Time-varying effects of prognostic factors associated with long-term survival in breast cancer

The impact of some prognostic factors on breast cancer survival has been shown to vary with time since diagnosis. However, this phenomenon has not been evaluated in Asians. In the present study, 4886 patients were recruited from the Shanghai Breast Cancer Survival Study, a longitudinal study of patients diagnosed during 2002-2006, with a median follow-up time of 11.2 years. Cox model incorporating time-by-covariate interactions was used to describe the time-varying effects of prognostic factors related to overall survival and disease-free survival. Age >= 65 years showed a progressively negative effect on breast cancer prognosis over time, whereas tumour size >2cm had a lasting and constant impact. Age significantly modified the effects of the tumour grade, nodal status and oestrogen receptor (ER) status on breast cancer survival. The detrimental effect of poorly differentiated tumours was time limited and more obvious in patients aged 45-54 years. Having >= 4 positive lymph nodes had a persistent and negative impact on prognosis, although it attenuated in later years; the phenomenon was more prominent in the 55-64-year age group. ER-positive status was protective in the first 3 years after diagnosis but was related to a higher risk of recurrence in later years; the time-point when ER-positive status turned into a risk factor was earlier in younger patients. These results suggest that older age, positive lymph node status, larger tumour size and ER-positive status are responsible for late death or recurrence in Asian breast cancer survivors. Extended endocrine therapy should be given earlier in younger ER-positive patients.