Causes of death among women with breast cancer: A follow‐up study of 50 481 women with breast cancer in Finland

Our study aims to assess mortality from causes other than breast cancer among women with breast cancer with focus on indications of joint aetiology. Data on female breast cancer patients were obtained from the Finnish Cancer Registry and their underlying causes of death in 54 categories from the Statistics Finland. Standardised mortality ratios (SMR) and their 95% confidence intervals (CIs) were calculated for 50 481 patients diagnosed between 1971 and 2000 and followed until December 2012, stratified by histology, age at diagnosis and time since diagnosis. The expected numbers of deaths were based on respective mortality rates among the Finnish general population. Hazard ratio (HR) was estimated from Poisson regression model to compare risks of cause of death by histology. 41% of 30 841 deaths were due to causes other than breast cancer. Significant excess mortality was observed for stomach cancer (SMR 1.43, 95% CI 1.26‐1.62), circulatory system diseases (SMR 1.17, 95% CI 1.14‐1.20) and suicide (SMR 1.51, 95% CI 1.28‐1.78). In an age‐adjusted analysis, significantly higher relative risk of stomach cancer mortality was observed for lobular vs ductal subtype (HR 2.00, 95% CI 1.32‐3.02). Significantly increased SMRs were observed for cancers of respiratory organs among premenopausal women, and for other respiratory system diseases, dementia and Alzheimer disease among postmenopausal women. We conclude that female breast cancer patients are at increased risk of death from causes other than the breast cancer diagnosis including circulatory and respiratory system diseases and cancer of stomach, ovary and respiratory systems. The excess mortality because of different causes varies based on menopausal status and histology. There might be shared aetiological factors between the diagnosis of breast cancer and the causes of death among these patients.


| INTRODUCTION
Breast cancer is the most frequent cancer diagnosis and the leading cause of cancer death among women globally. 1 In 2018, there were 2.1 million estimated new breast cancer cases and 630 000 estimated deaths worldwide in women, which corresponds to about 24% of total cancer incidence and 15% of total cancer mortality. 1 In Finland, according to the latest estimates, breast cancer constituted 30% of all cancer cases and 15% of cancer deaths among women, and the 5-year relative survival was 90%. 2 The survival has increased mainly due to earlier diagnoses in mammographic screening and adjuvant treatments. 3,4 The risk of death among breast cancer patients also depends on many other factors such as hormone receptor status, histology of breast cancer and pre-existing health conditions. 5,6 Along with improving survival, the age and follow-up time increases after the breast cancer diagnosis, and there might be increased risk of dying from causes other than breast cancer, which might vary by the tumour characteristics, stage and calendar year at diagnosis. [7][8][9] Pulmonary circulation diseases, heart failures, other cardiovascular diseases, diabetes and gastrointestinal disease have been reported as causes of death among breast cancer patients more frequently than among other women. 10,11 What's new?
Gains in early detection, treatment, and survival in breast cancer presumably increase the chances that patients will die from causes other than breast cancer. Here, mortality patterns and indications of joint etiology with other diseases was investigated among Finnish breast cancer patients.
Observations indicate that female breast cancer patients experience excess mortality from circulatory system disease and from other cancers, particularly those of the lung, stomach, and ovary. Mortality from different causes varied by menopausal status and histology. The findings suggest that certain risk factors are involved in the etiology of breast cancer and eventual cause of mortality.
T A B L E 1 Observed (Obs) number of deaths and standardised mortality ratios (SMR) with 95% confidence interval (CI) for causes of death among women with breast cancer by menopausal status

| DISCUSSION
We utilised data from nationwide registries to identify the causes of death and changes in the causes of death over the time among women diagnosed with breast cancer in Finland. The SMR for breast cancer as well as all other causes combined was higher among women with breast cancer diagnosed before 50 years than among older women.

