- Email: [email protected]
Cardiorespiratory fitness, physical activity and cancer mortality in men
Accepted Manuscript Cardiorespiratory fitness, physical activity and cancer mortality in men
Baruch Vainshelboim, Jan Müller, Ricardo M. Lima, Kevin T. Nead, Cariad Chester, Khin Chan, Peter Kokkinos, Jonathan Myers PII: DOI: Reference:
S0091-7435(17)30134-2 doi: 10.1016/j.ypmed.2017.04.014 YPMED 4992
To appear in:
Preventive Medicine
Received date: Revised date: Accepted date:
11 January 2017 7 April 2017 9 April 2017
Please cite this article as: Baruch Vainshelboim, Jan Müller, Ricardo M. Lima, Kevin T. Nead, Cariad Chester, Khin Chan, Peter Kokkinos, Jonathan Myers , Cardiorespiratory fitness, physical activity and cancer mortality in men. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Ypmed(2017), doi: 10.1016/j.ypmed.2017.04.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Cardiorespiratory Fitness, Physical Activity and Cancer Mortality in Men
IP
T
Baruch Vainshelboim a, Jan Müller a,b, Ricardo M. Lima a,c , Kevin T. Nead d, Cariad Chester e , Khin Chana, Peter Kokkinos f, Jonathan Myers a
a
CR
Cardiology Division, Veterans Affairs Palo Alto Health Care System/ Stanford University, Palo Alto, California, US. b
US
Institute of Preventive Pediatrics, Technical University of Munich, Department of Pediatric Cardiology and Congenital Heart Disease, German Heart Center, Munich, Germany. Faculty of Physical Education, University of Brasília, Brasília, Brazil.
d
University of Pennsylvania, Philadelphia US.
AN
c
e
Washington DC Veterans Affairs Medical Center, Washington DC, US.
ED
f
M
Department of Medicine, Division of Oncology, Stanford University School of Medicine, Stanford, California, US.
PT
Short title: CRF, PA and Cancer Mortality
CE
Words count: Abstract-236, Text-2801.
AC
Corresponding Author: Baruch Vainshelboim, PhD VA Palo Alto Health Care System Cardiology 111C, 3801 Miranda Ave, Palo Alto CA 94304 Phone (650) 493-5000 x64661, Fax (650) 852-3473. Email: [email protected]
1
ACCEPTED MANUSCRIPT Abstract The preventive role of cardiorespiratory fitness (CRF) and physical activity (PA) in cancer mortality is not well-established. This study sought to evaluate the association between CRF, PA and cancer mortality in men.
IP
T
Maximal exercise testing was performed at the VA Palo Alto Health Care System in 5,876 male
CR
veterans (60.5±11 years) free from malignancy at baseline who were followed for mean of 9.9 (range 0.11 to 26.8) years. PA status was assessed in a sub-group of 4,034 participants. Relative
US
risks and population attributable risks (PAR%) for cancer-related mortality were determined.
AN
During the follow-up, 447 men (7.6%) died from cancer. Forty-nine percent of the sample was considered physically active (defined as meeting the minimal PA guidelines); this group
M
exhibited a 20% reduction in cancer mortality risk [95% confidence interval (0.67-0.97),
ED
p=0.02]. CRF was inversely associated with cancer death. For each 1 MET increase in CRF there was a 5% reduction in risk for cancer mortality (p=0.01). Compared to the least fit group (< 5.0
PT
METs), subjects with moderate to high CRF exhibited 26-46% reduced risks of cancer mortality
CE
(p for trend=0.002). The PARs% for low CRF and inactivity were 6.6% and 8.5%, respectively. Moderate and high CRF levels and meeting the minimal PA guidelines have protective benefits
AC
against cancer mortality in men. Eliminating inactivity and low CRF as risk factors could potentially prevent a considerable number of cancer deaths and reduce the associated societal and economic burden. Key words: Exercise Testing, Exercise Capacity, Cancer Death.
2
ACCEPTED MANUSCRIPT Introduction Cancer is the leading cause of death worldwide with more than 14 million new cancer cases and 8.2 million cancer deaths occurring in 2012 (Ferlay et al., 2015; World Health Organization, 2016). In the US approximately 600,000 people died from cancer and about 1.7 million new
T
cancers were diagnosed during 2015 (American Cancer Society, 2015). The American Cancer
IP
Society estimates lifetime probability of developing any type of cancer in men is 1 in 2, and 1 of
CR
4 will die from cancer (American Cancer Society, 2015). The financial cost of cancer was
US
estimated to be $895 billion globally and $88.7 billion in the US (American Cancer Society, 2015; LiveStrong and American Cancer Society, 2010).
AN
The World Cancer Research Fund and the American Cancer Research Institute have estimated
M
that up to two-thirds of cancer cases are related to lifestyle factors, and thus are potentially preventable (Wiseman, 2008; World Cancer Research Fund / American Institute for Cancer
ED
Research, 2007). Recently, meta-analyses of cardiorespiratory fitness (CRF) and physical
PT
activity (PA) patterns have reported inverse associations with cancer mortality, making CRF and PA potentially important factors for prevention (Li et al., 2015; Schmid and Leitzmann, 2015).
CE
Nevertheless, the public health implications of the existing CRF studies are somewhat
AC
challenging in terms of cancer death prevention. First, most previous reports have lacked a specific CRF threshold for cancer mortality risk reduction; thus, the public health recommendations need further exploration (Evenson et al., 2003; Farrell et al., 2007; Kampert et al., 1996; Sawada et al., 2003; Schmid and Leitzmann, 2015). Second, the available data clearly demonstrate protective benefits of high CRF levels (highest tertirle, quartile or quintile) however, the association between moderate CRF and cancer mortality has been inconsistent and needs further investigation to better understand the public health implications for cancer outcomes
3
ACCEPTED MANUSCRIPT (Evenson et al., 2003; Laukkanen et al., 2010; Schmid and Leitzmann, 2015). Finally, the attributable portion of inactivity, low CRF, or both, of cancer mortality, and the fraction of cancer deaths that might be prevented by eliminating these risk factors has not been previously described. Therefore, the aim of the current study was to further elucidate the association
CR
IP
T
between CRF, PA, and cancer mortality in men.
