Abstract
Purpose:
To prospectively investigate accelerated aging and its association with total mortality and breast-cancer specific mortality/recurrence among breast cancer survivors.
Methods:
This study included 4,218 female breast cancer patients enrolled into a population-based cohort study approximately 6-months post-diagnosis. Information on aging related symptoms (i.e., self-rated overall health condition, energy level, depression, sleep difficulty and quality) was collected at 18- and 36-month post-diagnosis surveys. Information on overall health, daily function impairments, survival status and recurrence was collected at 10-year post-diagnosis survey. Record linkages with vital statistics were conducted to collect mortality information. Cox proportional hazards model was applied.
Results:
Among 3,041 10-year survivors with a mean age of 63.7±9.7 years, respectively 52.3%, 19.0% and 27.6% reported poor health, limitation in daily activity and climbing floors. Age-specific prevalence revealed that breast cancer survivors reached similar prevalence of the functional limitations 5–10 years earlier than cancer-free women. At the 18-month post-diagnosis survey, respectively 47.0%, 72.5% and 25.1% of survivors reported unsatisfied overall health condition, reduced energy level and depression symptoms. After a median follow-up of 10.9 years, low self-rated overall health, low energy level and depression were significantly associated with increased total mortality, with hazard ratios (HRs; 95% confidence intervals [CI]) of 3.14 (2.43, 4.06), 1.49 (1.20, 1.84) and 1.59 (1.21, 2.09), respectively. Low self-rated health was associated with breast-cancer specific mortality/recurrence (HR=1.85, 95% CI: 1.30, 2.65). No significant association was found for sleep difficulty and quality.
Conclusions:
Aging related physical changes/symptoms are commonly presented at 18-months after breast cancer diagnosis and are associated with worse prognosis.
Impact:
Our findings highlight the concern of accelerated aging among breast cancer survivors.
Keywords: Accelerated aging, energy, self-rated health, depression, sleep, breast cancer, mortality
Introduction
Breast cancer is the most common cancer among women in both developed and developing countries/regions [1]. With advances in early detection and treatment, however, worldwide breast cancer mortality rates are decreasing [2]. For women diagnosed with breast cancer between 2005–2009, the estimated age-standardized 5-year survival reached 80% or higher in many countries around the world [3] and exceeded 90% in some high-income countries [4]. By 2018, there were approximately 6.9 million breast cancer survivors globally who had been diagnosed within the previous 5 years [5].
Despite the improving survival, a growing body of evidence supports that both cancer and cancer-related treatments may cause physical and psychological changes resulting in impaired physical and mental functioning [6–8], indicating accelerating aging among breast cancer survivors [6, 9]. Aging related physical changes/symptoms include, but are not limited to, fatigue, depression, decreased physical functioning, increased comorbidity, premature ovarian failure and pain [9, 10]. Fatigue, or low energy level, defined as a feeling of serious physical or emotional exhaustion [11], is one of the most frequent symptoms reported among off-treatment cancer survivors [12, 13]. Studies have shown that fatigue is especially common among breast cancer survivors and is associated with decreased physical functioning and increased psychological distress [14, 15]. Similarly, incidence of depressive disorders post-diagnosis is higher in cancer survivors in comparison to patients with other diseases [16, 17]. In addition, sleep disturbance frequently follows fatigue and depression and manifests as a cluster symptom [18, 19]. Particularly, a higher prevalence of sleep disturbance was observed in breast cancer survivors compared to survivors of other cancers [20, 21], and most patients claimed that the sleep disorder was caused or exacerbated by cancer or cancer-associated treatment, such as pain and stress [22].
In addition to being associated with worse health-related quality of life [7], accelerated aging may contribute to premature mortality in cancer survivors through underlying biological changes, such as chronic inflammation [9, 23]. For example, sleep disorder has been shown to be linked with enhanced inflammation [24], weakened immune function [25] and decreased effectiveness and tolerability of breast cancer treatments [26]. However, to date, few studies have comprehensively evaluated accelerated aging and its association with mortality and recurrence among breast cancer survivors. Only limited studies have evaluated the associations of individual post-diagnosis symptoms, such as sleep disturbance, with mortality in breast cancer survivors, yielding inconsistent results [27–29].
In the reported study, using data from a prospective longitudinal study of breast cancer patients, we evaluated overall health condition and function limitation among 10-year breast cancer survivors. We further comprehensively investigated self-reported symptoms suggestive of accelerated aging (i.e., depression, energy level and self-rated health condition, as well as sleep difficulty and sleep quality) at 18- and 36-months post-cancer diagnosis for their associations with all-cause mortality and cancer recurrence.
Methods
Data used in this study included women from the Shanghai Breast Cancer Survival Study, a population-based prospective cohort study of breast cancer survivors living in Shanghai, China. Detailed information on the study design and recruitment methods have been described previously [30]. Briefly, 6,299 patients diagnosed with primary breast cancer between March 2002 and April 2006 were identified through the population-based Shanghai Cancer Registry. Among them, 5,042 provided informed consent and were enrolled in the study at approximately 6-months post-cancer diagnosis (participation rate = 80%). Subsequent in-person follow-up surveys took place at 18-, 36-, 60- and 120-months after cancer diagnosis, with respective response rates of 92.8%, 88.2%, 82.5%, and 87.8%. Record linkages with the Shanghai Vital Statistics Registry were conducted, which captured virtually 100% vital information for cohort members. In the current study, we excluded participants who were diagnosed with breast cancer at an advanced stage (TNM stage IV, n=28), or who had missing information on TNM stage (n=230). We set the 18-months of follow-up survey as the study entry time for the current study to minimize the influence of active cancer treatments on energy level, depression, sleep disturbance and overall health assessment. As a result, 566 cases who had a recurrence of cancer (n=223, including 87 deaths) before the 18-month survey, or did not participate in the survey for miscellaneous reasons (n=343), were excluded from the study. The remaining 4,218 women were included in this study.
The study was performed in accordance with the Declaration of Helsinki, and the study protocols were approved by the Institutional Review Boards of participating institutes in both the People’s Republic of China and the United States. Written informed consent was acquired from all participants.
Data Collection
Structured in-person surveys were conducted by trained interviewers at 6-, 18-, 36-, 60- and 120-months post-diagnosis. Information was collected on demographic characteristics, selected lifestyle factors, menstrual and reproductive histories, dietary patterns and self-rated health conditions, as well as clinical factors including cancer diagnosis and treatment, cancer stage, comorbidity status, and estrogen receptor (ER) and progesterone receptor (PR) status. Weight and height were measured at the 6-month post-diagnosis survey, according to a standard protocol, and were used to estimate body mass index (BMI). Comorbidity status was created by the Charlson Comorbidity Index (≥1 or 0) using baseline information [31]. Cancer diagnosis, treatment and disease stage information were further verified by reviewing medical records and pathology slides.
