Breast cancer (BC) is the most frequently diagnosed cancer in women worldwide, representing 12% of all new cancer cases and 25% of all cancers in women.1 American women are having BC at a rate of 12% in lifetime risk, although 89% of BC patients survive more than 5 years.2 According to the American Cancer Society, in 2016, the incidence of BC increased 3-fold over the last 30 years.3 The growing populations of BC patients need to identify crucial factors that lead to improved outcomes for BC patients.
Physical activity (PA) is a known contributor to reduced BC-specific mortality (BCM) and all-cause mortality (AM), increased quality of life and physical function, and reduced fatigue and edema, which are related to cancer treatments.4,5 To maintain health, PA guidelines recommend participating in moderate PA at least 250 min/wk.6 There are several studies suggesting that PA helps reduce the risk of death in BC patients.5,7,8 A recent meta-analysis of 22 prospective cohort studies reported an inverse association between PA and mortality in BC survivors.9
Although participants of PA reduce their risk of death, it is important to know what the most effective level of PA intensity is for BC patients and how much the amount of PA changes after BC diagnosis. Some studies showed that vigorous PA and increased PA amount after BC diagnosis significantly reduced mortality; others found that moderate PA and maintaining PA amount after BC diagnosis reduced BCM. However, there is limited evidence showing that the intensity of PA and changes in the amount of PA influence mortality after BC diagnosis, which could be useful and applied to clinical practice in exercise prescription. Therefore, the purpose of this meta-analysis was to examine the association between PA and the risk of death in BC survivors and its dependency on PA intensity and amount of PA changes after BC diagnosis.
Methods
Search Strategy
We followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for the selection of eligible studies in this meta-analysis.10 We used the electronic databases MEDLINE and EMBASE to identify eligible studies published in English from January 1970 to February 2017. Reference lists from original and review articles were also examined to find additional relevant studies. The search strategy used the terms “physical activity,” “exercise,” “breast cancer,” “mortality,” “cancer-specific mortality,” and “survival.” Those key terms were used alone and in various combinations.
Eligibility Criteria
The studies were selected in an unblended standardized manner by 1 author (J.L.) and 1 research assistant (S.L.) who supported the selection process. Any discrepancies on study eligibility were resolved through consensus. The studies included in this meta-analysis met the following criteria: (1) studies that presented outcomes of BC or AM, (2) studies that offered relative risk (RR) of mortality and the confidence intervals (CIs), and (3) studies in which the exposure of interest was PA measured before or after diagnosis. If data sets were overlapped or duplicated, the most complete data sets were included.
Data Extraction
Two authors (J.L. and S.L.) independently assessed the records for eligibility according to the guidelines for the Meta-analysis of Observational Studies in Epidemiology.11 Any discrepancies in the selection of eligible studies were reevaluated by further discussion to reach a consensus. For each eligible study, basic information regarding the name of the first author, country, publication year, sample size, study recruitment period, follow-up period, study design, criteria of the cause of death, PA assessment, RR, 95% CIs, and adjustment factors are presented in Table 1. We assessed the quality of the study according to the Newcastle-Ottawa quality assessment scale.27 The following quality items were assessed: clarity of PA measurement timing (prediagnosis or postdiagnosis), PA type, adjustments for confounding factors (age, stage, tumor differentiation, etc), duration of follow-up, and study end points.
;)
Table 1: Characteristics of Selected Studies: Prediagnosis Physical Activity and Mortality
Statistical Analysis
To determine the cutoff points of PA amount, we calculated the averages of PA amount for 3 groups—low amount of PA, moderate amount of PA, and high amount of PA—reported by all included studies. We defined the cutoff points as low amount of PA (<300 min/wk), moderate amount of PA (300-500 min/wk), and high amount of PA (>500 min/wk). Fixed- and random-effect model parameters and 95% CIs were computed from each study RR, which were directly extracted from the published reports, by using the Comprehensive Meta-Analysis version 1.25 (Biostatic, Inc, Englewood, New Jersey) software program. The Q statistic was used to assess the statistical heterogeneity across the sampled studies (P < .10 was significant), and the I2 statistic was used to quantify inconsistency. Fixed effect models were used if the meta-analysis data were homogenous (P < .1), but random effect models were used if the meta-analysis data were heterogeneous (P > .1). A visual inspection of the funnel plot was performed to assess publication bias. A P < .05 was considered statistically significant. All statistical analyses were performed using Comprehensive Meta-Analysis version 1.25.
