OBJECTIVES: Staging investigations in newly diagnosed breast cancer (BC) women are important to decide management since the presence of distant metastases negatively affects survival. We aim to evaluate the value of staging investigation strategies stratified by T (tumour-size) and N (node) stage of disease.

METHODS: We performed a search in Medline, EMBASE, Cochrane Library, and NHS-Economic Evaluation Database (May 2017) to identify studies: a) reporting true distant metastasis and false positive results with conventional imaging (CI) tests (bone scan, X-Ray, computed-tomography, ultrasound) or 18FDG-PET/CT, or b) cost-effectiveness analysis. We only included studies reporting data during the perioperative period from asymptomatic or consecutively diagnosed BC patients, and presenting results stratified by stage (TN: I to III). We pooled proportions using a binomial analysis and a random effect assumption.

RESULTS: Eleven studies (14,709 patients) reported CI tests performance; for combined tests (bone scan plus any other test), the distant metastases detection rate was 0.9% (95% CI: 0% to 3.2%) with stage I breast cancer, 3.0% (95% CI: 1.8% to 4.5%) with stage II, and 15.7% (95% CI: 12.5 to 19.2%) with stage III. The rates of false positive results ranged from 5.5% to 8.3% across all stages. Eight studies (1,315 patients) reported the performance of 18FDG-PET/CT; the distant metastases detection rate was 2.1% (95% CI: 0% to 8.3%) with stage I, 7.3% (95% CI: 4.2% to 10.9%) with stage II, and 31.3% (95% CI: 24.2 to 38.9%) with stage III. The false positive rate ranged from 0.8% to 1.2% for stage II/III combined. We did not identify cost–effectiveness studies comparing CI tests and 18FDG-PET/CT per stage.

CONCLUSIONS: PET/CT showed a higher accuracy, however there were no appropriate direct comparisons against CI tests. Modelling studies could help to fill the gap between accuracy and clinical outcomes to support health care decision making.