Methods
US
Study Population
The Veterans Exercise Testing Study (VETS) has been previously described (Myers et al., 2015;
AN
Myers et al., 2002). In brief, the VETS cohort is an ongoing, prospective evaluation of Veteran
M
subjects (aged 21-89 years) referred for exercise testing for clinical reasons, designed to address exercise test, clinical, and lifestyle factors and their association with health outcomes. All
ED
subjects who underwent a maximal treadmill exercise test between 1992 to 2012 at the Veterans
PT
Affairs Palo Alto Health Care System were considered for inclusion. Clinical information on diagnoses, risk factors and health behaviors (smoking status, history of alcohol and drug abuse)
CE
were collected using the Veterans Affairs Computerized Patient Record System (CPRS) and by
AC
participants’ self-repot at the time of the exercise test. Of 6,496 subjects who completed the baseline evaluation, 620 subjects were excluded; women (n=352), those having a history any malignancy (n=76) and incomplete or prematurely terminated exercise tests (n=192). A total of 5,876 male veterans were included in the analysis. In a sub-group of 4,034 participants, PA status was also assessed (Figure 1).
4
ACCEPTED MANUSCRIPT Cardiorespiratory Fitness Subjects underwent maximal sign or symptom-limited exercise testing using an individualized ramp treadmill protocol (Myers et al., 1991). Standard criteria for test termination were used (American College of Sports Medicine., 2014; Fletcher et al., 2013). The Borg 6-20 perceived
T
exertion scale was used to quantify degree of effort (Borg, 1982). CRF (in metabolic equivalents
IP
[METs]) was estimated from peak treadmill speed and grade. CRF was analyzed both as a
CR
continuous and categorical variable divided to quartiles (< 5 METs, 5 to <8 METs, 8 to <10 METs and ≥10 METs) (American College of Sports Medicine., 2014; Farrell et al., 2007; Myers
US
et al., 2004).
AN
Physical Activity
Participation in regular physical activity was evaluated by a subject’s binary response to the
M
question: “Do you engage in some form of physical activity such as brisk walking, jogging,
ED
bicycling, or swimming, long enough to work up a sweat, get your heart thumping, or become
PT
short of breath at least 3 times per week?” Subjects who answered yes to this question were classified as Active and considered as meeting the minimal guidelines for physical activity,
CE
whereas subjects who answered no were classified as Inactive and considered as not meeting the
AC
minimal guidelines for physical activity (American College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Myers et al., 2015). Cancer Mortality Ascertainment The Veterans Affairs computerized medical records were used for capturing cancer-related death; all-type cancer death was the primary outcome. Death records were carefully reviewed by qualified medical personal who were otherwise blinded to treadmill test results and other study information. The vital status of each patient was ascertained as of August 2015. 5
ACCEPTED MANUSCRIPT Statistical Analysis SPSS (IBM, Chicago, IL, USA) version 23 was used for statistical analyses. The significance level was set at p<0.05. Demographic, clinical and physiological data of the participants are presented as mean ± SD. Categorical variables are presented in percentages. When CRF was
T
expressed categorically, the Cox proportional hazards model was adjusted for age, smoking
IP
status (no smokers, previous smokers and current smokers) history of alcohol and drug abuse and
CR
body mass index; the continuous model was adjusted for age. PA was analyzed using the dichotomous categories Active and Inactive in a univariate model. Population attributable risk
US
(PAR) was calculated for CRF and PA according to the equation: P(RR-1)/1+P(RR-1)*100
AN
(Northridge, 1995), where P= prevalence of the risk factor (lowest CRF quartile [<5 METs] and inactivity) and RR=relative risk calculated from the Cox model (Greenland, 2004; Song and
M
Giovannucci, 2016). Clinical and demographic comparisons between subjects experiencing and
ED
not experiencing cancer death were performed using independent sample t-tests for continuous variables and chi-square tests for categorical variables. Chi-square tests were also used for
CE
Results
PT
categorical comparisons for the interaction between cancer death, inactivity and CRF quartiles.
AC
The study sample included 5,876 male subjects with a mean age of 60.5±11 years. Clinical, demographic and physiological characteristics of the sample are presented in Table 1. Fifty six percent were Caucasian, 22% African-American, 9% Hispanic and 8.2% Asian. Most subjects (89%) had at least one cardiometabolic risk factor, 12.8% had coronary artery disease, 7.8% had pulmonary disease and 14.7% had a diagnosis of diabetes at baseline. Twenty eight percent were current smokers and 34% had a history of smoking.
6
ACCEPTED MANUSCRIPT Mean CRF was 7.3±3.3 METs. During a mean follow-up of 9.9 years (range 0.11 to 26.8), 1,013 (17.2%) were diagnosed with cancer and 447 (7.6%) died from cancer (Table 1). In a multivariate categorical model, CRF was inversely associated with cancer mortality (p for trend =0.002). Subjects in the highest quartile (≥10 METs) had a 46% [95% confidence interval (CI)
T
0.36-0.82), p=0.004] reduced risk for cancer death compared to subjects in the lowest quartile
IP
(<5 METs). CRF in the second (5 to <8 METs) and third quartiles (8 to <10 METs) had 26%
CR
(95% CI 0.6-0.92, p=0.006) and 40% (95% CI 0.42-0.85, p=0.005) lower risks for cancer mortality, respectively (Figure 2). In a continuous, age-adjusted model, every 1 MET increase in
US
CRF was associated with a 5% reduction in cancer mortality (95% CI 0.91-0.98, p=0.01).
AN
PA status was assessed in subgroup of 4,034 subjects, and 49% were considered to be physically
M
active (Table 1). In a univariate analysis, these subjects demonstrated a 20% reduction in risk of cancer mortality (95% CI 0.67-0.97, p=0.02) compared to the inactive group (Figure 2).
