|
An individual patient data meta-analysis of long supported adjuvant chemotherapy with oral carmofur in patients with curatively resected colorectal cancer |
Sakamoto J, Kodaira S, Hamada C, Ito K, Maehara Y, Takagi H, Sugimachi K, Nakazato H, Ohashi Y |
|
|
Authors' objectives To assess the effect of adjuvant chemotherapy with oral carmofur on survival and disease-free survival in people with curatively resected colorectal cancer.
Searching MEDLINE, the PDQ database and conference abstracts were searched; the search terms and search dates were not reported. The authors also approached personal contacts for additional studies.
Study selection Study designs of evaluations included in the reviewThe review included individual patient data (IPD) from randomised controlled trials (RCTs).
Specific interventions included in the reviewStudies were eligible for inclusion in the review if they compared adjuvant chemotherapy with carmofur versus placebo or no other treatment following short induction therapy, or ad libitum control after curative resection. The studies included in the review used induction therapy of either mitomycin C, 5-fluorouracil, or surgery alone. Three different carmofur regimens were included in the primary studies: 8 mg/kg per day, every day for 12 months; 300 mg/day, every day for 12 months; and 600 mg/day, every day for 3 to 12 months.
Carmofur is a lipophilic masked compound of 5-flurouracil for oral use. It is not available in North America, but is widely used in China, Japan, Korea and Finland.
Participants included in the reviewStudies of people who underwent curative resection for colorectal adenocarcinoma with no severe complications were eligible for inclusion. In two of the three included trials, the participants had to have stage II to III disease and be aged younger than 75 years (74 years in one trial). Studies of people with multiple cancers, marrow or renal impairment, or cardiac or other medical conditions where chemotherapy was contraindicated were excluded, as were those who had undergone prior chemotherapy or radiotherapy. The authors did not report the proportion of women included in the primary studies. All participants in eligible trials, not solely evaluable patients, were included.
Outcomes assessed in the reviewAll of the included studies provided information about survival and disease-free survival. The primary outcomes in the review were survival and disease-free survival, measured at 5 and 7 years.
How were decisions on the relevance of primary studies made?The authors contacted the trial investigators to ensure that only eligible trials were included in the analysis.
Assessment of study quality The authors collated IPD from each trial, regardless of whether the patients were eligible or properly treated to allow an intention-to-treat analysis. They collected information from the original investigators about patient characteristics, follow-up and outcomes, and checked each trial for imbalance between the groups in terms of prognostic factors and the duration of follow-up. The authors did not state how the papers were assessed for validity, or how many reviewers performed the validity assessment.
Data extraction The trial investigators provided IPD including baseline characteristics, disease type, treatment assigned by randomisation, date of recurrence, date of second look operation if any, and date of death or last visit.
Methods of synthesis How were the studies combined?The authors combined the IPD using a meta-analysis. Survival times were analysed using a stratified log rank test and survival curves. Proportions were analysed using the Mantel-Haenszel method. A Cox regression model was used to assess the effect of covariates on survival and disease-free survival. Data from all patients, not just evaluable patients, were included in the analyses.
How were differences between studies investigated?The authors performed formal tests for heterogeneity, but did not report the type of test used. Differences between the studies and tabulated trial characteristics were described, to allow comparisons between the studies. Subgroup analyses were also presented according to the cancer location (colon or rectal cancer) and Dukes' stage. The authors also performed an analysis of evaluable patients to confirm that this did not impact on the findings.
Results of the review Three RCTs with 614 participants were included. IPD were available for all trials.
The addition of oral carmofur to adjuvant chemotherapy following curative resection increased survival and disease-free survival in colorectal cancer.
Disease-free survival (3 trials).
Five-year disease-free survival was 75.6% for the carmofur group and 67% for the controls. Seven-year survival was 71.2% for the carmofur group and 63.9% for the controls. The overall disease-free survival risk ratio (RR) was 0.71 in favour of carmofur (95% confidence interval, CI: 0.53, 0.95, P=0.021). There was no heterogeneity in the treatment effect between studies (P=0.424, Q not reported). The subgroup analysis found that carmofur significantly increased disease-free survival in colon cancer (RR 0.61, 95% CI: 0.39, 0.95, P=0.024), but not in rectal cancer (RR 0.72, 95% CI: 0.47, 1.09, P=0.107). Carmofur increased disease-free survival in people with Dukes' C stage disease (RR 0.48, 95% CI: 0.32, 0.73, P=0.0004), but the effect was not as marked as in Dukes' A and B patients (statistics not reported).