Results
Literature Search
Details of the studies' selection process for this meta-analysis are presented in Figure 1. A total of 545 studies were investigated during our initial search. Forty-seven studies, which were review studies, case studies, and duplicated studies, were excluded while screening the title of the articles. Twenty-three studies were excluded because they did not report survival outcomes or the level of PA. Finally, 24 studies were included for this meta-analysis. Basic descriptions of those selected studies are in Tables 1 and 2.
Figure 1: Selection process for meta-analysis. Details of the selection process are presented.
Study Characteristics
All 24 studies were included for this meta-analysis: 17 studies for prediagnosis analysis,12–26,28 12 studies for postdiagnosis analysis,8,18,22,28–36 and 3 studies for both prediagnosis and postdiagnosis analyses.18,24,28 Our prediagnosis independent variables were moderate amount of PA, vigorous amount of PA, moderate-intensity PA × moderate amount of PA, and moderate-intensity PA × high amount of PA. Our postdiagnosis independent variables were moderate amount of PA, high amount of PA, moderate-intensity PA × moderate amount of PA, moderate-intensity PA × high amount of PA, vigorous-intensity PA × moderate amount of PA, vigorous-intensity PA × high amount of PA, moderate- to vigorous-intensity PA × moderate amount of PA, moderate- to vigorous-intensity PA × high amount of PA, increased and decreased PA, and whether the patients met the current PA guidelines. We analyzed the relationship between these prediagnosis and postdiagnosis PA independent variables and each dependent variable, BCM and AM. Additional postdiagnosis independent variables, decreased and increased PA, were only tested against AM. The amount of PA was divided into 3 groups based on an average amount of PA found in the studies: low amount of PA (<300 min/wk), moderate amount of PA (300-500 min/wk), and high amount PA (>500 min/wk). PA intensity was reported in the selected studies by self-reported responses to PA questionnaires. The PA intensity was divided into low intensity, moderate intensity, moderate to vigorous intensity, and vigorous intensity. In addition, all the selected studies provided the measurement timing of PA as both before and after BC diagnosis. All findings were presented based on the measurement timing: prediagnosis and postdiagnosis.
Table 2: Characteristics of Selected Studies: Postdiagnosis Physical Activity and Mortality
Prediagnosis Amount of PA and Mortality
Fifteen studies were included to analyze the associations between prediagnosis PA and BCM and prediagnosis PA and AM (Figure 2). This meta-analysis found that the association between the amount of prediagnosis PA with BCM had an RR of 0.87 (95% CI, 0.78-0.96; P < .05) in moderate versus low amount of PA and 0.80 (95% CI, 0.72-0.89; P < .05) in high versus low amount of PA. The AM associated with the amount of prediagnosis PA had an RR of 0.78 (95% CI, 0.71-0.85; P < .05) in moderate versus low amount of PA and 0.74 (95% CI, 0.68-0.80; P < .05) in high versus low amount of PA. There was no statistically significant heterogeneity among the studies in any analysis. There was neither evidence of publication bias in the analyses nor apparent influence of unpublished data in any analyses using the trim and fill method.
Figure 2: Prediagnosis amounts of PA versus BCM and AM. Results of studies for the associations between prediagnosis amounts of PA and BCM and prediagnosis amounts of PA and AM.