ED
However, PA did not add significantly to risk in a multivariate model. Cancer mortality and
PT
inactivity were more prevalent in the lower quartiles of CRF, with significantly reduced numbers of cases in the higher quartiles (p<0.001) (Figure 3). Subjects not experiencing cancer death were
CE
younger, smoked less and had a lower prevalence of cardiometabolic risk factors and pulmonary
AC
disease. CRF was significantly higher among subjects who were free from cancer death compared to cancer decedents, and cancer decedents had a higher prevalence of inactivity (Table 2). The PARs of low CRF and inactivity were 6.6% and 8.5%, respectively (Table 3).
Discussion The current study sought to evaluate the association between CRF, PA, and cancer mortality in a relatively large cohort of men referred for exercise testing who were followed for a mean of approximately 10 years. The main findings indicate that CRF is inversely associated with death 7
ACCEPTED MANUSCRIPT from cancer, with an overall 5% reduction in risk for each MET achieved. Compared to low fit individuals (< 5.0 METs), subjects with moderate to high CRF had 26% to 46% reductions in cancer mortality (Figure 2). In addition, based on subjects’ self-report, being physically active was associated with a 20% decreased risk for cancer mortality (Figure 2). Cancer death and
T
inactivity were more prevalent in the lower quartiles of CRF while subjects free from cancer
IP
mortality were characterized by lower smoking prevalence, lower prevalence of comorbidities
CR
and higher CRF compared to cancer decedents (Table 2, Figure 3). Independently, relatively fit (≥5.0 METs) and physically active subjects exhibited 43% and 20% reductions in cancer
US
mortality risk, respectively. Subjects achieving <5.0 METs and who were inactive had 6.6% and
AN
8.5% PARs for cancer mortality, respectively (Table 3). These findings strengthen existing evidence supporting the importance of CRF and PA as modifiable risk factors for cancer
M
mortality. Specifically, they underscore potential cancer-protective benefits of being moderately
ED
fit and meeting the minimum recommendations for PA. From a public health perspective, both cancer-related mortality and its associated economic burden can potentially be lessened by
PT
reducing the prevalence of inactivity and low CRF. These data should encourage health care
CE
providers to assess CRF and PA, and advise men to achieve at least moderate CRF (≥5 to 10 METs) along with participating in at least the minimum amount of PA recommended by many
AC
health organizations in order to reduce the risk for cancer mortality (American College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Kushi et al., 2012; Pate et al., 1995; World Health Organization, 2010). The current results are consistent with previous reports generally showing an inverse association between CRF, PA and cancer mortality (Li et al., 2015; Schmid and Leitzmann, 2015). The findings also add novel insights to existing knowledge, particularly in terms of the impact of
8
ACCEPTED MANUSCRIPT moderate levels of CRF and their potential protective benefits against cancer mortality. In addition, to the best of our knowledge, the present study is the first to demonstrate population attributable risks of CRF and PA related to cancer mortality in men. The latter results are important with respect to resource allocation and public health decisions for primary cancer
T
prevention programs. Previous studies have reported some inconsistency regarding moderate
IP
CRF levels and cancer mortality (Evenson et al., 2003; Farrell et al., 2007; Kampert et al., 1996;
CR
Laukkanen et al., 2010; Sawada et al., 2003; Schmid and Leitzmann, 2015). While several studies have reported reduced cancer mortality risk in both high and moderate CRF levels
US
(Farrell et al., 2007; Kampert et al., 1996; Sawada et al., 2003), others have demonstrated this
AN
association only with relatively high CRF (Evenson et al., 2003; Laukkanen et al., 2010). Importantly, the current and previous data are consistent in that most low fit individuals are at
M
higher risk for cancer-related outcomes (Evenson et al., 2003; Farrell et al., 2007; Kampert et al.,
ED
1996; Sawada et al., 2003; Schmid and Leitzmann, 2015). However, most of these individuals have only a small chance of reaching high levels of CRF (highest quartile or quintile) even with
PT
supervised exercise training programs. It is well-documented that untrained subjects can
CE
typically improve CRF by 10-20% after a structured exercise training intervention using directly measured maximal oxygen uptake (VO2max) (Kenney et al., 2012). This degree of improvement
AC
is unlikely to move most individuals from the lowest to highest quartile or quintile of CRF; thus the cancer protective benefits are unlikely to be achieved. Our study sheds light on this issue by demonstrating 26% and 40% reductions in cancer mortality with moderate CRF levels (≥5 to <8 and 8 to<10 METS, respectively) (Figure 2). Achieving a modestly higher CRF level (moving from the first to the second quartile) was associated with a substantial risk reduction in cancer
9
ACCEPTED MANUSCRIPT mortality. These results underscore the concept that being moderately fit, a level achievable by most individuals, provides considerable benefit for prevention of cancer mortality. We also observed a 20% reduced risk of cancer mortality among subjects considered to be physically active, defined as meeting the minimal guidelines on PA (Figure 2) (American
IP
T
College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Kushi et al., 2012;
CR
Pate et al., 1995; World Health Organization, 2010). This finding is consistent with a recent meta-analysis (Li et al., 2015), showing a 13% reduced cancer mortality risk among those who
US
met the minimal PA recommendations. Taken together, the current and previous observations provide further support for the widely-held public-health message of meeting the 150 min/week
AN
moderate intensity PA (American College of Sports Medicine., 2014; Garber et al., 2011;
M
Haskell et al., 2007; Kushi et al., 2012; Pate et al., 1995; World Health Organization, 2010),
conditions, including cancer.