Self-reported information on selected aging related conditions was collected at the 18-month and 36-month post-diagnosis surveys. We assessed participants’ energy levels by asking: “How was your energy in the last week?” with five answer options: 1) Always felt tired; 2) Often felt tired; 3) Sometimes felt tired; 4) Often felt energetic; 5) Always felt energetic. The question we used to assess depressive symptoms was: “Did you often feel depressed during the past week? How serious was your depression?” with answer choices: 1) Most serious; 2) Very serious; 3) Serious; 4) Minor; 5) Not at all. To determine the self-perceived overall health condition of our study participants, we asked: “How is your health condition now? Are you satisfied with your current health condition?” with answer options: 1) Very unsatisfied; 2) Unsatisfied; 3) Somewhat satisfied; 4) Satisfied; 5) Very satisfied. Information on sleeping attributes, including sleep difficulty and sleep quality, was also collected by asking the following questions. Sleep difficulty: “How was your sleep during the last week?” with five answer options: 1) Had difficulty falling asleep or staying asleep every night; 2) Usually had difficulty falling or staying asleep; 3) Often had difficulty falling or staying asleep; 4) Sometimes had difficulty falling or staying asleep; 5) Never had difficulty falling or staying asleep. Sleep quality: “Did you wake up refreshed, relaxed and well-rested in the morning during the past week?” with answer choices: 1) Never; 2) Rarely; 3) Sometimes; 4) Most of the time; 5) Almost every day. For each of these five questions, answers were further categorized into three groups for analytical purposes, according to the following categorization rules: individuals answering option one or two were classified as one group, those answering option three were categorized as one group, and those selecting option four or five were classified as one group.
In addition, at the 10-year post-diagnosis follow-up survey, information was collected on self-reported health condition by asking: “How is your health condition now?” with answer options: 1) Excellent; 2) Very good; 3) Good; 4) Not bad; 5) Bad. Participants who answered option 4 or 5 were categorized as reporting poor health condition. Information was also collected on daily activities by asking the following question: “Did your health limit the following of your daily activities during the past 4 weeks?” with three answer options: 1) Very much; 2) Slightly; 3) Not at all. Examples of moderate activities included in the survey were: moved the table, pushed the vacuum, bowled, played golf, did Tai Chi and climbed several stairs. Participants who answered option 1 or 2 were categorized as having limitation in daily function.
We also collected information on disease recurrence and survival status during the in-person surveys. In addition, we periodically linked unique resident IDs of study participants with the Vital Statistics Registry database to ascertain survival status for all study participants. The most recent vital record linkage was conducted on December 31, 2017.
Statistical Analysis
We evaluated the prevalence rate of self-reported health conditions and limitations on daily activities among study participants based on the 10-year post-diagnosis data survey and compared them with data from participants of the Shanghai Women’s Health Study (SWHS). The SWHS is an ongoing, prospective population-based cohort study, at baseline from March 1997 to May 2000, which included 74,941 women aged 40–70 years from the same area [32]. In-person follow-up surveys were conducted every 2–3 years. The fourth follow-up survey (2008–2011) asked the same questions on self-reported health conditions, with a response rate of 92.0% (N=48,010; age range: 51–89 year at the survey).
Primary end points for the survival analyses were any death (all-cause mortality) and breast cancer-related death or cancer recurrence (breast cancer-specific mortality/recurrence). Our exposure variables, including self-rated health status, energy level, depression, sleep difficulty and sleep quality, were treated as either static variables that were assessed at 18-months after breast cancer diagnosis, or as time-dependent variables that were measured at both 18-months and 36-months post-diagnosis.
We applied Chi-square tests and analyses of variance (ANOVA) to evaluate associations between covariates and exposure variables of interest, measured at the 18-month survey. Total mortality and recurrence/breast cancer-specific mortality in each patient subgroup, with entry time set as 18-months after diagnosis, were estimated using the Kaplan-Meier method. Cox proportional hazards models were used to estimate hazard ratios (HR) and 95% confidence intervals (CIs), separately, for the associations of self-rated health status, energy level, depression, sleep difficulty and sleep quality, with total mortality or breast cancer-specific mortality/recurrence, using age as the time scale [33]. Subjects with the lowest study variables score (i.e., satisfied self-rated health condition, no depression, high energy level, rare sleep difficulty or high sleep quality) were treated as the reference group. Entry time was defined as age at 18-month post-diagnosis survey. Exit time was defined as age at event (either loss to follow-up or December 31, 2017, whichever occurred first). Potential confounding variables identified a priori included BMI (<18.5 kg/m2, 18.5–24.9 kg/m2, 25–29.9 kg/m2, ≥30 kg/m2), education (less than high school, high school graduate, some college, or more), monthly income per person (<1,000, 1,000–1,999, ≥2000), exercise (absence/presence), menopause (yes/no), tamoxifen use (yes/no), chemotherapy (yes/no), radiotherapy (yes/no), comorbidity (any/none), TNM stage (0, I, IIA, IIB, III), ER status (negative, positive, missing), PR status (negative, positive, missing), mastectomy (yes/no) and menopausal symptoms (not applicable, yes, no). To adjust for covariates in the analyses, we used information collected at the 18-month survey, when available. For education, income, comorbidity, radiotherapy, TNM stage, ER status and PR status, we used information collected at the 6-month post-diagnosis survey. Covariates with missing values (1 subject on education, 3 subjects on income and 2 subjects on BMI) were assigned to the most frequent category.
All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, United States). Statistical tests were based on two-tailed probability with a significance level set at P<0.05.
Results
The average age of the 4,218 study participants was 53.5 years (standard deviation [SD] = 9.9, range: 20–75 years) at breast cancer diagnosis, 55.1 years (SD = 9.9, range: 21–76 years) at the 18-month post-diagnosis survey, and 63.7 years (SD = 9.7, range: 32–88 years) at the 10-year post-diagnosis survey.
A total of 3,041 breast cancer survivors completed the 10-year post-diagnosis follow-up survey. Of these, 52.3% reported having a poor health condition. For 2,748 survivors who responded to the function related questions, 19.0% and 27.6% reported restriction to perform moderate daily activities or a limited ability to climb stairs, respectively. Higher prevalence of functional limitations among breast cancer survivors was observed across all age groups of SBCSS participants in comparison with SWHS participants. Apparently, breast cancer survivors reached similar prevalence of the functional limitations 5–10 years earlier than cancer-free women. A similar pattern was observed for self-reported poor health condition (Figure 1).
Figure 1.
Age-specific prevalence of aging-related symptoms among breast cancer survivors and cancer-free women. A. Prevalence of restriction to perform moderate daily activities; B. Prevalence of limited ability to climb stairs. C. Prevalence of poor self-rated health condition.