Postdiagnosis Amount of PA and Mortality
Twelve cohort studies reported associations between postdiagnosis amount of PA and BCM and AM (Figure 3). Breast-cancer–specific mortality was associated with moderate versus low amount of postdiagnosis PA (0.72 [95% CI, 0.62-0.83], P < .05]) and high versus low amount of PA (0.71 [95% CI, 0.64-0.78], P < .05). The association between AM and postdiagnosis PA had an RR of 0.74 (95% CI, 0.66-0.83; P < .05) in moderate versus low amount of PA and 0.61 (95% CI, 0.56-0.66; P < .05) in high versus low amount of PA. There was no statistically significant heterogeneity among the studies in any analysis. There was no evidence of publication bias in any analyses. In addition, there was no apparent influence of unpublished data in any analyses using the trim and fill method.
Figure 3: Postdiagnosis amounts of PA versus BCM and AM. Results of studies for the associations between postdiagnosis amounts of PA and BCM and postdiagnosis amounts of PA and AM.
Prediagnosis Amount of PA and Mortality According to PA Intensity
This meta-analysis provided results of studies for the associations between prediagnosis amount of PA and BCM and prediagnosis amount of PA and AM based on PA intensity (Figure 4). In moderate-intensity PA, BCM showed an RR of 0.65 (95% CI, 0.51-0.83; P < .05) in moderate versus low amount of prediagnosis PA and 0.62 (95% CI, 0.46-0.84; P < .05) in high versus low amount of prediagnosis PA. The association between AM and moderate-intensity prediagnosis PA showed an RR of 0.67 (95% CI, 0.55-0.81; P < .05) in moderate versus low amount of PA and 0.73 (95% CI, 0.58-0.91; P < .05) in high versus low amount of PA.
Figure 4: Associations between prediagnosis amounts of PA and BCM and MA according to PA intensity. Results of studies for the association between prediagnosis amounts of PA and BCM and prediagnosis amounts of PA and MA based on PA intensity.
This analysis only included moderate versus low amount of PA for the association between vigorous-intensity prediagnosis PA and BCM and AM because of insufficient studies to allow for an additional analysis of high versus low amount of prediagnosis PA with vigorous-intensity prediagnosis PA. Breast-cancer–specific mortality was associated with moderate versus low amount of prediagnosis PA (0.75 [95% CI, 0.57-0.98], P < .05]), but AM did not have a significant association with moderate versus low amount of prediagnosis PA (0.90 [95% CI, 0.74-1.09], P = .29). The analyses did not show evidence of publication bias or any apparent influence of unpublished data using the trim and fill method.
Postdiagnosis Amount of PA and Mortality According to PA Intensity
The results for the association between postdiagnosis amount of PA and BMC and postdiagnosis amount of PA and AM based on PA intensity are shown in Figures 5 and 6. Breast-cancer–specific mortality in moderate-intensity postdiagnosis PA showed an RR of 0.69 (95% CI, 0.47-0.995; P < .05) when associated with moderate versus low amount of postdiagnosis PA and 0.58 (95% CI, 0.39-0.90; P < .05) when associated with high versus low amount of postdiagnosis PA. The association of moderate-intensity postdiagnosis PA with AM had an RR of 0.67 (95% CI, 0.55-0.83; P < .05) in moderate versus low amount of PA and 0.53 (95% CI, 0.38-0.75; P < .05) in high versus low amount of PA.
Figure 5: Associations between postdiagnosis amounts of PA and BCM and AM according to PA intensity. Results of studies for the association between postdiagnosis amounts of PA and BCM and postdiagnosis amounts of PA and AM based on PA intensity.
Figure 6: Associations between postdiagnosis amounts of PA and BCM and AM according to PA intensity. Results of studies for the association between postdiagnosis amounts of PA and BCM and postdiagnosis amounts of PA and AM based on PA intensity.
The association between vigorous-intensity postdiagnosis PA and BCM showed an RR of 0.76 (95% CI, 0.59-0.98; P < .05) in moderate versus low amount of postdiagnosis PA, but BCM did not have a significant association with high versus low amount of postdiagnosis PA (RR, 1.07 [95% CI, 0.72-1.58]; P < .05). The association of AM with vigorous-intensity postdiagnosis PA had an RR of 0.76 (95% CI, 0.59-0.98; P < .05) in moderate versus low amount of postdiagnosis PA, but AM did not show a significant association with an RR of 0.91 (95% CI, 0.71-1.17; P = .02) for high versus low amount of postdiagnosis PA.
