This review found that addition of bevacizumab improved progression-free overall survival in patients with advanced colorectal cancer who received first- or second-line fluoropyrimidine-based chemotherapy, but at the expense of increased incidence of toxicity. These conclusions were supported by the data, but should be interpreted with some caution due to the possibility of missing studies and the heterogeneity between studies.

To determine whether the evidence supports the addition of bevacizumab to cytotoxic chemotherapy for patients with locally advanced non-resectable or metastatic colorectal cancer who are considered suitable candidates for systemic therapy.

MEDLINE, EMBASE and The Cochrane Library were searched from 1996 to April 2008 for studies published in English. Search terms were reported. Relevant conference proceedings were screened. Abstracts were eligible, but those that reported only preliminary or interim data were excluded.

Randomised controlled trials (RCTs) that compared chemotherapy plus bevacizumab with chemotherapy alone in the treatment of adult patients with advanced colorectal cancer were eligible for inclusion. Studies had to report data on overall survival, progression-free survival and/or objective response rate.

Treatment regimens assessed in the included studies consisted of IFL (bolus 5-fluorouracil/folinic acid/irinotecan), FOLFOX4 (infusion 5-fluorouracil/bolus folinic acid/oxaliplatin), XELOX (capecitabine/oxaliplatin) and 5-FU/FA (5-fluorouracil/folinic acid) alone (or with placebo) compared to the same drugs combined with bevacizumab. The exact dose and schedule varied between studies. Median follow-up (where reported) was 28 months.

Two reviewers independently assessed studies for inclusion. Disagreements were resolved by consensus.

Study quality was assessed based on criteria of funding source, randomisation method, concealment of treatment allocation, blinding, baseline characteristics, follow-up and type of analysis.

The authors did not state how many reviewers performed the quality assessment.

Hazard ratios (HRs) were extracted directly from the included studies for time-to-event outcomes. Variances of hazard ratios were estimated from confidence intervals or log-rank p-values. Dichotomous data were extracted as number of responses, which were used to calculated risk ratios (RRs).

The authors did not state how many reviewers performed the data extraction.

Summary hazard ratios and risk ratios, together with 95% confidence intervals (CIs), were estimated using inverse-variance weighted random-effects models. Statistical heterogeneity was assessed using X² and I² statistics.

Five RCTs were included. An individual patient meta-analysis based on three of these studies was also identified. Two studies were phase II studies and three were phase III. Four studies were funded by the industry. Method of randomisation and concealment of treatment allocation was adequate in three trials; the other two studies reported insufficient information to make a judgement. Three trials were double blinded and placebo controlled. Outcome assessors were blinded in one trial. One trial was open label.

The addition of bevacizumab to chemotherapy resulted in improved overall survival (HR 0.79, 95% 0.69 to 0.90; four RCTs), progression-free survival (HR 0.63, 95% CI 0.49 to 0.81; four RCTs) and response rate (RR 1.50, 95% CI 1.06 to 2.10; five RCTs). There was significant heterogeneity for progression-free survival (p=0.003) and response rate (p<0.001), but not for overall survival (p=0.14). The individual patient data meta-analysis reported similar results.

The most commonly reported adverse events related to bevacizumab included hypertension, proteinuria, bleeding, and thrombosis. Gastrointestinal perforation and poor wound healing were observed, but their incidence was rare.

For patients with advanced colorectal cancer who received first- or second-line fluoropyrimidine-based chemotherapy, addition of bevacizumab improved progression-free survival and overall survival at the expense of increased incidence of toxicity. The magnitude of benefit may differ based on the chemotherapy regimen with which bevacizumab is partnered.

The review addressed a focused question supported by clearly defined inclusion criteria. The literature search was adequate for published studies. The restriction to published English-language studies risked language and publication biases. Appropriate steps were taken to minimise bias and errors when selecting studies; it was unclear whether such steps were taken when extracting data and assessing quality. Study quality was formally assessed using relevant criteria. Methods used to pool data were appropriate, but there was substantial heterogeneity for two of the analyses and so the results should be interpreted with caution.

The conclusions were supported by the data, but should be interpreted with some caution due to the possibility of missing studies and the heterogeneity between studies.

Practice: The authors stated that bevacizumab should not be administered to patients with cerebral metastases, uncontrollable hypertension, severe proteinuria, advanced atherosclerotic disease, bleeding diatheses and to those with non-healing wounds or recent surgery or trauma.

Research: The authors stated that further RCTs were required to investigate whether bevacizumab, when used as part of first-line therapy, should be continued on progression as part of a second-line regimen.

Cancer Care Ontario; Ministry of Health and Long-Term Care.

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.