A total of 3,041 10-year breast cancer survivors from SBCSS (age range: 32–88 year) and 48,010 cancer-free women from SWHS (age range: 51–89 year) were included in this analysis, of whom 293 breast cancer survivors did not answer the two questions on functional limitations.
Abbreviations: SBCSS, Shanghai Breast Cancer Survivor Study; SWHS, Shanghai Women Health Study.
At the 18-month post-diagnosis survey, 47.0% of participants reported a self-rated health status of unsatisfied (43.5% somewhat satisfied and 3.5% not satisfied), 72.5% reported reduced energy level (56.7% moderate and 15.8% low level), and 25.1% reported depression (21.2% moderate and 3.9% severe depression). In addition, 26.2% had sleep difficulty (14.8% sometimes and 11.4% always), and 19.3% reported impaired sleep quality (13.5% moderate and 5.8% low quality). At the 36-month post-diagnosis survey, a higher proportion of participants reported unsatisfied self-rated health condition (50.3%), reduced energy level (74.1%), depression (27.8%), sleep difficulty (28.9%), or impaired sleep quality (24.0%; data is not shown in tables). Women with higher self-rated health status or higher energy levels, with less depression, or with better sleeping attributes, were more likely to be diagnosed with earlier TNM stages of breast cancer, have higher education levels, higher incomes and were more likely to participate in regular exercise. They were younger on average and less likely to be in menopause, have comorbidities, or receive chemotherapy. In addition, women with better sleep attributes tended to have better self-rated health condition and higher energy levels, and they were less likely to be depressed. (Tables 1 and 2).
Table 1.
Demographic and clinical characteristics of study participants by self-rated health condition, depression, and energy level assessed 18-months after cancer diagnosis
| Characteristics | Total | Self-rated health condition | Energy level | Depression | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| (N=4,218) | Satisfied | Somewhat satisfied | Not satisfied | P value | High | Moderate | Low | P value | Not at all | Moderate | Severe | P value | |
| (N=2,236) | (N=1,836) | (N=146) | (N=1,160) | (N=2,392) | (N=666) | (N=3,159) | (N=894) | (N=165) | |||||
| Age at diagnosis, years | 53.5 (9.9) | 53.0 (10.0) | 53.9 (9.7) | 57.4 (10.7) | <0.01 | 51.7 (9.4) | 53.7 (9.9) | 56.1 (10.5) | <0.01 | 53.2 (9.9) | 53.8 (9.9) | 57.7 (10.1) | <0.01 |
| Age at 18-month post-diagnosis, years | 55.1 (9.9) | 54.6 (10.0) | 55.4 (9.7) | 58.9 (10.7) | <0.01 | 53.2 (9.4) | 55.2 (9.9) | 57.6 (10.5) | <0.01 | 54.8 (9.9) | 55.3 (9.9) | 59.3 (10.1) | <0.01 |
| Body mass index, kg/m2 | 24.4 (3.4) | 24.5 (3.2) | 24.5 (3.5) | 24.0 (3.7) | 0.27 | 24.4 (3.3) | 24.4 (3.3) | 24.4 (3.9) | 0.09 | 24.5 (3.3) | 24.2 (3.4) | 24.0 (3.9) | <0.01 |
| Education | 0.08 | 0.04 | 0.68 | ||||||||||
| <High school | 84.4 | 83.0 | 86.0 | 87.0 | 82.9 | 84.2 | 87.8 | 84.4 | 84.3 | 86.1 | |||
| High school | 8.8 | 9.5 | 8.0 | 8.2 | 9.0 | 9.0 | 7.7 | 8.6 | 9.6 | 7.3 | |||
| >High school | 6.8 | 7.6 | 6.1 | 4.8 | 8.0 | 6.9 | 4.5 | 7.0 | 6.0 | 6.7 | |||
| Income, yuan/month/capital | <0.01 | <0.01 | <0.01 | ||||||||||
| <1,000 | 57.2 | 53.5 | 60.8 | 67.2 | 51.7 | 58.0 | 63.8 | 55.7 | 61.4 | 63.7 | |||
| 1,000–1,999 | 31.1 | 32.3 | 30.0 | 26.0 | 32.0 | 31.5 | 28.2 | 31.7 | 29.9 | 26.7 | |||
| ≥2,000 | 11.7 | 14.2 | 9.2 | 4.8 | 16.3 | 10.5 | 8.0 | 12.7 | 8.7 | 9.7 | |||
| Exercise | 74.1 | 75.6 | 73.3 | 61.6 | <0.01 | 75.3 | 75.8 | 65.9 | <0.01 | 74.6 | 74.5 | 61.8 | <0.01 |
| Menopause | 73.1 | 71.2 | 74.7 | 81.5 | <0.01 | 69.1 | 73.7 | 77.8 | <0.01 | 72.1 | 73.4 | 89.1 | <0.01 |
| Tamoxifen use | 52.5 | 51.5 | 54.1 | 45.9 | 0.15 | 52.2 | 53.6 | 49.0 | 0.05 | 52.8 | 51.2 | 52.1 | 0.77 |
| Chemotherapy | 4.7 | 4.3 | 4.9 | 8.2 | 0.09 | 5.3 | 4.2 | 5.6 | 0.21 | 4.8 | 4.5 | 4.9 | 0.93 |
| Radiotherapy | 30.7 | 31.7 | 30.0 | 24.0 | 0.11 | 34.5 | 29.6 | 28.1 | <0.01 | 31.4 | 29.0 | 25.5 | 0.12 |
| Mastectomy | 95.2 | 95.3 | 95.2 | 93.8 | 0.74 | 95.0 | 95.2 | 95.4 | 0.90 | 95.3 | 95.1 | 93.9 | 0.74 |
| Comorbidity | 20.2 | 17.9 | 22.2 | 29.5 | <0.01 | 15.0 | 20.3 | 28.7 | <0.01 | 18.8 | 22.6 | 33.9 | <0.01 |
| TNM stage | 0.34 | 0.06 | 0.14 | ||||||||||
| 0-I | 39.8 | 38.4 | 41.1 | 41.7 | 36.6 | 41.2 | 39.9 | 39.1 | 41.8 | 41.8 | |||
| II | 52.2 | 53.3 | 50.9 | 52.1 | 54.9 | 50.5 | 53.6 | 52.6 | 50.6 | 52.1 | |||
| III | 8.1 | 8.3 | 8.0 | 6.2 | 8.5 | 8.3 | 6.5 | 8.3 | 7.6 | 6.1 | |||
| Estrogen receptor status | 0.60 | 0.12 | 0.61 | ||||||||||
| Negative | 33.9 | 34.6 | 32.8 | 34.9 | 36.7 | 33.1 | 31.7 | 34.2 | 33.5 | 30.3 | |||
| Positive | 65.4 | 64.6 | 66.5 | 65.1 | 62.5 | 66.1 | 67.9 | 65.1 | 65.9 | 69.7 | |||
| Unknown | 0.7 | 0.8 | 0.7 | 0.0 | 0.8 | 0.8 | 0.5 | 0.8 | 0.7 | 0.0 | |||
| Progesterone receptor status | 0.12 | 0.45 | 0.72 | ||||||||||
| Negative | 39.5 | 39.3 | 39.9 | 49.7 | 38.4 | 39.0 | 41.3 | 39.4 | 40.0 | 40.6 | |||
| Positive | 59.5 | 59.8 | 59.7 | 50.7 | 57.9 | 59.8 | 60.8 | 59.6 | 59.1 | 59.4 | |||
| Unknown | 1.0 | 1.0 | 1.1 | 0.0 | 0.8 | 1.2 | 0.8 | 1.1 | 0.9 | 0.0 | |||
Values are presented as mean (standard deviation) or percentage.
