This review examined the safety and efficacy of pre-operative aromatase inhibitors compared with tamoxifen in postmenopausal women with hormone receptor positive breast cancer. The authors concluded that aromatase inhibitors improved breast conserving surgery rate, and were not more toxic than tamoxifen. Problems with the review methodology and reporting mean that the reliability of these results is uncertain.

To examine the safety and efficacy of pre-operative aromatase inhibitor compared with tamoxifen in postmenopausal women with hormone receptor positive breast cancer.

Searches of MEDLINE and ISI Web of Knowledge were performed and search terms were listed. These were supplemented by handsearches of the bibliographies of retrieved articles. No language restriction was applied.

Double-blind, phase III, randomised controlled trials (RCTs), published or presented prior to November 2007, which compared a pre-operative aromatase inhibitor with tamoxifen, in postmenopausal women with hormone receptor positive breast cancer, were eligible for inclusion.

The included studies investigated letrozole, anastrazole, or exemestane and treatment duration was three to four months. No further details were presented. Measures of effectiveness were the primary outcomes: clinical objective response rate, ultrasound objective response rate, and breast conserving surgery rate. Toxicities (hot flushes, nausea, headache, and fatigue) were secondary outcomes.

The authors did not state how the papers were selected, nor how many reviewers performed the selection.

The authors did not state that they assessed validity.

Risk ratios (RRs) with 95% confidence intervals (CIs) were calculated. It appears that any adverse event was extracted if it occurred at a rate of 5% of more, in either treatment arm. The authors did not state how many reviewers performed the data extraction.

Studies were combined using fixed-effect meta-analysis, or, in the presence of heterogeneity (p<0.1), a DerSimonian and Laird random-effects model. Heterogeneity between studies was investigated using the Q and I² statistics.

Four studies (1,160 participants) were included in the review; three of these contributed data to the toxicity analyses. Sample sizes ranged from 151 to 451 participants.

Clinical objective response rate: Aromatase inhibitor was associated with a higher response than tamoxifen, RR: 1.29 (95% CI: 1.14 to 1.47). There was significant heterogeneity between studies (p=0.005); the studies of letrozole and exemestane showed beneficial results compared with tamoxifen, whereas the studies of anastrazole did not.

Ultrasound objective response rate: Aromatase inhibitor was associated with a higher response than tamoxifen, RR: 1.29 (95% CI: 1.10 to 1.51). This effect was driven by the exemestane trial, but there was no statistical heterogeneity between studies (p=0.203).

Breast conserving surgery rate: Aromatase inhibitor was associated with a higher rate than tamoxifen, RR: 1.36 (95% CI: 1.16 to 1.59).

There was no significant difference in the rates of hot flushes, nausea, or fatigue in the aromatase inhibitor compared with tamoxifen groups. Headache was more common in the aromatase inhibitor-treated women, RR: 2.02 (95% CI: 1.18 to 3.45).

Pre-operative aromatase inhibitor had a better breast conserving surgery rate than tamoxifen, and was not inferior to tamoxifen in terms of toxicity.

This review addressed a clearly stated research question and used appropriate inclusion criteria. The authors did not specifically state which databases were searched (ISI Web of Knowledge covers five databases), so it is unclear whether the search was comprehensive. There appeared to be no search for unpublished studies, which means that the results could have been affected by publication bias. No language restriction means that language bias did not affect the results. Insufficient details of the methods of study selection were reported to know whether this was done in a way that minimised errors and bias. It appears that the authors did not assess differences between groups for less common adverse events. The results of individual studies were not provided for two of the three primary outcomes. No assessment of study quality was reported. The methods of synthesis were appropriately used and reported.

Although the authors' conclusions were based on their results, the limitations of the methodology and reporting mean that these conclusions should be interpreted with caution.

Practice: Pre-operative aromatase inhibitor was proposed as a replacement for tamoxifen for postmenopausal women, with hormone receptor positive breast cancer, who were not eligible for chemotherapy.

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

Partly funded by the Brain Korea 21 Project and the Korean Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (Grant No. 01-PG3-PG6-01GN07-0004).

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.