ED
which appears to have a considerable impact on reducing mortality from many chronic
PT
A variety of physiological mechanisms associated with CRF and PA may be involved in
CE
mediating cancer mortality (Brown et al., 2012; Li et al., 2015; Schmid and Leitzmann, 2015). Cancer is a broad and varied group of diseases, and therefore the biological pathways by which
AC
CRF and PA might influence cancer likely differ based on cancer type and site (Brown et al., 2012; Li et al., 2015). Potential protective mechanisms might include improved insulin sensitivity, reduced chronic inflammation, enhanced regulation of sex steroid hormones and growth-related hormones, decreased adipose tissue, optimized immune function, elevated antioxidant capacity, enhanced DNA repair, cell proliferation and apoptosis (Brown et al., 2012; Li et al., 2015; Schmid and Leitzmann, 2015). These mechanisms likely interact in a complex manner by blocking cancer cell initiation and countering cancer cell replication among fit and 10
ACCEPTED MANUSCRIPT regularly active individuals (Brown et al., 2012; Kiningham, 1998; Laukkanen et al., 2010; Sawada et al., 2003). However, despite growing observational evidence supporting the hypothesis that CRF and PA have a role in cancer prevention, prospective controlled studies addressing the mechanisms associated with exercise intervention on cancer genesis are lacking.
IP
T
The strengths of this study with respect to internal and external validity include its relatively
CR
large sample size (n=5,876), extended follow-up time [mean of approximately 10 years (range 0.11 to 26.8)] and a prospective assessment of cancer outcomes. In addition, cancer mortality
US
was verified through the Veterans Affairs Computerized Patient Record System which has been demonstrated to be comparatively accurate and complete (Page et al., 1996). CRF was measured
AN
objectively from treadmill speed and grade using an established technique (American College of
M
Sports Medicine., 2014; Fletcher et al., 2013). This method has been widely used in epidemiologic studies and has been shown to be strongly predictive for all-cause and
ED
cardiovascular mortality and other outcomes (Barry et al., 2014; Bouchard et al., 2015; Kodama
PT
et al., 2009; Williams, 2001). Our study also has several limitations. First, we used a brief selfreport for assessing PA status; while this has obvious limitations, self-report PA questionnaires
CE
are commonly used in epidemiologic studies, including those assessing cancer outcomes, and
AC
have been well-accepted as a tool for PA evaluation (Arem et al., 2014; Li et al., 2015; Li et al., 2016). Second, we considered active individuals as those who met the minimal recommendations for PA (American College of Sports Medicine., 2014; Haskell et al., 2007; Pate et al., 1995). While this approach cannot precisely quantify the minimal threshold of 150 min/week of moderate intensity PA, it has been a widely utilized by leading medical organizations (Haskell et al., 2007). Third, our multivariate hazards model, although adjusted for established covariates similar to those in previous studies of exercise and cancer, dietary habits were not collected in
11
ACCEPTED MANUSCRIPT the current cohort and were not included in the risk model; this is also consistent with previous studies (Evenson et al., 2003; Farrell et al., 2007; Sawada et al., 2003). Fourth, Veteran subjects are a unique population with a rich mixture of co-morbidities that may have influenced the results; future studies should be undertaken to confirm our findings. Finally, as in the case with
T
all epidemiological studies, our findings provide an association between CRF, PA, and cancer
CR
IP
mortality, but do not show cause and effect relationship.
In summary, achieving moderate and high CRF levels and being physically active appear to have
US
protective benefits against cancer-related death in men. These findings underscore the
AN
importance of CRF and PA as modifiable risk factors in primary prevention of cancer mortality. Funding
M
No funding was received for this study
ED
Conflict of Interest
AC
CE
PT
All authors declare there is no conflict of interest
12
ACCEPTED MANUSCRIPT References American Cancer Society, 2015. Cancer Facts & Figures 2015. Atlanta: American Cancer Society. American College of Sports Medicine., 2014. ACSM's guidelines for exercise testing and
IP
T
prescription, 9th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health, Philadelphia. Arem, H., Moore, S.C., Park, Y., Ballard-Barbash, R., Hollenbeck, A., Leitzmann, M.,
CR
Matthews, C.E., 2014. Physical activity and cancer-specific mortality in the NIH-AARP Diet and
US
Health Study cohort. Int J Cancer 135:423-31.
Barry, V.W., Baruth, M., Beets, M.W., Durstine, J.L., Liu, J., Blair, S.N., 2014. Fitness vs.
AN
fatness on all-cause mortality: a meta-analysis. Progress in cardiovascular diseases 56:382-90.
M
Borg, G.A., 1982. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14:377-81. Bouchard, C., Blair, S.N., Katzmarzyk, P.T., 2015. Less Sitting, More Physical Activity, or
ED
Higher Fitness? Mayo Clin Proc 90:1533-40.
PT
Brown, J.C., Winters-Stone, K., Lee, A., Schmitz, K.H., 2012. Cancer, physical activity, and exercise. Comprehensive Physiology 2:2775-809.
CE
Evenson, K.R., Stevens, J., Cai, J., Thomas, R., Thomas, O., 2003. The effect of
AC
cardiorespiratory fitness and obesity on cancer mortality in women and men. Med Sci Sports Exerc 35:270-7.
Farrell, S.W., Cortese, G.M., LaMonte, M.J., Blair, S.N., 2007. Cardiorespiratory fitness, different measures of adiposity, and cancer mortality in men. Obesity (Silver Spring) 15:3140-9. Ferlay, J., Soerjomataram, I., Dikshit, R., Eser, S., Mathers, C., Rebelo, M., Parkin, D.M., Forman, D., Bray, F., 2015. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 136:E359-86.
13
ACCEPTED MANUSCRIPT Fletcher, G.F., Ades, P.A., Kligfield, P., Arena, R., Balady, G.J., Bittner, V.A., Coke, L.A., Fleg, J.L., Forman, D.E., et al., 2013. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation 128:873-934. Garber, C.E., Blissmer, B., Deschenes, M.R., Franklin, B.A., Lamonte, M.J., Lee, I.M., Nieman,
T
D.C., Swain, D.P., American College of Sports, M., 2011. American College of Sports Medicine
IP
position stand. Quantity and quality of exercise for developing and maintaining
CR
cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc 43:1334-59.