Percentages may not add up to 1 because of rounding.
Table 2.
Demographic and clinical characteristics of study participants by sleeping attributes assessed 18-months after cancer diagnosis
| Characteristics | Sleep difficulty | Sleep quality | ||||||
|---|---|---|---|---|---|---|---|---|
| Rarely | Sometimes | Always | P value | High | Moderate | Low | P value | |
| (N=3,113) | (N=623) | (N=482) | (N=3,404) | (N=568) | (N=246) | |||
| Age at diagnosis, years | 53.0 (9.9) | 54.7 (10.0) | 55.2 (10.0) | <0.01 | 53.3 (9.8) | 54.1 (10.3) | 54.9 (10.4) | 0.02 |
| Age at 18-months post-diagnosis, years |
54.5 (9.9) | 56.3 (10.0) | 56.8 (10.0) | <0.01 | 54.9 (9.8) | 55.6 (10.3) | 56.4 (10.4) | <0.01 |
| Body mass index, kg/m2 | 24.5 (3.4) | 24.3 (3.5) | 24.1 (3.2) | 0.04 | 24.5 (3.3) | 24.2 (3.6) | 23.8 (3.3) | <0.01 |
| Education | <0.01 | 0.16 | ||||||
| <High school | 84.1 | 81.4 | 90.5 | 84.1 | 83.8 | 89.8 | ||
| High school | 8.9 | 10.6 | 5.4 | 8.8 | 9.7 | 6.1 | ||
| >High school | 7.0 | 8.0 | 4.2 | 7.1 | 6.5 | 4.1 | ||
| Income, yuan/month/capital | <0.01 | 0.15 | ||||||
| <1,000 | 56.3 | 55.2 | 65.3 | 56.2 | 60.9 | 62.2 | ||
| 1,000–1,999 | 31.5 | 32.6 | 26.6 | 31.5 | 29.2 | 30.1 | ||
| ≥2,000 | 12.2 | 12.2 | 8.1 | 12.3 | 9.9 | 7.7 | ||
| Exercise | 74.0 | 75.3 | 73.4 | 0.74 | 74.6 | 73.8 | 67.5 | 0.05 |
| Menopause | 71.5 | 76.6 | 78.6 | <0.01 | 72.5 | 74.8 | 77.6 | <0.01 |
| Tamoxifen use | 52.0 | 53.9 | 53.9 | 0.56 | 52.6 | 51.8 | 52.0 | 0.59 |
| Chemotherapy | 4.2 | 6.4 | 5.6 | 0.04 | 4.8 | 4.6 | 4.5 | 0.97 |
| Radiotherapy | 31.4 | 28.7 | 28.4 | 0.22 | 31.3 | 25.9 | 33.3 | 0.02 |
| Mastectomy | 95.1 | 95.0 | 95.9 | 0.76 | 95.0 | 96.1 | 95.5 | 0.48 |
| Comorbidity | 18.8 | 20.4 | 28.6 | <0.01 | 19.0 | 23.9 | 26.8 | <0.01 |
| TNM stage | 0.49 | <0.01 | ||||||
| 0-I | 39.6 | 38.5 | 42.3 | 39.6 | 40.1 | 40.7 | ||
| II | 52.1 | 52.0 | 49.8 | 52.3 | 52.7 | 49.1 | ||
| III | 7.8 | 9.5 | 7.9 | 8.1 | 7.2 | 10.2 | ||
| Estrogen receptor status | 0.53 | 0.19 | ||||||
| Negative | 34.6 | 32.6 | 30.9 | 34.4 | 31.7 | 31.3 | ||
| Positive | 64.7 | 66.8 | 68.3 | 64.8 | 68.1 | 68.3 | ||
| Unknown | 0.7 | 0.6 | 0.8 | 0.9 | 0.2 | 0.4 | ||
| Progesterone receptor status | 0.79 | 0.13 | ||||||
| Negative | 39.9 | 39.3 | 37.6 | 39.6 | 40.5 | 36.6 | ||
| Positive | 59.1 | 59.4 | 61.6 | 59.2 | 59.3 | 63.0 | ||
| Unknown | 1.0 | 1.3 | 0.8 | 1.2 | 0.2 | 0.4 | ||
| Self-rated health condition | <0.01 | <0.01 | ||||||
| Satisfied | 58.3 | 43.7 | 30.9 | 58.1 | 35.6 | 23.2 | ||
| Somewhat satisfied | 39.3 | 51.9 | 60.2 | 39.9 | 57.9 | 61.0 | ||
| Not satisfied | 2.4 | 4.5 | 8.9 | 2.1 | 6.5 | 15.9 | ||
| Energy level | <0.01 | <0.01 | ||||||
| High | 31.9 | 17.5 | 12.2 | 31.9 | 11.6 | 3.7 | ||
| Moderate | 56.4 | 61.8 | 52.1 | 57.1 | 59.3 | 44.7 | ||
| Low | 11.7 | 20.7 | 35.7 | 11.0 | 29.1 | 51.6 | ||
| Depression | <0.01 | <0.01 | ||||||
| Not at all | 80.0 | 63.6 | 56.6 | 80.4 | 56.5 | 40.7 | ||
| Moderate | 17.6 | 31.3 | 31.3 | 17.4 | 34.9 | 41.9 | ||
| Severe | 2.4 | 5.1 | 12.0 | 2.1 | 8.6 | 17.5 | ||
Values are presented as mean (standard deviation) or percentage.
Percentages may not add up to 1 because of rounding.