Survival (3 trials).
Five-year overall survival was 79.2% for the carmofur group and 73.1% for the controls. Seven-year survival was 76.1% for the carmofur group and 67.8% for the controls. The survival RR was 0.70 in favour of carmofur (95% CI: 0.51, 0.96, P=0.032). There was no heterogeneity in the treatment effect between studies (P=0.984, Q not reported). Carmofur was associated with a trend towards increased survival in colon cancer, of borderline significance (RR 0.62, 95% CI: 0.38, 1.01, P=0.059). There was no survival benefit in rectal cancer (RR 0.67, 95% CI: 0.43, 1.06, not significant, P-value not reported). Carmofur increased survival in people with Dukes' C stage disease (RR 0.54, 95% CI: 0.35, 0.84, P=0.004), but had no significant effect on survival in people with Dukes' A or B stage disease (statistics not reported).
Authors' conclusions In people with curatively resected colorectal cancer, long-term administration of oral carmofur improved survival and disease-free survival compared with short induction therapy or surgery alone. The authors suggested that while the magnitude of the benefit may be small, it is clinically significant.
CRD commentary The objective of this meta-analysis was clear. The authors specified the research question clearly and described the general inclusion and exclusion criteria. The search strategy appears to have been somewhat limited, although personal contact supplemented database searches. The authors did not provide sufficient information for the reader to feel confident that all eligible trials were sought. It is possible that some studies meeting the inclusion criteria were not identified by the search strategy. Carmofur is most widely used in Asia, yet the authors did not report any special strategies used to find studies published in Chinese or Japanese language journals. It was unclear whether there were any language restrictions.
Appropriately, the authors communicated with the trial investigators to check the eligibility of each trial and the accuracy of the data. It was unclear, however, whether any identified trials were excluded and who made decisions about inclusion and exclusion.
The authors addressed their research question clearly. They reported the findings of each study included in the review and pooled the data appropriately, after checking for differences in prognostic and design factors. However, the authors did not state what test was used to assess heterogeneity. Pooled statistics for survival and disease-free survival were presented without a corresponding time interval.
The authors' conclusions appear to be supported by the data presented. However, these conclusions were based on only three studies, with a total of 614 participants. Caution is advised when generalising these conclusions given: the apparent paucity of data on this topic; the age of the primary studies included in the review (accrual between 1981 and 1990); and the fact that carmofur is not used widely in the western world.
Implications of the review for practice and research Practice: The authors stated that long-term administration of oral carmofur improves survival and disease-free survival over short induction therapy, or surgery alone, in people with curatively resected colorectal cancer. They reported that although the survival benefit from carmofur may be small (about 8%), this could have important clinical implications for people with colorectal cancer.
Research: The authors did not state any implications for further research.
Funding Japanese Society of Strategies for Cancer Research.
Bibliographic details Sakamoto J, Kodaira S, Hamada C, Ito K, Maehara Y, Takagi H, Sugimachi K, Nakazato H, Ohashi Y. An individual patient data meta-analysis of long supported adjuvant chemotherapy with oral carmofur in patients with curatively resected colorectal cancer. Oncology Reports 2001; 8(3): 697-703 Indexing Status Subject indexing assigned by NLM MeSH Administration, Oral; Adult; Aged; Antineoplastic Agents /therapeutic use; Chemotherapy, Adjuvant; Colonic Neoplasms /drug therapy /mortality /surgery; Disease-Free Survival; Female; Fluorouracil /analogs & Humans; Male; Middle Aged; Neoplasm Recurrence, Local; Prognosis; Randomized Controlled Trials as Topic; Rectal Neoplasms /drug therapy /mortality /surgery; Survival Rate; Treatment Outcome; derivatives /therapeutic use AccessionNumber 12001001053 Date bibliographic record published 31/12/2004 Date abstract record published 31/12/2004 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|