Health outcomes of women with breast cancer

Abstract

The overall survival of breast cancer patients has increased quite remarkably in the past decades in the developed countries due to substantial improvements in diagnosis and treatment. As a consequence, the proportion of women alive after a breast cancer diagnosis is currently increasing. It is therefore becoming of outmost importance to also focus on medium- and long-term health outcomes of women with breast cancer.

Swedish population registers were used to study time-dependent survival of breast cancer patients according to age and tumor characteristics for the following health outcomes: causes of death, distant metastasis, risk of hospitalization due to a bone fracture, and risk of hospitalization due to an infection. Different survival analysis methodologies were applied including Cox regression models, Poisson regression models and flexible parametric survival models. Several measures were used to assess the outcomes of interest: rates, ratios and cumulative incidences. Comparisons with the general population using standardized incidence and mortality ratios were also performed.

The risk of dying from breast cancer varied by age, and by tumor characteristics in a time-dependent fashion. Circulatory system disorders were an important cause of death in our study population, in particular among women diagnosed with breast cancer after the age of 60 years. The risk of distant metastasis was still non-negligible after five years from breast cancer diagnosis in most subgroups of patients. Women with breast cancer were at increased risk of being hospitalized with a bone fracture or with an infection for at least ten years since diagnosis. The risk of hospitalization due to an infection was particularly increased for skin infections and sepsis. Women with breast cancer were also at significantly increased risk of dying after being hospitalized with a bone fracture or an infection.

Lymph node status at breast cancer diagnosis was not only found to be an important long-term predictor of overall and disease-free survival, but also of risk of hospitalization due to bone fracture or infection. Women with estrogen receptors-negative breast tumors showed a worse overall prognosis as compared with patients with estrogen receptors-positive breast tumors only in the first five years after diagnosis. Estrogen-receptor positive tumors carried a low but persistent risk of distant recurrence and death. A breast tumor size of more than 20mm at diagnosis was mainly associated with a worse short-term prognosis, however a mild significant association was detectable for more than five years from diagnosis.

In conclusion, there is no evidence to support discontinuation of clinical follow-up in breast cancer patients. Further investigation on more targeted approaches for different subgroups should be considered and more attention to medical conditions not directly related to breast cancer would be probably beneficial for these women. Preventive measures for bone fractures and infections could be taken into more consideration for all breast cancer patients at increased risk.