After a median follow-up of 10.9 years (range: 0.1–14.2 years) after the 18-month survey, 823 total deaths and 687 recurrences or breast cancer-related deaths were recorded. After adjustment for potential confounders, both poor self-rated health condition and low energy level were significantly associated with both total mortality and breast cancer-specific mortality/recurrence. The respective HRs were 1.53 (95% CI: 1.12, 2.09) and 1.32 (1.06, 1.64) for total mortality, and 1.50 (95% CI: 1.04, 2.17) and 1.34 (95% CI: 1.04, 1.72) for breast cancer-specific mortality/recurrence. Depression was not significantly associated with total mortality (HR=0.86, 95% CI: 0.60, 1.23) or breast cancer-specific mortality/recurrence (HR=1.15, 95% CI: 0.78, 1.68). When treated as time-dependent variables, poor self-rated health condition, low energy level and depression were all significantly associated with total mortality, with respective HRs of 3.14 (95% CI: 2.43, 4.06), 1.49 (95% CI: 1.20, 1.84) and 1.59 (95% CI: 1.21, 2.09). Only self-rated health condition was associated with breast cancer-specific mortality/recurrence, with a HR of 1.85 (95% CI: 1.30, 2.65; Table 3).
Table 3.
HRs (95% CI) for total mortality and recurrence/breast cancer-specific mortality in association with health condition, energy level and depression for participants who survived for at least 18 months since cancer diagnosis *
| Total mortality | Recurrence/breast cancer-specific mortality | |||||||
|---|---|---|---|---|---|---|---|---|
| Death | 18-months HR (95% CI) | 18- and 36-months HR (95% CI) | Recurrence/breast cancer-specific death | 18-months HR (95% CI) | 18- and 36-months HR (95% CI) | |||
| Number | Rate (%) # | Number | Rate (%) # | |||||
| Self-rated health condition | ||||||||
| Satisfied | 427 | 21.7 | 1.00 | 1.00 | 355 | 36.5 | 1.00 | 1.00 |
| Somewhat satisfied | 351 | 20.4 | 1.03 (0.89, 1.19) | 1.25 (1.08, 1.45) | 300 | 36.7 | 1.03 (0.88, 1.21) | 1.08 (0.93, 1.27) |
| Not satisfied | 45 | 31.1 | 1.53 (1.12, 2.09) | 3.14 (2.43, 4.06) | 32 | 38.3 | 1.50 (1.04, 2.17) | 1.85 (1.30, 2.65) |
| Energy level | ||||||||
| High | 199 | 19.1 | 1.00 | 1.00 | 158 | 19.4 | 1.00 | 1.00 |
| Moderate | 461 | 21.4 | 1.15 (0.97, 1.36) | 1.28 (1.07, 1.53) | 416 | 41.2 | 1.38 (1.14, 1.66) | 1.14 (0.95, 1.37) |
| Low | 163 | 25.8 | 1.32 (1.06, 1.64) | 1.49 (1.20, 1.84) | 113 | 36.5 | 1.34 (1.04, 1.72) | 1.08 (0.85, 1.37) |
| Depression | ||||||||
| Not depressed | 625 | 22.1 | 1.00 | 1.00 | 513 | 38.8 | 1.00 | 1.00 |
| Depressed | 165 | 19.5 | 0.95 (0.79, 1.13) | 1.18 (1.00, 1.39) | 145 | 31.8 | 1.04 (0.86, 1.25) | 1.07 (0.89, 1.29) |
| Very depressed | 33 | 20.1 | 0.86 (0.60, 1.23) | 1.59 (1.21, 2.09) | 29 | 29.2 | 1.15 (0.78, 1.68) | 1.28 (0.92, 1.79) |
HRs were adjusted for menopause status, education, income, exercise, mastectomy, type of therapy received (chemotherapy, radiotherapy), tamoxifen use, Charlson index of comorbidity, menopausal symptoms, BMI, TNM stage, ER/PR status.
Rates of total death or recurrence/breast cancer-specific death, with entry time set as 18-months after diagnosis, were estimated using the Kaplan-Meier method.
Abbreviation: Hazard Ratio, HR; Confidence Interval, CI.
Sleep difficulty and low sleep quality were not significantly associated with total mortality or breast cancer-specific mortality/recurrence. When modeled as constant variables, respective HRs for sleep difficulty and sleep quality, assessed at 18-months, were 0.92 (95% CI: 0.74, 1.14) and 1.11 (95% CI: 0.84, 1.46) for total mortality, and 1.07 (95% CI: 0.84, 1.35) and 1.13 (95% CI: 0.83, 1.53) for breast cancer-specific mortality/recurrence. When treated as time-dependent variables, respective HRs were 1.04 (95% CI: 0.85, 1.28) and 1.02 (95% CI: 0.79, 1.31) for total mortality, and 1.01 (95% CI: 0.80, 1.27) and 0.93 (95% CI: 0.69, 1.25) for breast cancer-specific mortality/recurrence. However, compared to patients who rarely had sleep difficulty, those who sometimes did showed a lower risk of total mortality (HR=0.79, 95% CI: 0.64, 0.97) when sleep difficulty was modeled as a constant variable and a lower risk of breast cancer-specific mortality/recurrence (HR=0.79, 95% CI: 0.63, 0.99) when sleep difficulty was modeled as a time-dependent variable. These associations remained statistically significant after additional adjustment for self-rated health condition, energy level and depression (Table 4).
Table 4.
HRs (95% CI) for total mortality and recurrence/breast cancer-specific mortality in association with sleep attributes for participants who survived for at least 18 months since cancer diagnosis
| Total mortality | Recurrence/breast cancer-specific mortality | |||||||
|---|---|---|---|---|---|---|---|---|
| Death | 18-months HR (95% CI) | 18- and 36-months HR (95% CI) | Recurrence/breast cancer-specific death | 18-months HR (95% CI) | 18- and 36-months HR (95% CI) | |||
| Number | Rate (%) # | Number | Rate (%) # | |||||
| Minimal adjustment * | ||||||||
| Sleep difficulty | ||||||||
| Rarely | 613 | 21.7 | 1.00 | 1.00 | 503 | 33.5 | 1.00 | 1.00 |
| Sometimes | 111 | 18.7 | 0.79 (0.64, 0.97) | 0.86 (0.70, 1.05) | 100 | 47.7 | 0.89 (0.71, 1.10) | 0.79 (0.63, 0.99) |
| Always | 99 | 23.0 | 0.92 (0.74, 1.14) | 1.04 (0.85, 1.28) | 84 | 28.5 | 1.07 (0.84, 1.35) | 1.01 (0.80, 1.27) |
| Sleep quality | ||||||||
| High | 665 | 21.6 | 1.00 | 1.00 | 545 | 30.3 | 1.00 | 1.00 |
| Moderate | 100 | 19.2 | 0.86 (0.70, 1.07) | 0.90 (0.74, 1.09) | 96 | 37.0 | 1.00 (0.80, 1.25) | 0.94 (0.76, 1.17) |
| Low | 58 | 25.2 | 1.11 (0.84, 1.46) | 1.02 (0.79, 1.31) | 46 | 22.2 | 1.13 (0.83, 1.53) | 0.93 (0.69, 1.25) |
| Further adjustment for health condition, energy level and depression | ||||||||
| Sleep difficulty | ||||||||
| Rarely | 613 | 21.7 | 1.00 | 1.00 | 503 | 33.5 | 1.00 | 1.00 |
| Sometimes | 111 | 18.7 | 0.79 (0.64, 0.97) | 0.84 (0.69, 1.02) | 100 | 47.7 | 0.87 (0.70, 1.09) | 0.77 (0.61, 0.96) |
| Always | 99 | 23.0 | 0.89 (0.71, 1.11) | 0.99 (0.81, 1.23) | 84 | 28.5 | 1.02 (0.80, 1.30) | 0.96 (0.76, 1.22) |
| Sleep quality | ||||||||
| High | 665 | 21.6 | 1.00 | 1.00 | 545 | 30.3 | 1.00 | 1.00 |
| Moderate | 100 | 19.2 | 0.87 (0.70, 1.07) | 0.87 (0.72, 1.07) | 96 | 37.0 | 0.98 (0.79, 1.23) | 0.90 (0.73, 1.12) |
| Low | 58 | 25.2 | 1.07 (0.81, 1.42) | 0.97 (0.75, 1.26) | 46 | 22.2 | 1.06 (0.77, 1.46) | 0.88 (0.65, 1.19) |
HRs were adjusted for menopause status, education, income, exercise, mastectomy, type of therapy received (chemotherapy, radiotherapy), tamoxifen use, Charlson index of comorbidity, menopausal symptoms, BMI, TNM stage, ER/PR status.
