Radiologic analysis of interval cancers in the screening program called Florence Woman Project

PURPOSE: The aim of the present study was to provide data on interval cancer incidence after the first round of the screening program in the city of Florence, and to provide and discuss the results of the review of previous screening mammograms. MATERIAL AND METHODS: The screening program "Progetto Firenze Donna" involves all women age 50 to 69 years resident in the city of Florence. In the interval from the end of the first and the start of the second screening round 51 interval cancers were identified, of which 41 were available for radiological review. Films were reviewed by expert screening radiologists, not involved in the program and blinded to the final diagnosis, according to the review criteria recommended by the UK Guidelines for radiologists (blind and informed review). RESULTS: Sensitivity (screening errors/total) was different among blind reviewers (reader A = 7.3%, reader B = 14.6%), as well as specificity which shows an inverse pattern (reader A = 98.4%, reader B = 97.6%). Informed review classified 5 cases as "screening error" (12.2%), 11 as "minimal sign" (26.8%) and 25 as "occult" (61%). Informed review classified a higher proportion of cases as "minimal sign" (minimal abnormalities are better perceived with the knowledge of the site and the pattern of subsequent cancer). DISCUSSION AND CONCLUSIONS: Analysis of interval cancers is fundamental for the quality control of a screening program. According to this experience, informed review tends to overestimate "minimal signs" in the cancer site and should not be used (or "minimal signs" should be anyhow assumed as negative). Identification of interval cases as "minimal signs" may be influenced by individual variability. Standard criteria should be adopted to identify and review/classify interval cancers to allow comparisons between different programs.