Moderate- to vigorous-intensity postdiagnosis PA and BCM were not significantly associated, with an RR of 0.91 (95% CI, 0.68-1.22; P = .2) in moderate versus low amount of postdiagnosis PA and 0.90 (95% CI, 0.51-1.58; P = .72) in high versus low amount of postdiagnosis PA. The association of AM with postdiagnosis moderate- to vigorous-intensity PA had an RR of 0.60 (95% CI, 0.43-0.86; P < .05) in high versus low amount of postdiagnosis PA, but AM in moderate- to vigorous-intensity postdiagnosis PA was not significantly associated, with an RR of 0.83 (95% CI, 0.60-1.16; P = .02) in moderate versus low amount of postdiagnosis PA. No evidence of publication bias and no apparent influence of unpublished data using the trim and fill method were observed in these analyses.
Changes in PA After BC Diagnosis and Mortality
The results for changes in PA after diagnosis versus BCM and AM are shown in Figure 7. Decreased PA postdiagnosis was associated with an RR of 2.36 (95% CI, 1.09-5.12; P < .05) in AM. The association between postdiagnosis increased PA and AM was not significant, with an RR of 0.99 (95% CI, 0.98-1.00; P = .05). There were insufficient studies to test BCM for postdiagnosis decreased PA and postdiagnosis increased PA.
Figure 7: Associations between changes in PA after diagnosis and AM. Results of studies for the association between changes in PA after diagnosis and AM.
When patients met the PA guidelines postdiagnosis, BCM and AM had RRs of 0.79 (95% CI, 0.68-0.93; P < .05) and 0.72 (95% CI, 0.66-0.80; P < .05), respectively. The analyses did not show evidence of publication bias or any apparent influence of unpublished data using the trim and fill method.
Discussion
This meta-analysis of 24 studies examined the association between PA and mortality. In addition, we conducted this meta-analysis to estimate the most effective intensity and amount of PA and postdiagnosis changes in PA for improved BC outcomes. We found that prediagnosis and postdiagnosis PAs were associated with reduced BCM and M. Additional findings from this meta-analysis show that moderate-intensity PA and high amount of PA, both prediagnosis and postdiagnosis, had more beneficial effects on the risk of death than other intensities and amounts of PA. However, decreased PA postdiagnosis showed the worst risk of mortality. Consequently, high amounts of moderate-intensity PA, both prediagnosis and postdiagnosis, might be required to reduce the risk of BC death and prevent a decrease in PA postdiagnosis.
Breast cancer patients who participated in moderate and high amounts of PA, both prediagnosis and postdiagnosis, had reductions in BCM and AM compared with those who participated in low amounts of PA. The risk of BCM and AM among those who participated in high amounts of prediagnosis PA was 0.80 and 0.74 times, respectively, in comparison with the risks of BCM and M among those who participated in low amounts of PA. Those who had high amounts of postdiagnosis PA had 0.71 and 0.61 times the risk of BCM and AM, respectively, compared with those who participated in low amounts of PA. In addition, the risk of BCM and AM among those who participated in moderate amounts of prediagnosis PA was 0.87 and 0.78 times, respectively, in comparison with the risk of BCM and AM among those who participated in low amounts of PA. Those who had moderate amounts of postdiagnosis PA had 0.72 and 0.74 times the risk of BCM and AM, respectively, compared with those who participated in low amounts of PA. The results of this meta-analysis study are consistent with previous meta-analyses.5,37 Patients who participated in PA had better mortality outcomes, a fact that is supported by previous studies because of a major role of PA in decreasing the risk of cardiovascular disease in women.38,39 Our meta-analysis may suggest that more than 300 min/wk of PA can lead to favorable outcomes of survival for BC patients. However, it should be noted that each included study in this meta-analysis had mixed types of PA, such as recreational PA, occupational PA, and exercise.
