The authors concluded that longer time to initiation of adjuvant chemotherapy following surgery for colorectal cancer was associated with worse overall and disease-free survival. Potential for bias in the search and unclear study quality mean that some caution is warranted when assessing the reliability of the authors' conclusions.

To determine the impact of time to initiation of adjuvant chemotherapy on survival outcomes in patients with colorectal cancer.

MEDLINE (1975 to January 2011), EMBASE, Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Trials (CENTRAL) were searched. Key terms were reported. Reference lists were scanned. Abstracts from American Society of Clinical Oncology and European Society for Medical Oncology conference proceedings (2007 to 2010) were consulted. Only English-language studies were eligible for inclusion.

Studies of all patients with colorectal cancer treated with adjuvant chemotherapy that reported on the relationship between survival outcomes and the time interval from surgery to initiation of adjuvant chemotherapy were eligible for inclusion. Studies had to have a clearly defined time interval measure. Non-standard forms of adjuvant chemotherapy (such as perioperative chemotherapy) and cross trial comparisons were excluded.

Only studies considered to be high quality were included in the review. Most of the included studies were conducted in USA or Canada; two were UK-based. Mean or median age of patients ranged from 25 to 87 years. Most patients had colon cancer stage II or III. Chemotherapy regimens (where reported) were largely fluorouracil-based. Waiting time ranged from less than one month to at least 16 weeks. Overall survival and disease-free survival were the outcomes of interest. One study reported relapse-free survival. Various representations of waiting time between surgery and adjuvant chemotherapy were reported.

Two reviewers independently selected studies for inclusion. Disagreements were resolved by consensus with a third reviewer.

Study quality appeared to be assessed using an unspecified methodology used previously by the authors. High quality studies were those with relevant, adequately described and balanced prognostic factors between the study groups.

It appeared that two reviewers independently carried out the quality assessment. Disagreements were resolved by consensus with a third reviewer.

Data were extracted to enable the calculation of hazard ratios (HR) and 95% confidence intervals (CI) for overall survival and disease-free survival. Adjusted regression coefficients were calculated to provide a continuous representation of waiting time (per four weeks).

Two reviewers independently extracted data. Disagreements were resolved by consensus with a third reviewer.

Hazard ratios and 95% CIs were pooled in a fixed-effect meta-analysis using inverse variance weighting. Statistical heterogeneity was assessed using X² and I² statistics. Publication bias was assessed in funnel plots and with the Egger test. The trim-and-fill method was applied to adjust for publication bias. Sensitivity analysis was conducted by removal of the largest studies one by one. A separate analysis for cancer-specific survival was conducted.

Ten studies (n=15,422 participants, sample size 179 to 6,059) were included in the review. There were nine cohort studies and one secondary analysis from a randomised controlled trial. Individual quality assessment results were not presented. All 10 studies reported on overall survival; six studies (n=12,522) also reported on disease-free survival.

Statistically significant decreases in overall survival (HR 1.14, 95% CI 1.10 to 1.17) and disease-free survival (HR 1.14, 95% CI 1.10 to 1.18) were associated with increased time interval (per four weeks of waiting time) between surgery and initiation of adjuvant chemotherapy. There was no statistically significant heterogeneity and no evidence of publication bias.

Sensitivity analysis did not materially alter the main findings. A separate analysis of three studies that reported cancer-specific survival was similar to overall survival (HR 1.15, 95% CI 1.10 to 1.19). Other survival data (such as five-year overall survival) were reported.

Longer time to adjuvant chemotherapy was associated with worse survival outcomes in patients with resected colorectal cancer.

The review question was clear and supported by potentially reproducible inclusion criteria. The search strategy included some relevant sources. Published and unpublished studies were eligible, which minimised potential for publication bias. Language bias was a possibility. The review process included sufficient attempts to minimise errors and bias. Some study quality assessment criteria were applied, but details were sparse and the results were not reported clearly, which meant that the authors' interpretation of high standard was the only available measure of quality. Study characteristics were provided, but there was limited detail on the chemotherapy regimens. The chosen method of synthesis was appropriate in the absence of statistical heterogeneity.

The authors' conclusion reflected the evidence presented. The authors acknowledged limitations associated with retrospective data. Potential for bias in the search strategy and unclear quality assessment criteria mean that a degree of caution is warranted when assessing the reliability of this conclusion.

Practice: The authors stated that clinicians should consider timing carefully when discussing adjuvant chemotherapy for patients with resected colorectal cancer.

Research: The authors stated that the timing of chemotherapy should be a strictly controlled variable in future trials of adjuvant treatment.

Canadian Institutes of Health Research.

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.