Critical assessment of the clinical TNM system in breast cancer. Report from the Yorkshire Breast Cancer Group.

If useful conclusions are to result from trials of treatment in breast cancer it is vital that the system of classification is accurate and reproducible. In 348 patients with operable breast cancer the TNM classification (1974 version) was examined. Surgeons, radiologists, and pathologists varied considerably in measuring tumour size (T), surgeons and radiologists agreeing in only 39% of cases, surgeons and pathologists in 54%, and radiologists and pathologists in 59%. Surgeons and pathologists agreed on the state of the regional lymph nodes (N) in 68% of cases. Thus the 1974 version of the TNM system is an inaccurate method of classifying breast cancer. The revised 1978 version, however, incorporating histopathological findings, should improve the accuracy of classification, and it is strongly recommended that future trials use this new version.