US
Greenland, S., 2004. Interval estimation by simulation as an alternative to and extension of
AN
confidence intervals. International journal of epidemiology 33:1389-97. Haskell, W.L., Lee, I.M., Pate, R.R., Powell, K.E., Blair, S.N., Franklin, B.A., Macera, C.A.,
M
Heath, G.W., Thompson, P.D., et al., 2007. Physical activity and public health: updated
ED
recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation 116:1081-93.
PT
Kampert, J.B., Blair, S.N., Barlow, C.E., Kohl, H.W., 3rd, 1996. Physical activity, physical
CE
fitness, and all-cause and cancer mortality: a prospective study of men and women. Annals of epidemiology 6:452-7.
AC
Kenney, W.L., Wilmore, J.H., Costill, D.L., 2012. Physiology of sport and exercise, 5th ed. Human Kinetics, Champaign, IL. Kiningham, R.B., 1998. Physical activity and the primary prevention of cancer. Primary care 25:515-36. Kodama, S., Saito, K., Tanaka, S., Maki, M., Yachi, Y., Asumi, M., Sugawara, A., Totsuka, K., Shimano, H., et al., 2009. Cardiorespiratory fitness as a quantitative predictor of all-cause
14
ACCEPTED MANUSCRIPT mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA 301:2024-35. Kushi, L.H., Doyle, C., McCullough, M., Rock, C.L., Demark-Wahnefried, W., Bandera, E.V., Gapstur, S., Patel, A.V., Andrews, K., et al., 2012. American Cancer Society Guidelines on
IP
food choices and physical activity. CA Cancer J Clin 62:30-67.
T
nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy
CR
Laukkanen, J.A., Pukkala, E., Rauramaa, R., Makikallio, T.H., Toriola, A.T., Kurl, S., 2010. Cardiorespiratory fitness, lifestyle factors and cancer risk and mortality in Finnish men. Eur J
US
Cancer 46:355-63.
AN
Li, T., Wei, S., Shi, Y., Pang, S., Qin, Q., Yin, J., Deng, Y., Chen, Q., Wei, S., et al., 2015. The
cohort studies. Br J Sports Med.
M
dose-response effect of physical activity on cancer mortality: findings from 71 prospective
ED
Li, Y., Gu, M., Jing, F., Cai, S., Bao, C., Wang, J., Jin, M., Chen, K., 2016. Association between
Cancer 138:818-32.
PT
physical activity and all cancer mortality: Dose-response meta-analysis of cohort studies. Int J
CE
LiveStrong and American Cancer Society, 2010. The Global Economic Cost of Cancer. Myers, J., Buchanan, N., Walsh, D., Kraemer, M., McAuley, P., Hamilton-Wessler, M.,
AC
Froelicher, V.F., 1991. Comparison of the ramp versus standard exercise protocols. J Am Coll Cardiol 17:1334-42.
Myers, J., Kaykha, A., George, S., Abella, J., Zaheer, N., Lear, S., Yamazaki, T., Froelicher, V., 2004. Fitness versus physical activity patterns in predicting mortality in men. Am J Med 117:912-8.
15
ACCEPTED MANUSCRIPT Myers, J., Nead, K.T., Chang, P., Abella, J., Kokkinos, P., Leeper, N.J., 2015. Improved reclassification of mortality risk by assessment of physical activity in patients referred for exercise testing. Am J Med 128:396-402. Myers, J., Prakash, M., Froelicher, V., Do, D., Partington, S., Atwood, J.E., 2002. Exercise
T
capacity and mortality among men referred for exercise testing. N Engl J Med 346:793-801.
CR
public health action. Am J Public Health 85:1202-4.
IP
Northridge, M.E., 1995. Public health methods--attributable risk as a link between causality and
Page, W.F., Mahan, C.M., Kang, H.K., 1996. Vital status ascertainment through the files of the
US
Department of Veterans Affairs and the Social Security Administration. Annals of epidemiology
AN
6:102-9.
Pate, R.R., Pratt, M., Blair, S.N., Haskell, W.L., Macera, C.A., Bouchard, C., Buchner, D.,
M
Ettinger, W., Heath, G.W., et al., 1995. Physical activity and public health. A recommendation
Medicine. JAMA 273:402-7.
ED
from the Centers for Disease Control and Prevention and the American College of Sports
PT
Sawada, S.S., Muto, T., Tanaka, H., Lee, I.M., Paffenbarger, R.S., Jr., Shindo, M., Blair, S.N.,
CE
2003. Cardiorespiratory fitness and cancer mortality in Japanese men: a prospective study. Med Sci Sports Exerc 35:1546-50.
AC
Schmid, D., Leitzmann, M.F., 2015. Cardiorespiratory fitness as predictor of cancer mortality: a systematic review and meta-analysis. Ann Oncol 26:272-8. Song, M., Giovannucci, E., 2016. Preventable Incidence and Mortality of Carcinoma Associated With Lifestyle Factors Among White Adults in the United States. JAMA oncology. Williams, P.T., 2001. Physical fitness and activity as separate heart disease risk factors: a metaanalysis. Med Sci Sports Exerc 33:754-61.
16
ACCEPTED MANUSCRIPT Wiseman, M., 2008. The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. The Proceedings of the Nutrition Society 67:253-6. World Cancer Research Fund / American Institute for Cancer Research, 2007. Food, Nutrition,
T
Physical Activity, and the Prevention of Cancer: a Global Perspective. , Washington, DC.
IP
World Health Organization, 2010. Global Recommendations on Physical Activity for Health.,
CR
Geneva.
World Health Organization, 2016. Global Health Observatory, NCD mortality and morbidity
AC
CE
PT
ED
M
AN
US
2012.