| CANCER MORTALITY
We observed that risk of dying from other malignancies in women with breast cancer was higher than that in the general population.
Excess mortality was most marked for cancer of stomach in both premenopausal (SMR 2.25) and postmenopausal (SMR 1.31) women.
Excess mortality was also observed for ovarian cancer (SMR 1.49) among women with premenopausal but not among women with postmenopausal breast cancer. These excesses might have some association with the germline mutations in the breast cancer type 1/2 (BRCA1/2) genes. Germline mutations in the tumour suppressor genes BRCA1 and BRCA2 expose individuals to early onset breast and ovarian cancer. [14][15][16] Mutations in BRCA1 and BRCA2 have also been reported to be associated with increased risk of developing stomach cancer. [17][18][19] Increased SMRs observed in our study for cancers of the lung, stomach and ovary observed in our study might be related to tobacco smoking. Nicotine is likely to promote proliferation and progression of cells as well as increased cell migration in cell lines of gastric, breast and ovarian cancer cells. [20][21][22][23] Several studies consistently show an association between cigarette smoking and breast cancer. 20,24-28 A recent study based on a large Norwegian cohort estimated that among smokers, one in nine breast cancer cases could have been avoided in the absence of active smoking. 29 The carcinogens found in tobacco are described more effective in inducing cancer in less differentiated breast tissue, which might be the reason for the stronger increase in breast cancer risk among younger women. 30,31 Smoking is an established strong risk factor for lung cancer and has repeatedly been shown to also be associated with stomach cancer with a dose-T A B L E 2 Observed (Obs) numbers of deaths, standardised mortality ratio (SMR) and 95% confidence interval (CI) for causes of death among women with breast cancer, by histology response relationship. 32 A large meta-analysis indicated that smoking significantly increases the risk of ovarian cancer among premenopausal women but not among postmenopausal women. 33 Excess mortality of ovarian cancer observed only among premenopausal women in our study is consistent with this finding.
A large prospective study of European countries showed a significant association of alcohol intake with breast cancer and more than 4% increase in risk for each 10 g/d increase in alcohol intake. 34,35 Alcohol consumption is also suggested to increase the risk of stomach cancer. 36 Hence, our finding on an increased SMR of stomach cancer might also point to the common risk associated with alcohol consumption. Additionally, studies have shown that alcohol consumption is strongly associated with the increased risk up to two-fold of lobular breast cancer than ductal breast cancer as alcohol increases the production of oestrogen. [37][38][39] The significantly higher HR of stomach cancer mortality among women with lobular breast cancer as compared to ductal subtype might be another indication of an association with alcohol consumption.
In the line of our study, earlier studies have reported an increased risk of developing ovarian cancer after breast cancer 40 and increased ovarian cancer mortality among breast cancer patients. 41 In addition to the earlier discussed risk factors such as genetic mutations and smoking, shared risk factors associated with childbirth and female hormones might explain the association between breast and ovarian cancer. [42][43][44] Obesity increases the risk of both breast cancer and ovarian cancer 45 and may hence offer another partial explanation to the coexistence of these diseases. Conversely, lobular breast cancer is more strongly associated with increased exposure to endogenous oestrogens such as early menarche, late menopause and exogenous hormones. 51 Therefore, the higher level of endogenous hormones and use of hormonal therapy leading to lobular breast cancer might have worked as the protective factor against circulatory system diseases among these women.
Type 2 diabetes mellitus (T2DM) has been reported as an important cause of death among breast cancer patients in previous studies. 41,52 A meta-analysis reported that women with T2DM are at increased risk of developing breast cancer. 53 The association between breast cancer and T2DM could possibly be due to common risk factors for these two diseases such as physical inactivity, dietary patterns and obesity. However, in our study among Finnish female breast cancer patients, no excess mortality was observed for T2DM in any age and histology category. A recent large Finnish study also indicated that there is no excess incidence of breast among women with T2DM. 54 Hence, these two diseases seem not to have joint aetiology in Finland.
Diseases of the respiratory system were observed as a cause of significant excess mortality among women diagnosed with breast cancer after age 50. It could be possible that breast cancer patients have a weakened immune system as compared to other women of the same age and are more prone to infections, including the respiratory system.
We observed significantly elevated mortality from accidents and violence among postmenopausal breast cancer patients, which is hardly due to shared risk factors with breast cancer. Women treated with chemotherapy after breast cancer treatment are at increased risk of osteoporosis and fractures. 55 An increased risk of suicides was observed among both premenopausal and postmenopausal women diagnosed with breast cancer who might find it difficult to cope better with cancer diagnosis and its threat to the functionality of life. 56

| STRENGTHS AND LIMITATIONS OF THE STUDY
The strengths of our study include the long-term follow-up of a large number of Finnish women diagnosed with breast cancer and completeness of the information. The study is based on high-coverage and accurate nationwide registers that allow us to overcome the selection bias. We studied a wide range of causes of death among breast cancer patients, some of which may not be directly important for the main objective of our study-to get indirect hints of aetiology of breast cancer-but may provide a reference for further studies. A limitation of our study is that we are unable to separate the proportion of the mortality attributed to shared risk factors from the effects of treatment of the breast cancer. The treatment procedures may be different for cancer patients with different histology and of different age and have advanced over decades, which could partly explain differences in mortality risks between subgroups of breast cancer patients. Similarly, it was not within the scope of our study to trace back the lifestyle and hormonal factors of breast cancer patients in order to associate them with breast cancer and also the mortality cause.

| CONCLUSION
Our study concludes that there might be shared genetic, reproductive, lifestyle and environmental risk factors between breast cancer and mortality outcomes among breast cancer patients. The mortality patterns provide indirect information on which risk factors might have played a role in the development of breast cancer. It is noticeable that modification of lifestyle and risk behaviours would reduce both the incidence of breast cancer and other related diseases.