resistant to therapy. The ORR in the AI+sorafenib group was 79% (72%+7%) and 11 patients (15%) were resistant. We analyzed OS by dividing patients based on the day of PB blast clearance: group 1 (0–1 days), 2 (2–3 days), 3 (4–5 days), 4 (6–8 days), and 5 (>8 days). For the patients receiving AI induction therapy, a total of 32, 59, 38, 18, and 2 patients were included in groups 1–5, and median OS for groups 1–4 were 167, 48, 67, and 25 weeks respectively. A logrank test comparison revealed these four OS curves were significantly different (p-value=0.0015). For the AI treatment group, earlier blast disappearance corresponded with better OS. In contrast, for patients in the AI+sorafenib group, the correlation between day of PB blast clearance and OS was less clear. A total of 10, 23, 25, 6, and 3 patients were included in groups 1–5. The median OS for groups 1, 3, and 4 were 101, 74, and 66 weeks respectively, and the median survival for group 2 could not be estimated because the curve did not fall below 0.5. The difference between these four survival curves was not significantly different (p-value=0.35). Survival differences for the AI group were more clearly demarcated when patients were divided based on blast disappearance within 0–5 days or >5 days (p-value=0.0004). A similar analysis in the AI+sorafenib group revealed no significantly different OS (p-value=0.13). Among patients in the AI+sorafenib treatment group, the mean day of blast disappearance for FLT3-ITD versus FLT3-negative patients was 5.2 vs. 4.5 days, and this difference was not statistically significant (p-value=0.33). While FLT3-ITD patients who cleared their blasts within 0–3 days (n=10) tended to have better OS survival than FLT3-ITD patients who cleared their blasts after 3 days (n=13), the difference in OS curves did not reach significance (p-value=0.18). Conclusion:The day of PB blast clearance is prognostic among patients receiving classic induction chemotherapy, where early clearance predicts better long-term outcomes. The addition of a targeted agent to a standard induction regimen limits the prognostic power of early PB blast clearance. We are currently investigating a novel mathematical approach to evaluate the prognostic value of clearance of peripheral blasts compared to WBC after initiation of therapy in patients receiving targeted agents during induction therapy; the approach aims to integrate the prognostic impact of FLT3 and NPM1 in combination with cytoreduction kinetics to better identify prognostically distinct groups. Disclosures:Off Label Use: Plerixafor plus G-CSF as a part of conditioning in allogenic transplant for AML/MDS.

878 - The Role of Nephrectomy in Metastatic Renal Cell Carcinoma - A Single Centre Analysis

ABSTRACT The approval of new therapeutic options has increased the survival of metastatic renal cell carcinoma (mRCC) patients (pts). These drugs were mostly studied in pts who underwent nephrectomy (Sx) prior to the diagnosis of metastatic disease (MD). Sx is also indicated for pts with MD who are suitable and have good performance status (PS). The aim of this work is to retrospectively analyse the benefit of Sx in the pts with mRCC referred to the Oncology Division at Santa Maria Hospital from 01/2006 to 10/2010. Two groups of pts were defined: pts who had Sx and pts who had no Sx. The first group was further divided into pts who had Sx whilst with non-MD (SxNMD) and pts who did it whilst with MD (SxMD). With significance set at p = 0.05 and the aid of IBM® SPSS® Statistics 20, statistical analysis was performed using descriptive statistics and Fisher's exact test. Overall survival (OS) - time from diagnosis of MD to death - was plotted using the Kaplan–Meier method and compared by log-rank test. The population had 44 pts: 72,7% (n = 32) male, 59,1% (n = 26) under 65 years old, 90,9% (n = 40) with Karnofsky PS (KPS) ≥80 and 59,1% (n = 26) with clear cell carcinoma. Half were diagnosed with non-MD (NMD) (n = 22) and progressed to MD in a mean of 45,9 months (m). Thirty-eight pts (86,4%) did medical therapy, of those 86,8% (n = 33) had favourable or intermediate risk according to the Memorial Sloan Kettering Cancer Center (MSKCC) criteria and 86,8% (n = 33) did target therapy. Thirty pts underwent Sx (68,2%) and 14 didn't (31,8%). There was no statistical difference between groups regarding gender, age, KPS, histology, MSKCC risk and medical therapy. The median OS was 14 and 27 m respectively for the no Sx group versus the Sx group (p = 0,027). In the pts who had Sx, 21 (70%) were part of the SxNMD group and 9 (30%) of the SxMD group. Except for Fuhrman grade (p = 0,022), there was no statistical difference between groups regarding gender, age, KPS, histology, MSKCC risk and medical therapy. The median OS for SxNMD had not been reached while the median OS for SxMD was 21 m (p = 0,682). In the population studied there was no difference in OS curves between the SxNMD group and the SxMD group. This suggests that Sx has a valuable role in pts diagnosed with MD in addition to medical therapy, namely target therapy. Disclosure All authors have declared no conflicts of interest.