17
ACCEPTED MANUSCRIPT Table 1 Demographic and Clinical Characteristics of the Population (n=5,876). Clinical History and Demographics
Value
Age (years)
60.5±12
Race 56%
African-American
22%
IP
T
Caucasian
Hispanic
9%
CR
Asian Other Body Mass Index (kg/m2)
8.8% 4.2%
US
27.7±5.5
Clinical History
AN
Any Cardiometabolic Risk Factors
89% 22%
Hypertension
54%
M
Family History of Coronary Artery Disease
Dyslipidemia Obesity (Body Mass Index≥30)
12.8% 7.7%
PT
Any Pulmonary Diseases
33%
ED
Coronary Artery Disease
35%
History of Alcohol Abuse
3% 1.5%
Diabetes
14.7%
CE
History of Drugs Abuse
38%
Current Smokers
28%
Previous Smokers
34%
AC
Non Smokers
Pack/Years
16.8±27.3
Medications Anti-Hypertensive Drugs
22.5%
Anti-Hyperlipidemias Drug
9.9%
Cardiorespiratory Fitness and Physical Activity
18
ACCEPTED MANUSCRIPT Exercise Capacity (METs)
7.3±3.3
Physical Activity Status (n=4,037) Physically Active
49%
Prostate
27%
Skin
26%
Colon and Rectum
T
Cancer Diagnosis (n=1,013)
IP
16%
Lung and Bronchus
8%
CR
Head & Neck, Oral Cavity and Pharynx Liver & Intrahepatic Bile Duct and Pancreas
US
Kidney & Renal Pelvis Other Cancers
AN
Cancer Death (n/% of total cohort)
AC
CE
PT
ED
M
Data presented as mean ± standard deviation and % for categorical variables.
19
5% 4% 3%
11% 447/7.6
ACCEPTED MANUSCRIPT Table 2 Comparison of Characteristics between Cancer Decedents and Subjects free from Cancer Death. Variables
Free from Cancer Death
Cancer Decedents
p-value
n=447 n=5429
67.6±10.1
<.001
60.6%
78.1%
<.001
31.4±39
.003
28±5.2
.058
96%
<.001
Current and Previous smokers Pack/year
16.1±26.5
Body Mass Index (kg/m2)
28.8±8.9 89%
US
Any Cardiometabolic Risk Factors
T
59.9±11
CR
Age(yr)
IP
Demographics
12.6%
15%
.142
Pulmonary Diseases
7.4%
12.4%
<.001
15.3%
18.1%
.120
7.4±3.3
5.8±2.4
<.001
49.9%
42.2%
.002
AN
Coronary Artery Disease
Diabetes
M
Exercise Capacity (METs)
ED
Physically Active
AC
CE
PT
Data presented as mean and SD or categorical variables as %.
20
ACCEPTED MANUSCRIPT Table 3 Prevalence, Relative Risks and Population Attributable Risks for Cancer Mortality
Cardiorespiratory fitness < 5 METs
Inactivity
Relative risk
PAR%
95% (CI)
95% (CI)
1.43
6.6%
(1.17-1.75)
(3.2- 9.3)
1.2
8.5%
(1.03-1.33)
(1.5-12.7)
21.8%
51%
T
Prevalence
IP
Risk Factor/Variables
p-value
<.001
.024
AC
CE
PT
ED
M
AN
US
CR
PAR; population attributable risk. CI; confidence interval. Relative risk of cardiorespiratory fitness was calculated using Cox proportional hazard model adjusted for age, smoking, drug and alcohol abuse, and body mass index. Relative risk of physical inactivity was calculated using bivariate Cox proportional hazard model.
21
ACCEPTED MANUSCRIPT Legends for Figures. Figure 1 Flowchart of Study Design Figure 2 Relative Risks of Cardiorespiratory Fitness, Physical Activity and Cancer Mortality in Men. MET; metabolic equivalent
IP
T
Figure 3 Prevalence of Cancer Mortality and Inactivity in different Cardiorespiratory Fitness Categories in Men.
AC
CE
PT
ED
M
AN
US
CR
P value was derived from comparison between cardiorespiratory fitness categories, activity status and cancer death using Chi square test for categorical variables.
22
PT
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
Figure 1
23
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 2
24
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 3
25
ACCEPTED MANUSCRIPT Highlights
T IP CR US AN M ED PT
CE
CRF and PA status are inversely associated with cancer mortality in men. Higher CRF and being physically active were associated with reduced risk for cancer death. Moderate and high CRF and being active have protective benefits against cancer mortality. Low CRF and inactivity have considerable attributable risks for cancer mortality.
AC
26
Baruch Vainshelboim, Jan Müller, Ricardo M. Lima, Kevin T. Nead, Cariad Chester, Khin Chan, Peter Kokkinos, Jonathan Myers PII: DOI: Reference:
S0091-7435(17)30134-2 doi: 10.1016/j.ypmed.2017.04.014 YPMED 4992
To appear in:
Preventive Medicine
Received date: Revised date: Accepted date:
11 January 2017 7 April 2017 9 April 2017
Please cite this article as: Baruch Vainshelboim, Jan Müller, Ricardo M. Lima, Kevin T. Nead, Cariad Chester, Khin Chan, Peter Kokkinos, Jonathan Myers , Cardiorespiratory fitness, physical activity and cancer mortality in men. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Ypmed(2017), doi: 10.1016/j.ypmed.2017.04.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Cardiorespiratory Fitness, Physical Activity and Cancer Mortality in Men
IP
T
Baruch Vainshelboim a, Jan Müller a,b, Ricardo M. Lima a,c , Kevin T. Nead d, Cariad Chester e , Khin Chana, Peter Kokkinos f, Jonathan Myers a
a
CR
Cardiology Division, Veterans Affairs Palo Alto Health Care System/ Stanford University, Palo Alto, California, US. b
US
Institute of Preventive Pediatrics, Technical University of Munich, Department of Pediatric Cardiology and Congenital Heart Disease, German Heart Center, Munich, Germany. Faculty of Physical Education, University of Brasília, Brasília, Brazil.
d
University of Pennsylvania, Philadelphia US.
AN
c
e
Washington DC Veterans Affairs Medical Center, Washington DC, US.
ED
f
M
Department of Medicine, Division of Oncology, Stanford University School of Medicine, Stanford, California, US.
PT
Short title: CRF, PA and Cancer Mortality
CE
Words count: Abstract-236, Text-2801.