Rates of total death or recurrence/breast cancer-specific death, with entry time set as 18-months after diagnosis, were estimated using the Kaplan-Meier method.
Abbreviation: Hazard Ratio, HR; Confidence Interval, CI.
Discussion
In this large prospective cohort study, we found 10-year breast cancer survivors had increased prevalence of poor health condition and daily function limitation compared to cancer-free, age-appropriate women from the same area. Several aging related measurements, such as low energy level, depression and poor self-rated health condition, were highly prevalent among breast cancer survivors 18-months after cancer diagnosis and were associated with elevated total mortality or breast cancer-specific mortality/recurrence. We did not find sleep difficulty or poor sleep quality, assessed at 18- and 36-months after cancer diagnosis, to be significantly associated with increased all-cause and breast cancer-specific mortality/recurrence among women with breast cancer, regardless of adjustment for self-rated health condition, energy level and depression.
Emerging evidence shows that cancer survivors experience a high burden of chronic health problems characteristic of aging, including fatigue, depression and sleep disturbance [7], indicative of accelerated aging. In our study, breast cancer survivors had higher prevalence of functional limitations than cancer-free women in the same study area across all age groups, and the age of function limitation appears to shift about 5–10 years earlier among the cancer survivors, supporting the accelerated aging hypothesis. We also found a substantial proportion of breast cancer survivors who reported one or more aging related condition as early as 18-months after cancer diagnosis. In our study, greater than two-thirds of survivors reported reduced energy level at 18-month post-diagnosis, which was substantially higher than that previously reported among the general healthy population (5% to 45%) [34]. In our study, this proportion further increased to 74.1% 36-months after cancer diagnosis. Consistent with our findings, Hagen et al. [35] showed that, 1–2 years after surgery, breast cancer survivors had more self-reported health complaints than healthy controls and experienced fatigue (76% in patients vs 41% in controls), depression (34% vs 17%) and anxiety (20% vs 12%). Their study also showed a higher proportion of breast cancer patients who reported sleep problems in comparison to healthy controls (63% vs 40%). The differences between cancer patients and controls observed in Hagen et al’s study is larger than that we observed in our study. This inconsistency may be due to the different questionnaires used to assess these symptoms and different demographic characteristics (Norway vs China) between the two studies. Both cancer itself, and cancer treatment, may contribute to the accelerated aging process, though the underlying biologic mechanisms have not been well illuminated [36]. However, studies have proposed that modification of lifestyle factors, such as exercise, may help mitigate the physiological changes associated with accelerated aging [6, 7, 23, 37]. Future studies focusing on how the accelerated aging process contributes to cancer prognosis, as well as preventive measurements, are needed to deliver better health care and guidance for the growing breast cancer survivor population, especially for those with accelerated aging issues.
In our study, we found low self-rated health condition, depression and low energy level to be significantly associated with poor breast cancer survival. This is consistent with findings from previous studies, although the latter focused only on individual symptoms [38–41]. Self-reported health conditions and energy levels may reveal patients’ self-perceptions of their disease status, which may not be captured by routine health evaluations [42]. Symptoms of depression after cancer diagnosis may reflect patients’ emotional stress responses, which could lead to elevated levels of the stress hormone cortisol, immunologic changes and chronic inflammation. The biologic changes elicited from depression may increase patients’ vulnerability to cancer progression [43–45]. In addition, depression has been shown to be associated with telomere length shortening, an important signature of accelerated biological aging [46, 47], further supporting our observed positive association between depression and mortality.
Findings from studies focusing on sleep attributes have been mixed. For patients with stages I-III breast cancer, the Nurses’ Health Study showed that longer sleep duration and sleep difficulty, measured within 4 years after cancer diagnosis, were associated with poorer breast cancer survival [28] when sleep variables were treated as constant variables. However, the Women’s Healthy Eating and Living Study observed no such significant associations when time-varying models were applied [27]. None of the previous studies, however, took into consideration the confounding or modifying effects of factors such as sleep quality, energy level and depression, which, as pointed out by Stamatakis, could be major concerns [48]. Also, the potential for reverse causation was not adequately addressed in these studies. In our study, when all these concerns were taken into account, we did not find sleep attributes were associated with breast cancer outcomes.
Our study has several strengths. First, we collected data on aging related measurements at multiple time points, making it possible to model these variables as time-varying exposures instead of using single time-point models. Second, we carefully evaluated potential modifying effects of self-rated health condition, energy level and depression for associations between sleep attributes and breast cancer outcomes. These factors are often entangled with sleep conditions and breast cancer outcomes but were not assessed in previous studies [33,41]. Furthermore, the possibility of reverse causality is minimized in our study by excluding observations during the first 18-months post-cancer diagnosis and by restricting the analyses to patients with stage I to III cancer. Additional strengths of our study include its large sample size (the largest study to date), long follow-up (over 10-years), as well as high participation and follow-up rates. On the other hand, several limitations need to be acknowledged. First, we acknowledge that our study was not designed to specifically assess accelerated aging. We employed several self-reported aging related measurements as a proxy, and no objective geriatric or frailty assessment, or potential biomarker (e.g., inflammatory markers), was measured; thus, misclassification could not be ruled out. Second, information on self-rated health condition, energy level and depression status were collected only by a single question; thus, measurement is not comprehensive. Third, the results from our time-varying analyses should be interpreted with caution, as the possibility of reverse-causality could come into play. Lack of comparable statistics for age-appropriate, general, cancer-free populations for some conditions evaluated is another limitation of our study.
