Is breast cancer curable: a study of long-term crude cumulative incidence

The problem of the search for a proportion of cured patients of breast cancer and the relative time to cure breast cancer has been known since the 1950s. The literature on this topic has developed during the last decades thanks to the availability of studies with long follow-up times, which was required by the type of disease in question. The aim of this study is to estimate, if it exists, the proportion of cured breast cancer patients with a parametric method accounting for competing causes of death.

In this work we used data from a two-group randomized controlled trial. From 1973 to 1980, 701 women were recruited with breast cancer measuring no more than 2 cm in diameter. They were randomly assigned to radical mastectomy (349 patients) or breast-conserving surgery followed by radiotherapy to the ipsilateral mammary tissue (352 patients). Thanks to a reliable classification of the causes of death, it was possible to analyze data using cause-specific mortality.

The parametric crude cumulative incidence estimates by the Gompertz distribution accorded closely with the non-parametric estimates. The analysis was also performed for subsets of patients based on menopausal status and number of positive lymph nodes. Statistical evidence for the existence of a cure fraction ranging from 0.48 to 0.71, depending on the subset, was found. However, the estimated proportion of cured patients has not been reached after 30 years' follow-up for most of the subsets.

Although statistical evidence has been found for the presence of a cure fraction, in practical terms it seems that 30 years after a breast cancer diagnosis cure cannot yet be claimed.