Changes in survival during the past two decades for breast cancer at the Kyoto University Hospital

Aims To report the changes in survival over 20 years of 775 breast cancer women operated between 1982 and 2003 at the Kyoto University Hospital in Japan, reflecting changes in clinical practice over that period. Results Survival curves have significantly improved between the periods 1982–1989 and 1990–2003. The 5- and 10-year survival rates between these periods were 80.3% and 85.1%, and 67.5% and 75.0%, respectively. Moreover, there was a difference in overall survival curves of patients of stages II and III, of 35–54 ages, or of positive estrogen receptor (ER) status between these periods. Conclusion The present study presented the recent advance of the survival rates might be due to the rational development of breast cancer treatment, and suggested the possibility that the patients of stages II and III, of 35–54 ages, or of positive ER status were received benefits by these treatments.

Continuation Electroconvulsive Therapy vs Pharmacotherapy for Relapse Prevention in Major Depression

Although electroconvulsive therapy (ECT) has been shown to be extremely effective for the acute treatment of major depression, it has never been systematically assessed as a strategy for relapse prevention. To evaluate the comparative efficacy of continuation ECT (C-ECT) and the combination of lithium carbonate plus nortriptyline hydrochloride (C-Pharm) in the prevention of depressive relapse. Multisite, randomized, parallel design, 6-month trial performed from 1997 to 2004. Five academic medical centers and their outpatient psychiatry clinics. Two hundred one patients with Structured Clinical Interview for DSM-IV-diagnosed unipolar depression who had remitted with a course of bilateral ECT. Random assignment to 2 treatment groups receiving either C-ECT (10 treatments) or C-Pharm for 6 months. Relapse of depression, compared between the C-ECT and C-Pharm groups. In the C-ECT group, 37.1% experienced disease relapse, 46.1% continued to have disease remission at the study end, and 16.8% dropped out of the study. In the C-Pharm group, 31.6% experienced disease relapse, 46.3% continued to have disease remission, and 22.1% dropped out of the study. Both Kaplan-Meier and Cox proportional hazards regression analyses indicated no statistically significant differences in overall survival curves and time to relapse for the groups. Mean +/- SD time to relapse for the C-ECT group was 9.1 +/- 7.0 weeks compared with 6.7 +/- 4.6 weeks for the C-Pharm group (P = .13). Both groups had relapse proportions significantly lower than a historical placebo control from a similarly designed study. Both C-ECT and C-Pharm were shown to be superior to a historical placebo control, but both had limited efficacy, with more than half of patients either experiencing disease relapse or dropping out of the study. Even more effective strategies for relapse prevention in mood disorders are urgently needed.

Analysis of Survival, AML Transformation, and Transfusion Independence in Patients with High-Risk Myelodysplastic Syndromes (MDS) Receiving Azacitidine Determined Using a Prognostic Model.