In summary, in this prospective study with more than 10 years of follow-up of breast cancer patients, we found aging related conditions are highly prevalent among breast cancer survivors. A poor self-rated health condition and low energy level at 18- to 36-months post-diagnosis were associated with increased risk of total mortality and breast cancer-specific mortality/recurrence, while depression was associated with elevated risk of total mortality. Our study suggests that a simple assessment of these aging related conditions (i.e., self-rated health conditions, energy levels, and depression symptoms) may identify high-risk breast cancer survivors who may benefit from greater surveillance and possible interventions.
Acknowledgements and Funding Information
We are indebted to the research team and participants of the Shanghai Breast Cancer Survival Study for their contributions to this study. We would like to thank Dr. Shannon Byers, PhD, MS for her assistance on editing and preparing the manuscript. Fei Wang is supported by the program of China Scholarship Council (201806225032). The SBCSS was supported by grants from the US Department of Defense Breast Cancer Research Program (DAMD17-02-1-0607) and the National Cancer Institute (R01 CA118229).
Footnotes
Ethical Standard
The study was performed in accordance with the Declaration of Helsinki, and the study protocols were approved by the Institutional Review Boards of participating institutes, i.e., the Vanderbilt University School of Medicine, Vanderbilt University, Nashville, Tennessee, United States; the Shanghai Cancer Institute, Shanghai, China; and Shanghai Municipal Center for Disease Prevention & Control, Shanghai, China. All participants provided written, informed consent.
Informed Consent
Informed consent was obtained from all individual participants included in the study.
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Disclosure of Potential Conflicts of Interest
The authors declare that they have no conflict of interest.
References
- 1.Ferlay J, Colombet M, Soerjomataram I, et al. (2019) Estimating the global cancer incidence and mortality in 2018: GLOBOCAN sources and methods. Int J Cancer 144:1941–1953. 10.1002/ijc.31937 [DOI] [PubMed] [Google Scholar]
- 2.Bray F, Ferlay J, Soerjomataram I, et al. (2018) Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68:394–424. 10.3322/caac.21492 [DOI] [PubMed] [Google Scholar]
- 3.Allemani C, Weir HK, Carreira H, et al. (2015) Global surveillance of cancer survival 1995–2009: analysis of individual data for 25,676,887 patients from 279 population-based registries in 67 countries (CONCORD-2). Lancet Lond Engl 385:977–1010. 10.1016/S0140-6736(14)62038-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.DeSantis C, Ma J, Bryan L, Jemal A (2014) Breast cancer statistics, 2013. CA Cancer J Clin 64:52–62. 10.3322/caac.21203 [DOI] [PubMed] [Google Scholar]
- 5.Ferlay J, Ervik M, Lam F, et al. (2018) Global Cancer Observatory: Cancer Today. International Agency for Research on Cancer, Lyon, France [Google Scholar]
- 6.Hurria A, Jones L, Muss HB (2016) Cancer Treatment as an Accelerated Aging Process: Assessment, Biomarkers, and Interventions. Am Soc Clin Oncol Educ Book 35:e516–522. 10.1200/EDBK_156160 [DOI] [PubMed] [Google Scholar]
- 7.Henderson TO, Ness KK, Cohen HJ (2014) Accelerated aging among cancer survivors: from pediatrics to geriatrics. Am Soc Clin Oncol Educ Book 34: e423–430. 10.14694/EdBook_AM.2014.34.e423 [DOI] [PubMed] [Google Scholar]
- 8.Chang L, Weiner LS, Hartman SJ, et al. (2019) Breast cancer treatment and its effects on aging. J Geriatr Oncol 10:346–355. 10.1016/j.jgo.2018.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Alfano CM, Peng J, Andridge RR, et al. (2017) Inflammatory Cytokines and Comorbidity Development in Breast Cancer Survivors Versus Noncancer Controls: Evidence for Accelerated Aging? J Clin Oncol 35:149–156. 10.1200/JCO.2016.67.1883 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hewitt M, Greenfield S, Stovall E (2006) From Cancer Patient to Cancer Survivor: Lost in Transition. the National Academies Press, Washington DC [Google Scholar]
- 11.Cella D, Davis K, Breitbart W, et al. (2001) Cancer-related fatigue: prevalence of proposed diagnostic criteria in a United States sample of cancer survivors. J Clin Oncol 19:3385–3391. 10.1200/JCO.2001.19.14.3385 [DOI] [PubMed] [Google Scholar]
- 12.Whitehead LC, Unahi K, Burrell B, Crowe MT (2016) The Experience of Fatigue Across Long-Term Conditions: A Qualitative Meta-Synthesis. J Pain Symptom Manage 52:131–143.e1. 10.1016/j.jpainsymman.2016.02.013 [DOI] [PubMed] [Google Scholar]
- 13.Bruera E, Yennurajalingam S (2010) Challenge of managing cancer-related fatigue. J Clin Oncol 28:3671–3672. 10.1200/JCO.2010.29.8984 [DOI] [PubMed] [Google Scholar]
- 14.Singer S, Kuhnt S, Zwerenz R, et al. (2011) Age- and sex-standardised prevalence rates of fatigue in a large hospital-based sample of cancer patients. Br J Cancer 105:445–451. 10.1038/bjc.2011.251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rotonda C, Guillemin F, Bonnetain F, et al. (2013) Factors associated with fatigue after surgery in women with early-stage invasive breast cancer. The Oncologist 18:467–475. 10.1634/theoncologist.2012-0300 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hu L-Y, Liu C-J, Yeh C-M, et al. (2018) Depressive disorders among patients with gastric cancer in Taiwan: a nationwide population-based study. BMC Psychiatry 18:. 10.1186/s12888-018-1859-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fan C-Y, Chao H-L, Lin C-S, et al. (2018) Risk of depressive disorder among patients with head and neck cancer: A nationwide population-based study. Head Neck 40:312–323. 10.1002/hed.24961 [DOI] [PubMed] [Google Scholar]
- 18.Ho S-Y, Rohan KJ, Parent J, et al. (2015) A longitudinal study of depression, fatigue, and sleep disturbances as a symptom cluster in women with breast cancer. J Pain Symptom Manage 49:707–715. 10.1016/j.jpainsymman.2014.09.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Trudel-Fitzgerald C, Savard J, Ivers H (2013) Which symptoms come first? Exploration of temporal relationships between cancer-related symptoms over an 18-month period. Ann Behav Med 45:329–337. 10.1007/s12160-012-9459-1 [DOI] [PubMed] [Google Scholar]