This meta-analysis study tested the association between PA and the risk of mortality according to intensity. This meta-analysis indicated that BC patients who participated in moderate and high amounts of PA at both moderate and vigorous intensities had reductions in BCM and AM, compared with those who participated in low amounts of PA, both prediagnosis and postdiagnosis. The most effective intensity and amount of PA postdiagnosis were moderate intensity and high amounts of PA. The risk of BCM and AM among those who participated in moderate-intensity PA and high amounts of PA postdiagnosis was 0.58 and 0.53 times, respectively, in comparison with the risk of BCM and AM among those who participated in moderate-intensity PA and low amounts of PA. The average of moderate and high amounts of PA was between 300 and 500 min/wk and more than 500 min/wk, respectively. Thus, this meta-analysis suggests that the PA amount and intensity in BC patients may be higher than current recommendations, that is, moderate to vigorous PA for a minimum of 30 min and 5 days per week or 150 min/wk. Considering the results of this meta-analysis, the recommendation should be increased from 150 to 300 min/wk to increase the chances of BC survival. This meta-analysis proposes that at least 300 min/wk of moderate-intensity PA should be recommended to reduce mortality in BC patients.
It may be dangerous for patients to decrease PA after BC diagnosis. This meta-analysis found that BC patients who decreased PA postdiagnosis had a poorer risk of mortality. On the other hand, BC patients who met the PA recommendations and increased PA postdiagnosis had better mortality outcomes, although the difference was not significant in increased PA postdiagnosis. The reason for no significant differences in the mortality outcomes in increased PA postdiagnosis is that each study had different PA references, such as inactivity after diagnosis,30 decreasing PA after diagnosis,35 or maintained PA after diagnosis.28 Considering those differences, further studies with more consistent data sets should be conducted.
There are possibly several mechanisms that explain the inverse association between PA and the risks of mortality. First, PA changes adipocytokine profiles, increasing anti-inflammatory adipocytokines and decreasing inflammatory adipocytokines.40 The role of inflammation in cancer is physiologically complex; therefore, any further discussion is beyond the scope of this study. Second, PA is related to decreased insulin resistance. Third, PA helps reduce fat accumulation in the breast.41 It is important to note that the relative proportion of epithelial cells to adipose tissue in the breast may determine the risk of BC.42 Women who participated in running and walking had less fat accumulation and more epithelial cells than those who did not participate in any PA.26 Preventing adipose tissue accumulation might help reduce BCM and AM.
Several limitations of this meta-analysis should be addressed. First, each individual study in this meta-analysis had heterogeneous measures of PA and length of follow-up. Each study in this meta-analysis used different PA questionnaires; the follow-up periods ranged 2.9 to 12.7 years. Second, the included studies had different cutoff points for PA amount categories, including low, moderate, and high. Thus, this meta-analysis used calculated average cutoff points of PA amount categories. Third, this meta-analysis did not limit the adjusted values, such as disease state at diagnosis, stage of BC at death, and type of BC treatment.
This meta-analysis of 24 studies concluded that prediagnosis and postdiagnosis PAs were associated with a reduction in the risk of both BCM and AM. In addition, decreasing PA may lead to a worse prognosis of BC. Patients should participate in increased PA. This study's findings provided clinical implications for exercise prescription to get the most effective PA amount and intensity for BC patients. We suggest that BC patients should get at least 300 min/wk of moderate-intensity PA to improve prognosis and outcomes. Further studies must be conducted to confirm the most effective PA recommendations, including intensity and frequency, for BC patients.
References
1. Bandera EV, Fay SH, Giovannucci E, et al. The use and interpretation of anthropometric measures in cancer epidemiology: a perspective from the world cancer research fund international continuous update project.
Int J Cancer. 2016;139(11):2391–2397.
2. DeSantis CE, Fedewa SA, Goding Sauer A, Kramer JL, Smith RA, Jemal A. Breast cancer statistics, 2015: convergence of incidence rates between black and white women.
