The authors concluded that the addition of bevacizumab to chemotherapy improved survival and response rates, and that adverse effects were predictable, manageable and did not compound chemotherapy toxicity. In view of the questionable quality of the primary studies and methodological problems in the review (particularly the failure to address heterogeneity between the studies), these conclusions should be regarded with caution.

To evaluate the safety and efficacy of bevacizumab combined with chemotherapy for treating metastatic colorectal cancer (mCRC).

PubMed, EMBASE and The Cochrane Library were searched for articles published between 2003 and August 2008. Search terms were reported. Reference lists of articles retrieved were handsearched.

Randomised controlled trials (RCTs) that compared combined chemotherapy with bevacizumab for patients with metastatic colorectal cancer were eligible for inclusion, provided controls received the same chemotherapy regimen without bevacizumab. Outcomes of interest were overall survival, progression-free survival or median time to progression, overall response rate and toxicity. Progression-free survival and time to progression were defined as the time to first documented progression or death; overall response rate was defined as the sum of complete and partial responses (according to the Response Evaluation Criteria in Solid Tumours) and toxicity was graded by the Common Toxicity Criteria version two.

The mean or median age of participants in the included studies ranged from 60 to 71 years (where stated). About 40% to 56% of participants were women. Most participants had Eastern Cooperative Oncology Group Performance status of 0 or 1. Some 36% to 69% had more than one metastatic site (where stated). Prior cancer therapy included radiation therapy in 14% to 26% of cases, adjuvant chemotherapy in 14% to 28% of cases and surgery in 80% to 97% of cases (where stated). Chemotherapy comprised various combinations of oxaliplatin, fluorouracil, leucovorin, irinotecan and capecitabine. In most studies the control group received a placebo in place of bevacizumab. Both overall adverse event rates and specific adverse events were reported.

The authors stated neither how the papers were selected for the review nor how many reviewers performed the selection.

Study quality was evaluated using the Jadad scale for measures adequacy of randomisation, blinding and management of withdrawals and dropouts. Each study was awarded a score out of a maximum of 5 points. The authors did not state how the assessment was performed.

Hazard ratios (HRs), risk ratios (RRs) and odds ratios (ORs), together with their 95% confidence intervals (CIs), were calculated from the rates of events in the control and intervention groups of each study. If necessary, the methods of Parmar were used to estimate log HRs and their variance. Data were analysed by intention to treat. Data that related to a bevacizumab-only arm in one RCT were excluded from analysis. Data were extracted independently by two reviewers without disagreement.

Studies were combined using a fixed-effect model unless there was significant heterogeneity (p<0.1), in which case a DerSimonian and Laird random-effects model was used. Publication bias was assessed using the Egger weighted regression test. The impact of individual studies on the overall results was examined by omission from analysis of each study in turn.

Five RCTs were included in the review (n=3,103, range 104 to 1,400). The main limitation was failure to report adequate allocation concealment. The mean Jadad score was 2 out of 5. Three RCTs described their method of randomisation, two reported withdrawals and dropouts and none were double blinded. Participant characteristics appeared to be well-balanced between the groups in four RCTs.

The intervention was associated with a significantly lower risk of progression (HR 0.66, 95% CI 0.56 to 0.77, p<0.01; five RCTs), a significantly lower risk of death (HR 0.77, 95% CI 0.67 to 0.89, p<0.01; four RCTs) and significantly higher likelihood of response (RR 1.50, 95% CI 1.06 to 2.10, p=0.02; five RCTs). However, for all these analyses there was statistically significant heterogeneity, which in each case was attributed mainly to the same RCT. There was no statistically significant difference between the groups in all-cause mortality at 60 days (four RCTs).

In the intervention group, there were significantly more grade 3 or 4 adverse events (OR 1.79, 95% CI 1.52 to 2.11, p<0.01; five RCTs) and a significantly higher risk of discontinuing treatment due to adverse effects (OR 1.38, 96% CI 1.14 to 1.66, p<0.01; four RCTs), with no significant heterogeneity. The intervention was associated with significantly higher rates of grade 3 or 4 hypertension (five RCTs), thromboembolic/thrombotic events (three RCTs), bleeding (four RCTs), and gastrointestinal perforation (four RCTs). There was no statistically significant difference between the groups in rates of proteinuria, diarrhoea or leukopenia.

No obvious publication bias was detected.

The addition of bevacizumab to chemotherapy improved overall and progression-free survival and overall response rates. Adverse effects were predictable, manageable and did not compound chemotherapy toxicity.

The objectives and inclusion criteria of the review were clear and relevant sources were searched for studies. The apparent restriction to published papers meant that some studies may have been missed, but testing for publication bias showed no evidence of this. It was not stated whether the search was limited by language. Steps were taken to minimise the risk of reviewer bias and error by having more than one reviewer independently extract data, but it was not clear whether similar precautions were taken in the processes of study selection and validity assessment. Some relevant details about the primary studies (such as duration of follow-up) were not reported. Appropriate statistical techniques were used to combine the data and assess for publication bias, but the statistical test used to assess heterogeneity was not reported. Where significant statistical heterogeneity was detected, it was attributed mainly to a specific study, but no explanation for the difference in findings was suggested; the study in question did not appear to be an outlier on the forest plots. The authors’ conclusions about the manageability of adverse events associated with bevacizumab did not appear well supported by data presented in the review. In view of the questionable quality of the primary studies and methodological problems in the review (particularly the failure to address heterogeneity between the studies), the authors’ conclusions should be regarded cautiously.

Practice: The authors stated that risks associated with bevacizumab should be considered in balance with the benefit of increased survival. In patients who received bevacizumab, clinicians needed to carefully follow published advice for managing hypertension and proteinuria and monitor for signs of gastrointestinal perforation.

Research: The authors did not state any implications for further research.

Not stated.

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.