[Natural history of breast cancer: a systematic review of worldwide randomized controlled trials of mammography screening].

To parameterize the 1-year transition probabilities between different health status of the natural history of breast cancer based on the data of randomized controlled trial of X-ray mammography screening worldwide. Based on the breast cancer screening randomized controlled trials defined by a mammography screening review from the Cochrane 2013 and the International Agency for Research on Cancer, a systematic review was initiated in PubMed by searching names of the key investigators of the trials, combined with the diseases, screening intervention and outcome indicators. If applicable, all the original cumulative incidence rates were converted into one-year transition rate, using the life-table approach considering time length of follow-up. A total of 23 reports from 9 RCTs were included. The data on transition rate between the healthy status to precancerous lesions was absent. The 1-year transition rate from health to carcinoma in situ was 17.78 to 50.21 per 100 000 persons in the intervention group and 9.16 to 26.84 per 100 000 persons in the control group. Correspondingly, the 1-year transition rate from health to breast cancer (including carcinoma in situ and invasive cancer) were estimated as 143.75 to 316.97 per 100 000 persons in the intervention group, and 141.45 to 288.84 per 100 000 persons in the control group. Furthermore, the transition rate from the healthy status to invasive breast cancer was 159.79 to 264.60 per 100 000 persons in intervention group and 170.12 to 255.33 per 100 000 persons in control group. The transition rate from carcinoma in situ to invasive breast cancer varied among different pathological types. The most common natural history states of reported by the included trials involved the full healthy status, carcinoma in situ and invasive breast cancer. The findings of transition rates between different health statuses will be informative for future model development of natural history studies of breast cancer. Information in relation to breast precancerous lesions still limited and needs to be further addressed.