- 20.Davidson JR, MacLean AW, Brundage MD, Schulze K (2002) Sleep disturbance in cancer patients. Soc Sci Med 1982 54:1309–1321 [DOI] [PubMed] [Google Scholar]
- 21.Palesh OG, Roscoe JA, Mustian KM, et al. (2010) Prevalence, demographics, and psychological associations of sleep disruption in patients with cancer: University of Rochester Cancer Center-Community Clinical Oncology Program. J Clin Oncol 28:292–298. 10.1200/JCO.2009.22.5011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Savard J, Simard S, Blanchet J, et al. (2001) Prevalence, clinical characteristics, and risk factors for insomnia in the context of breast cancer. Sleep 24:583–590. 10.1093/sleep/24.5.583 [DOI] [PubMed] [Google Scholar]
- 23.Armenian SH, Gibson CJ, Rockne RC, Ness KK (2019) Premature Aging in Young Cancer Survivors. J Natl Cancer Inst 111:226–232. 10.1093/jnci/djy229 [DOI] [PubMed] [Google Scholar]
- 24.Irwin MR, Wang M, Ribeiro D, et al. (2008) Sleep loss activates cellular inflammatory signaling. Biol Psychiatry 64:538–540. 10.1016/j.biopsych.2008.05.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lange T, Dimitrov S, Born J (2010) Effects of sleep and circadian rhythm on the human immune system. Ann N Y Acad Sci 1193:48–59. 10.1111/j.1749-6632.2009.05300.x [DOI] [PubMed] [Google Scholar]
- 26.Sephton S, Spiegel D (2003) Circadian disruption in cancer: a neuroendocrine-immune pathway from stress to disease? Brain Behav Immun 17:321–328 [DOI] [PubMed] [Google Scholar]
- 27.Marinac CR, Nelson SH, Flatt SW, et al. (2017) Sleep duration and breast cancer prognosis: perspectives from the Women’s Healthy Eating and Living Study. Breast Cancer Res Treat 162:581–589. 10.1007/s10549-017-4140-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Trudel-Fitzgerald C, Zhou ES, Poole EM, et al. (2017) Sleep and survival among women with breast cancer: 30 years of follow-up within the Nurses’ Health Study. Br J Cancer 116:1239–1246. 10.1038/bjc.2017.85 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Palesh O, Aldridge-Gerry A, Zeitzer JM, et al. (2014) Actigraphy-measured sleep disruption as a predictor of survival among women with advanced breast cancer. Sleep 37:837–842. 10.5665/sleep.3642 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Shu XO, Zheng Y, Cai H, et al. (2009) Soy food intake and breast cancer survival. JAMA 302:2437–2443. 10.1001/jama.2009.1783 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 40:373–383 [DOI] [PubMed] [Google Scholar]
- 32.Xu WH, Zheng W, Xiang YB, et al. (2004) Soya food intake and risk of endometrial cancer among Chinese women in Shanghai: population based case-control study. BMJ 328:1285 10.1136/bmj.38093.646215.AE [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hosmer DWJ, Lemeshow S (1999) Applied Survival Analysis: Regression Modeling of Time-to-Event Data, 2nd Edition. John Wiley & Sons Inc, New York, NY [Google Scholar]
- 34.Finsterer J, Mahjoub SZ (2014) Fatigue in Healthy and Diseased Individuals. Am J Hosp Palliat Med 31:562–575. 10.1177/1049909113494748 [DOI] [PubMed] [Google Scholar]
- 35.Hagen KB, Aas T, Kvaløy JT, et al. (2016) Fatigue, anxiety and depression overrule the role of oncological treatment in predicting self-reported health complaints in women with breast cancer compared to healthy controls. The Breast 28:100–106. 10.1016/j.breast.2016.05.005 [DOI] [PubMed] [Google Scholar]
- 36.Hill A, Sadda J, LaBarge MA, Hurria A (2019) How cancer therapeutics cause accelerated aging: Insights from the hallmarks of aging. J Geriatr Oncol S1879406818303965. 10.1016/j.jgo.2019.03.007 [DOI] [PubMed] [Google Scholar]
- 37.Rebelo-Marques A, De Sousa Lages A, Andrade R, et al. (2018) Aging Hallmarks: The Benefits of Physical Exercise. Front Endocrinol 9:. 10.3389/fendo.2018.00258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Groenvold M, Petersen MA, Idler E, et al. (2007) Psychological distress and fatigue predicted recurrence and survival in primary breast cancer patients. Breast Cancer Res Treat 105:209–219. 10.1007/s10549-006-9447-x [DOI] [PubMed] [Google Scholar]
- 39.Hjerl K, Andersen EW, Keiding N, et al. (2003) Depression as a prognostic factor for breast cancer mortality. Psychosomatics 44:24–30. 10.1176/appi.psy.44.1.24 [DOI] [PubMed] [Google Scholar]
- 40.Watson M, Haviland JS, Greer S, et al. (1999) Influence of psychological response on survival in breast cancer: a population-based cohort study. Lancet 354:1331–1336. 10.1016/s0140-6736(98)11392-2 [DOI] [PubMed] [Google Scholar]
- 41.Eng JA, Clough-Gorr K, Cabral HJ, Silliman RA (2015) Predicting 5- and 10-year survival in older women with early-stage breast cancer: self-rated health and walking ability. J Am Geriatr Soc 63:757–762. 10.1111/jgs.13340 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Idler EL, Benyamini Y (1997) Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav 38:21–37 [PubMed] [Google Scholar]
- 43.Vodermaier A, Linden W, Rnic K, et al. (2014) Prospective associations of depression with survival: a population-based cohort study in patients with newly diagnosed breast cancer. Breast Cancer Res Treat 143:373–384. 10.1007/s10549-013-2795-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.McEwen BS, Seeman T (1999) Protective and damaging effects of mediators of stress. Elaborating and testing the concepts of allostasis and allostatic load. Ann N Y Acad Sci 896:30–47. 10.1111/j.1749-6632.1999.tb08103.x [DOI] [PubMed] [Google Scholar]
- 45.Armaiz-Pena GN, Lutgendorf SK, Cole SW, Sood AK (2009) Neuroendocrine modulation of cancer progression. Brain Behav Immun 23:10–15. 10.1016/Surl [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Simon NM, Smoller JW, McNamara KL, et al. (2006) Telomere shortening and mood disorders: preliminary support for a chronic stress model of accelerated aging. Biol Psychiatry 60:432–435. 10.1016/j.biopsych.2006.02.004 [DOI] [PubMed] [Google Scholar]
- 47.Verhoeven JE, Révész D, Epel ES, et al. (2014) Major depressive disorder and accelerated cellular aging: results from a large psychiatric cohort study. Mol Psychiatry 19:895–901. 10.1038/mp.2013.151 [DOI] [PubMed] [Google Scholar]
- 48.Stamatakis KA, Punjabi NM (2007) Long sleep duration: a risk to health or a marker of risk? Sleep Med Rev 11:337–339. 10.1016/j.smrv.2007.07.006 [DOI] [PMC free article] [PubMed] [Google Scholar]