Over the years three different concepts regarding the cure of treated
breast cancer have emerged. These are clinical cure, personal cure and
statistical cure. The latter is the most accurate estimate of the cur
ability of a disease which is presumed to be fatal unless treated. Sta
tistical cure is the elimination of the hazard of death in a treated g
roup compared with an age-matched control population. When statistical
cure is studied in patients treated for early breast cancer, it is cl
ear that breast cancer is an incurable disease. The expected gains fro
m the relatively recent introduction of adjuvant therapy are too small
to alter this concept. The significance of prognostic factors in a di
sease deemed to be incurable therefore requires re-examination. The co
nventional prognostic factors of tumour size, nodal status and a combi
nation of those two in staging systems significantly discriminates in
terms of survival in the short term. However, when the characteristics
of long-term survivors are examined, neither tumour size nor nodal st
atus discriminates effectively. If this is the case, then we need to r
econsider novel treatment strategies which have been introduced in the
hope of increasing the curability of the disease and the selection fo
r those treatment strategies of patients using the conventional progno
stic factors of tumour size or nodal involvement.