CA Cancer J Clin. 2016;66(1):31–42.
3. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016.
CA Cancer J Clin. 2016;66(1):7–30.
4. Warburton DE, Nicol CW, Bredin SS. Health benefits of physical activity: the evidence.
CMAJ. 2006;174(6):801–809.
5. Ibrahim EM, Al-Homaidh A. Physical activity and survival after breast cancer diagnosis: meta-analysis of published studies.
Med Oncol. 2011;28(3):753–765.
6. Schmitz KH, Courneya KS, Matthews C, et al. American College of Sports Medicine roundtable on exercise guidelines for cancer survivors.
Med Sci Sports Exerc. 2010;42(7):1409–1426.
7. Borch KB, Braaten T, Lund E, Weiderpass E. Physical activity before and after breast cancer diagnosis and survival—the Norwegian women and cancer cohort study.
BMC Cancer. 2015;15:967.
8. LeMasters TJ, Madhavan SS, Sambamoorthi U, Kurian S. Health behaviors among breast, prostate, and colorectal cancer survivors: a US population-based case-control study, with comparisons by cancer type and gender.
J Cancer Surviv. 2014;8(3):336–348.
9. Lahart IM, Metsios GS, Nevill AM, Carmichael AR. Physical activity, risk of death and recurrence in breast cancer survivors: a systematic review and meta-analysis of epidemiological studies.
Acta Oncol. 2015;54(5):635–654.
10. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration.
J Clin Epidemiol. 2009;62(10):e1–e34.
11. Stroup DF, Berlin J A, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
JAMA. 2000;283(15):2008–2012.
12. Abrahamson PE, Gammon MD, Lund MJ, et al. Recreational physical activity and survival among young women with breast cancer.
Cancer. 2006;107(8):1777–1785.
13. Cleveland RJ, Eng SM, Stevens J, et al. Influence of prediagnostic recreational physical activity on survival from breast cancer.
Eur J Cancer Prev. 2012;21(1):46–54.
14. Dal Maso L, Zucchetto A, Talamini R, et al. Effect of obesity and other lifestyle factors on mortality in women with breast cancer.
Int J Cancer. 2008;123(9):2188–2194.
15. Emaus A, Veierod MB, Tretli S, et al. Metabolic profile, physical activity, and mortality in breast cancer patients.
Breast Cancer Res Treat. 2010;121(3):651–660.
16. Enger SM, Bernstein L. Exercise activity, body size and premenopausal breast cancer survival.
Br J Cancer. 2004;90(11):2138–2141.
17. Friedenreich CM, Gregory J, Kopciuk KA, Mackey JR, Courneya KS. Prospective cohort study of lifetime physical activity and breast cancer survival.
Int J Cancer. 2009;124(8):1954–1962.
18. de Glas NA, Fontein DB, Bastiaannet E, et al. Physical activity and survival of postmenopausal, hormone receptor-positive breast cancer patients: results of the Tamoxifen Exemestane Adjuvant Multicenter Lifestyle study.
Cancer. 2014;120(18):2847–2854.
19. Hellmann SS, Thygesen LC, Tolstrup JS, Grønbaek M. Modifiable risk factors and survival in women diagnosed with primary breast cancer: results from a prospective cohort study.
Eur J Cancer Prev. 2010;19(5):366–373.
20. Irwin ML, Smith AW, McTiernan A, et al. Influence of pre- and postdiagnosis physical activity on mortality in breast cancer survivors: the health, eating, activity, and lifestyle study.
J Clin Oncol. 2008;26(24):3958–3964.
21. Keegan TH, Milne RL, Andrulis IL, et al. Past recreational physical activity, body size, and all-cause mortality following breast cancer diagnosis: results from the Breast Cancer Family Registry.
Breast Cancer Res Treat. 2010;123(2):531–542.
22. Rohan TE, Fu W, Hiller JE. Physical activity and survival from breast cancer.
Eur J Cancer Prev. 1995;4(5):419–424.
23. Schmidt ME, Chang-Claude J, Vrieling A, et al. Association of pre-diagnosis physical activity with recurrence and mortality among women with breast cancer.