AC
Corresponding Author: Baruch Vainshelboim, PhD VA Palo Alto Health Care System Cardiology 111C, 3801 Miranda Ave, Palo Alto CA 94304 Phone (650) 493-5000 x64661, Fax (650) 852-3473. Email: [email protected]
1
ACCEPTED MANUSCRIPT Abstract The preventive role of cardiorespiratory fitness (CRF) and physical activity (PA) in cancer mortality is not well-established. This study sought to evaluate the association between CRF, PA and cancer mortality in men.
IP
T
Maximal exercise testing was performed at the VA Palo Alto Health Care System in 5,876 male
CR
veterans (60.5±11 years) free from malignancy at baseline who were followed for mean of 9.9 (range 0.11 to 26.8) years. PA status was assessed in a sub-group of 4,034 participants. Relative
US
risks and population attributable risks (PAR%) for cancer-related mortality were determined.
AN
During the follow-up, 447 men (7.6%) died from cancer. Forty-nine percent of the sample was considered physically active (defined as meeting the minimal PA guidelines); this group
M
exhibited a 20% reduction in cancer mortality risk [95% confidence interval (0.67-0.97),
ED
p=0.02]. CRF was inversely associated with cancer death. For each 1 MET increase in CRF there was a 5% reduction in risk for cancer mortality (p=0.01). Compared to the least fit group (< 5.0
PT
METs), subjects with moderate to high CRF exhibited 26-46% reduced risks of cancer mortality
CE
(p for trend=0.002). The PARs% for low CRF and inactivity were 6.6% and 8.5%, respectively. Moderate and high CRF levels and meeting the minimal PA guidelines have protective benefits
AC
against cancer mortality in men. Eliminating inactivity and low CRF as risk factors could potentially prevent a considerable number of cancer deaths and reduce the associated societal and economic burden. Key words: Exercise Testing, Exercise Capacity, Cancer Death.
2
ACCEPTED MANUSCRIPT Introduction Cancer is the leading cause of death worldwide with more than 14 million new cancer cases and 8.2 million cancer deaths occurring in 2012 (Ferlay et al., 2015; World Health Organization, 2016). In the US approximately 600,000 people died from cancer and about 1.7 million new
T
cancers were diagnosed during 2015 (American Cancer Society, 2015). The American Cancer
IP
Society estimates lifetime probability of developing any type of cancer in men is 1 in 2, and 1 of
CR
4 will die from cancer (American Cancer Society, 2015). The financial cost of cancer was
US
estimated to be $895 billion globally and $88.7 billion in the US (American Cancer Society, 2015; LiveStrong and American Cancer Society, 2010).
AN
The World Cancer Research Fund and the American Cancer Research Institute have estimated
M
that up to two-thirds of cancer cases are related to lifestyle factors, and thus are potentially preventable (Wiseman, 2008; World Cancer Research Fund / American Institute for Cancer
ED
Research, 2007). Recently, meta-analyses of cardiorespiratory fitness (CRF) and physical
PT
activity (PA) patterns have reported inverse associations with cancer mortality, making CRF and PA potentially important factors for prevention (Li et al., 2015; Schmid and Leitzmann, 2015).
CE
Nevertheless, the public health implications of the existing CRF studies are somewhat
AC
challenging in terms of cancer death prevention. First, most previous reports have lacked a specific CRF threshold for cancer mortality risk reduction; thus, the public health recommendations need further exploration (Evenson et al., 2003; Farrell et al., 2007; Kampert et al., 1996; Sawada et al., 2003; Schmid and Leitzmann, 2015). Second, the available data clearly demonstrate protective benefits of high CRF levels (highest tertirle, quartile or quintile) however, the association between moderate CRF and cancer mortality has been inconsistent and needs further investigation to better understand the public health implications for cancer outcomes
3
ACCEPTED MANUSCRIPT (Evenson et al., 2003; Laukkanen et al., 2010; Schmid and Leitzmann, 2015). Finally, the attributable portion of inactivity, low CRF, or both, of cancer mortality, and the fraction of cancer deaths that might be prevented by eliminating these risk factors has not been previously described. Therefore, the aim of the current study was to further elucidate the association
CR
IP
T
between CRF, PA, and cancer mortality in men.
Methods
US
Study Population
The Veterans Exercise Testing Study (VETS) has been previously described (Myers et al., 2015;
AN
Myers et al., 2002). In brief, the VETS cohort is an ongoing, prospective evaluation of Veteran
M
subjects (aged 21-89 years) referred for exercise testing for clinical reasons, designed to address exercise test, clinical, and lifestyle factors and their association with health outcomes. All
ED
subjects who underwent a maximal treadmill exercise test between 1992 to 2012 at the Veterans
PT
Affairs Palo Alto Health Care System were considered for inclusion. Clinical information on diagnoses, risk factors and health behaviors (smoking status, history of alcohol and drug abuse)
CE
were collected using the Veterans Affairs Computerized Patient Record System (CPRS) and by
AC
participants’ self-repot at the time of the exercise test. Of 6,496 subjects who completed the baseline evaluation, 620 subjects were excluded; women (n=352), those having a history any malignancy (n=76) and incomplete or prematurely terminated exercise tests (n=192). A total of 5,876 male veterans were included in the analysis. In a sub-group of 4,034 participants, PA status was also assessed (Figure 1).
4
ACCEPTED MANUSCRIPT Cardiorespiratory Fitness Subjects underwent maximal sign or symptom-limited exercise testing using an individualized ramp treadmill protocol (Myers et al., 1991). Standard criteria for test termination were used (American College of Sports Medicine., 2014; Fletcher et al., 2013). The Borg 6-20 perceived
T
exertion scale was used to quantify degree of effort (Borg, 1982). CRF (in metabolic equivalents
IP
[METs]) was estimated from peak treadmill speed and grade. CRF was analyzed both as a
CR
continuous and categorical variable divided to quartiles (< 5 METs, 5 to <8 METs, 8 to <10 METs and ≥10 METs) (American College of Sports Medicine., 2014; Farrell et al., 2007; Myers
US
et al., 2004).