The International Prognostic Scoring System (IPSS) requires cytogenetic data to evaluate risk in MDS. Cytogenetic data were not available for all patients from the CALGB trial 9221 (Silverman et al, JCO 2002; 20:2429). Therefore, we developed an alternative prognostic model that identified a homogeneous subgroup of high-risk MDS patients based on the following risk factors: percent marrow blasts, number of cytopenias, age, gender, FAB classification, and time since diagnosis (IPSS used data at time of diagnosis; CALGB 9221 used data at time of randomization). The model was validated using 2,318 patients from the MDS Registry, Düsseldorf. Using these baseline prognostic factors, we predicted survival for each of the 191 CALGB patients, identified a high-risk subgroup with a survival prognosis of ≤ 1.2 years (the IPSS INT-2 survival median), and compared azacitidine to supportive care. Patients were analyzed as randomized (azacitidine or supportive care) according to the intention-to-treat (ITT) principle. The two groups had similar demographic and disease characteristics. All 70 high-risk patients were followed until death. There was a statistically significant difference in overall survival curves between the azacitidine and supportive care groups (p=0.03). The one-year survival rate was 63% (95% CI: 47 to 78%) in the azacitidine group and 37% (95% CI: 19 to 54%) in the supportive care group (p=0.03; 26% difference with a 95% CI of 3 to 49%). The two-year survival rate was 35% (95% CI: 20 to 50%) in the azacitidine group and 13% (95% CI: 1 to 25%) in the supportive care group (p=0.03; 22% difference with a 95% CI of 3 to 41%). Similarly, statistically significant differences in time to AML transformation, and death or AML transformation, were observed. Transfusion independence was defined as free from transfusions for at least 2 months. Among patients who were RBC transfusion dependent at baseline, a significantly greater number of patients from the azacitidine group (11/25, 44%) compared with patients from the supportive care group (1/14, 7%) achieved transfusion independence (p=0.03; 37% difference with a 95% CI of 7 to 59%). Additionally, patients in the azacitidine group with baseline transfusion independence experienced significantly prolonged duration of RBC (p<0.0001) and platelet (p=0.0002) transfusion independence compared with those in the supportive care group. Use of this prognostic model has identified a subgroup of high-risk MDS patients who significantly benefited from treatment with azacitidine. [Display omitted]

Significance of elevated serum lactate dehydrogenase (LDH) in patients undergoing salvage autologous bone marrow/peripheral stem cell transplantation (BM/PSCT) for relapsed/refractory Hodgkin’s lymphoma (R/R-HL)

6734 Background: High-dose chemoRx→ BM/PSCT is an acceptable salvage for pts with R/R-HL. In a reported single institution(Bierman PJ.et al. Ann Oncol. 2002;13:1370), low albumin (<4g/dl), anemia (<10.5g/dl), age (≥45 y) and lymphocytopenia (<600/mm3 and/or<8% of WBC) were associated with inferior event-free and overall survival (OS) in pts with HL undergoing auto BM/PSCT. Elevated LDH levels, as in the IPI and age-adjusted index, adversely impact outcome in pts with aggressive NHL. The significance of elevated LDH in pts with R/R-HL s/p salvage auto BM/PSCT is less clearly defined and controversial. Objective: Evaluate a potential correlation between OS and elevated LDH levels (at transpl) in pts s/p salvage auto BM/PSCT for R/R-HL. Methods: Retrospective, single-institution of consecutive pts with R/R-HL who received salvage auto-BM/PSCT over 14 y (Nov 01,90-Nov 30,04). LDH level (U/L) at time of BM/PSCT (day 0±1) was noted. Pts were divided as normal (A) and high-LDH group (B), if LDH were <1x or ≥1x normal (institution normal:<190 U/L), respectively. OS was defined as the time-period in months from day of transpl to death, until last day of f/u, or until last day of study, for pts who died, were lost to f/u, or still alive, respectively. Survival curves were estimated according to the Kaplan-Meier method, and compared between the 2 groups with the log-rank test. One-sided P-value of <0.05 was significant. Results: 49 pts (33 men, 16 women) were identified, who were 13–69 yrs (median: 32 yrs) at time of transpl. 30 pts (61%) died during f/u, 12 pts (25%) were lost to f/u at 2 to 84 mos (median: 25 mos) after transpl, and 7 pts (14%) are alive at time of analysis, 7–75 mos (median: 40 mos) after transpl. LDH at transpl was not available in 5 pts. Median OS for A (n1=26, 15/26 [58%]died) was 15.6 mos (95% CI: 8.4–69.4 mos) and for B (n2=18, 13/18 [72%] died) was 33.7 mos (95% CI: 18.1–58.7 mos). The difference in OS curves was not significant (p=0.601,log-rank). Conclusion: Our results show that elevated serum LDH levels don’t correlate with inferior OS in pts with R/R-HL s/p salvage auto BM/PSCT. A prospective study is warranted. No significant financial relationships to disclose.

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.