Int J Cancer. 2013;133(6):1431–1440.
24. Tao MH, Hainaut P, Marian C, et al. Association of prediagnostic physical activity with survival following breast cancer diagnosis: influence of TP53 mutation status.
Cancer Causes Control. 2013;24(12):2177–2186.
25. West-Wright CN, Henderson KD, Sullivan-Halley J, et al. Long-term and recent recreational physical activity and survival after breast cancer: the California Teachers Study.
Cancer Epidemiol Biomarkers Prev. 2009;18(11):2851–2859.
26. Williams PT. Breast cancer mortality vs. exercise and breast size in runners and walkers.
PLoS One. 2013;8(12):e80616.
27. Wells KF. Newcastle disease.
Can J Comp Med Vet Sci. 1948;12(4):101–104.
28. Irwin ML, McTiernan A, Manson JE, et al. Physical activity and survival in postmenopausal women with breast cancer: results from the women's health initiative.
Cancer Prev Res (Phila). 2011;4(4):522–529.
29. Beasley JM, Kwan ML, Chen WY, et al. Meeting the physical activity guidelines and survival after breast cancer: findings from the after breast cancer pooling project.
Breast Cancer Res Treat. 2012;131(2):637–643.
30. Bertram LA, Stefanick ML, Saquib N, et al. Physical activity, additional breast cancer events, and mortality among early-stage breast cancer survivors: findings from the WHEL Study.
Cancer Causes Control. 2011;22(3):427–435.
31. Bradshaw PT, Ibrahim JG, Khankari N, et al. Post-diagnosis physical activity and survival after breast cancer diagnosis: the Long Island Breast Cancer Study.
Breast Cancer Res Treat. 2014;145(3):735–742.
32. Chen X, Lu W, Zheng W, et al. Exercise after diagnosis of breast cancer in association with survival.
Cancer Prev Res (Phila). 2011;4(9):1409–1418.
33. Holick CN, Newcomb PA, Trentham-Dietz A, et al. Physical activity and survival after diagnosis of invasive breast cancer.
Cancer Epidemiol Biomarkers Prev. 2008;17(2):379–386.
34. Holmes MD, Chen WY, Feskanich D, Kroenke CH, Colditz GA. Physical activity and survival after breast cancer diagnosis.
JAMA. 2005;293(20):2479–2486.
35. Patterson RE, Saquib N, Natarajan L, et al. Improvement in self-reported physical health predicts longer survival among women with a history of breast cancer.
Breast Cancer Res Treat. 2011;127(2):541–547.
36. Sternfeld B, Weltzien E, Quesenberry CP Jr, et al. Physical activity and risk of recurrence and mortality in breast cancer survivors: findings from the LACE study.
Cancer Epidemiol Biomarkers Prev. 2009;18(1):87–95.
37. Wu Y, Zhang D, Kang S. Physical activity and risk of breast cancer: a meta-analysis of prospective studies.
Breast Cancer Res Treat. 2013;137(3):869–882.
38. Kodama S, Saito K, Yachi Y, et al. Association between serum uric acid and development of type 2 diabetes.
Diabetes Care. 2009;32(9):1737–1742.
39. Saquib N, Pierce JP, Saquib J, et al. Poor physical health predicts time to additional breast cancer events and mortality in breast cancer survivors.
Psychooncology. 2011;20(3):252–259.
40. Friedenreich CM. The role of physical activity in breast cancer etiology.
Semin Oncol. 2010;37(3):297–302.
41. Hoffman-Goetz L, Apter D, Demark-Wahnefried W, Goran MI, McTiernan A, Reichman ME. Possible mechanisms mediating an association between physical activity and breast cancer.
Cancer. 1998;83(3 suppl):621–628.
42. Pettersson A, Hankinson SE, Willett WC, Lagiou P, Trichopoulos D, Tamimi RM. Nondense mammographic area and risk of breast cancer.
Breast Cancer Res. 2011;13(5):R100.
