The review found that addition of bevacizumab to chemotherapy improved progression-free and overall survival and partial response rate, but increased risks of hypertension in advanced colorectal cancer patients. The conclusions reflect the limited evidence presented, but a lack of reporting of quality, some significant variation and potential for bias in the review make the reliability of the conclusions uncertain.

To assess the efficacy and safety of the addition of bevacizumab to first-line chemotherapy for advanced colorectal cancer.

PubMed was searched up to March 2009. American Society of Clinical Oncology (ASCO), ASCO-Gastrointestinal Symposium, European Society for Medical Oncology (ESMO) and Federation of Cancer Societies (FECS) websites were searched. There were no language restrictions. Search terms were reported. Reference lists of retrieved studies and reviews and lectures from ASCO, ASCO-GI, ESMO and European Cancer Organisation (ECCO) were searched.

Prospective phase II or III randomised controlled trials (RCTs) of previously untreated patients with advanced or metastatic colorectal cancer were eligible. Participants were required to be randomised to either bevacizumab plus chemotherapy or chemotherapy alone, regardless of drug, schedule, dosages and duration. Primary outcomes were progression free survival and overall survival. Secondary outcomes included objective response rate, partial response rate, grade 3 to 4 hypertension rate, grade 3 to 4 bleeding rate and grade 3 to 4 proteinuria rate, if reported in at least 50% of selected trials.

In the included studies, bevacizumab was combined with either irinotecan, 5-fluorouracil and leucovorin or either 5-fluorouracil, leucovorin and oxaliplatin or capecitabine and oxaliplatin. The proportion of women ranged from 39 to 46%. The proportion of participants with the rectum as the primary site (where reported) ranged from 19 to 29%; the proportion with adjuvant treatment (where reported) ranged from 19% to 26%. Other outcomes measured in the trials included duration of response, quality of life, time to treatment failure and time to treatment progression.

The authors did not state how many reviewers selected studies for the review.

The authors did not state whether studies were assessed for quality.

Data were extracted and hazard ratios (HRs), together with 95% confidence intervals (CIs) were calculated for primary outcomes and log relative risks (RRs) and 95% CIs calculated for secondary outcomes. For each study, data were collected at the last available update.

Five reviewers separately extracted data from the included studies.

Studies were pooled in meta-analyses and summary estimates (HRs or RRs) with 95% confidence intervals calculated using a random-effects model (inverse variance and Mantel-Haenszel). Absolute benefits were derived using a formula modified by Parmar. The number of patients needed to treat for one single patient to benefit (NNT) and the number of patients needed to treat to cause a significant harm in a patient (NNH) were calculated. Heterogeneity was assessed with X². Meta-regression was undertaken to assess the influence of negative prognostic factors in the studies. Sensitivity analysis was performed to assess possible differential effects of trial phase-design.

Five RCTs (2,728 participants) were included in the review. Three studies were phase III trials and two studies were phase II trials.

Primary outcomes: Compared to chemotherapy alone, addition of bevacizumab to chemotherapy was associated with significantly increased progression-free survival (HR 0.62, 95% CI 0.48 to 0.69, absolute difference 17.1%; NNT 6; significant heterogeneity; four RCTs) and significantly increased overall survival (HR 0.78, 95% CI 0.66 to 0.94, absolute difference 8.6%, NNT 12; four RCTs) without significant interaction according to study setting (phase II or III).

Secondary outcomes: Compared to chemotherapy alone, addition of bevacizumab to chemotherapy was associated with significantly increased partial response rate (RR 1.24, 95% CI 1.06 to 1.46, absolute difference 6.5%; NNT 15; no significant heterogeneity; four RCTs). There was no evidence of a significant difference in objective response rate between groups in five studies (although a heterogeneous trend favoured the addition of bevacizumab).

Compared to chemotherapy alone, addition of bevacizumab to chemotherapy was associated with a significantly increased risk of hypertension (RR 4.87, 95% CI 3.12 to 7.61; absolute difference 6.2%; NNH 16; no significant heterogeneity; five RCTs). There was no evidence of a significant difference between groups in risk of bleeding or proteinuria (although a trend against the addition of bevacizumab was found for risk of proteinuria).

Female gender and rectal primary site were significant predictors for progression-free survival benefit.

Addition of bevacizumab to conventional chemotherapy improved progression-free and overall survival and partial response rate, but increaseds the risk of hypertension in advanced colorectal cancer patients.

The review addressed a clear research question and inclusion criteria appeared appropriate. A range of relevant sources was used to identify studies; there were no language restrictions and attempts were made to find unpublished studies. Data were extracted by multiple reviewers; the authors did not state how many reviewers selected studies for the review, so reviewer error and bias could not be excluded. It did not appear that studies were assessed for quality, which made it difficult to determine the reliability of the results.

Trials were synthesised in meta-analyses. Heterogeneity was appropriately assessed. Meta-regression was used to explore the influence of prognostic factors on the results and sensitivity analysis was used to determine whether trial setting (as represented by trial design of phase II or phase III) influenced the findings.

The authors' conclusions reflect the limited evidence presented, but a lack of reporting of quality, some significant heterogeneity and potential for bias in the review processes make the reliability of the conclusions uncertain.

Practice: The authors stated that, in spite of costs and significant risk of hypertension, addition of bevacizumab to chemotherapy should be considered as first-line therapy for unselected advanced colorectal cancer patients.

Research: The authors stated that further head-to-head phase III RCTs should compare anti epidermal growth factor receptor cetuximab with vascular endothelial growth factor antibody bevacizumab. Research should be undertaken to confirm whether female gender and rectal primary site were significant predictors of progression-free survival benefit with combined bevacizumab and chemotherapy.

National Ministry of Health, Italy; Italian Association for Cancer 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.