AN
Physical Activity
Participation in regular physical activity was evaluated by a subject’s binary response to the
M
question: “Do you engage in some form of physical activity such as brisk walking, jogging,
ED
bicycling, or swimming, long enough to work up a sweat, get your heart thumping, or become
PT
short of breath at least 3 times per week?” Subjects who answered yes to this question were classified as Active and considered as meeting the minimal guidelines for physical activity,
CE
whereas subjects who answered no were classified as Inactive and considered as not meeting the
AC
minimal guidelines for physical activity (American College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Myers et al., 2015). Cancer Mortality Ascertainment The Veterans Affairs computerized medical records were used for capturing cancer-related death; all-type cancer death was the primary outcome. Death records were carefully reviewed by qualified medical personal who were otherwise blinded to treadmill test results and other study information. The vital status of each patient was ascertained as of August 2015. 5
ACCEPTED MANUSCRIPT Statistical Analysis SPSS (IBM, Chicago, IL, USA) version 23 was used for statistical analyses. The significance level was set at p<0.05. Demographic, clinical and physiological data of the participants are presented as mean ± SD. Categorical variables are presented in percentages. When CRF was
T
expressed categorically, the Cox proportional hazards model was adjusted for age, smoking
IP
status (no smokers, previous smokers and current smokers) history of alcohol and drug abuse and
CR
body mass index; the continuous model was adjusted for age. PA was analyzed using the dichotomous categories Active and Inactive in a univariate model. Population attributable risk
US
(PAR) was calculated for CRF and PA according to the equation: P(RR-1)/1+P(RR-1)*100
AN
(Northridge, 1995), where P= prevalence of the risk factor (lowest CRF quartile [<5 METs] and inactivity) and RR=relative risk calculated from the Cox model (Greenland, 2004; Song and
M
Giovannucci, 2016). Clinical and demographic comparisons between subjects experiencing and
ED
not experiencing cancer death were performed using independent sample t-tests for continuous variables and chi-square tests for categorical variables. Chi-square tests were also used for
CE
Results
PT
categorical comparisons for the interaction between cancer death, inactivity and CRF quartiles.
AC
The study sample included 5,876 male subjects with a mean age of 60.5±11 years. Clinical, demographic and physiological characteristics of the sample are presented in Table 1. Fifty six percent were Caucasian, 22% African-American, 9% Hispanic and 8.2% Asian. Most subjects (89%) had at least one cardiometabolic risk factor, 12.8% had coronary artery disease, 7.8% had pulmonary disease and 14.7% had a diagnosis of diabetes at baseline. Twenty eight percent were current smokers and 34% had a history of smoking.
6
ACCEPTED MANUSCRIPT Mean CRF was 7.3±3.3 METs. During a mean follow-up of 9.9 years (range 0.11 to 26.8), 1,013 (17.2%) were diagnosed with cancer and 447 (7.6%) died from cancer (Table 1). In a multivariate categorical model, CRF was inversely associated with cancer mortality (p for trend =0.002). Subjects in the highest quartile (≥10 METs) had a 46% [95% confidence interval (CI)
T
0.36-0.82), p=0.004] reduced risk for cancer death compared to subjects in the lowest quartile
IP
(<5 METs). CRF in the second (5 to <8 METs) and third quartiles (8 to <10 METs) had 26%
CR
(95% CI 0.6-0.92, p=0.006) and 40% (95% CI 0.42-0.85, p=0.005) lower risks for cancer mortality, respectively (Figure 2). In a continuous, age-adjusted model, every 1 MET increase in
US
CRF was associated with a 5% reduction in cancer mortality (95% CI 0.91-0.98, p=0.01).
AN
PA status was assessed in subgroup of 4,034 subjects, and 49% were considered to be physically
M
active (Table 1). In a univariate analysis, these subjects demonstrated a 20% reduction in risk of cancer mortality (95% CI 0.67-0.97, p=0.02) compared to the inactive group (Figure 2).
ED
However, PA did not add significantly to risk in a multivariate model. Cancer mortality and
PT
inactivity were more prevalent in the lower quartiles of CRF, with significantly reduced numbers of cases in the higher quartiles (p<0.001) (Figure 3). Subjects not experiencing cancer death were
CE
younger, smoked less and had a lower prevalence of cardiometabolic risk factors and pulmonary
AC
disease. CRF was significantly higher among subjects who were free from cancer death compared to cancer decedents, and cancer decedents had a higher prevalence of inactivity (Table 2). The PARs of low CRF and inactivity were 6.6% and 8.5%, respectively (Table 3).
Discussion The current study sought to evaluate the association between CRF, PA, and cancer mortality in a relatively large cohort of men referred for exercise testing who were followed for a mean of approximately 10 years. The main findings indicate that CRF is inversely associated with death 7
ACCEPTED MANUSCRIPT from cancer, with an overall 5% reduction in risk for each MET achieved. Compared to low fit individuals (< 5.0 METs), subjects with moderate to high CRF had 26% to 46% reductions in cancer mortality (Figure 2). In addition, based on subjects’ self-report, being physically active was associated with a 20% decreased risk for cancer mortality (Figure 2). Cancer death and
T
inactivity were more prevalent in the lower quartiles of CRF while subjects free from cancer
IP
mortality were characterized by lower smoking prevalence, lower prevalence of comorbidities
CR
and higher CRF compared to cancer decedents (Table 2, Figure 3). Independently, relatively fit (≥5.0 METs) and physically active subjects exhibited 43% and 20% reductions in cancer
US
mortality risk, respectively. Subjects achieving <5.0 METs and who were inactive had 6.6% and
AN
8.5% PARs for cancer mortality, respectively (Table 3). These findings strengthen existing evidence supporting the importance of CRF and PA as modifiable risk factors for cancer
M
mortality. Specifically, they underscore potential cancer-protective benefits of being moderately
ED
fit and meeting the minimum recommendations for PA. From a public health perspective, both cancer-related mortality and its associated economic burden can potentially be lessened by
PT
reducing the prevalence of inactivity and low CRF. These data should encourage health care
CE
providers to assess CRF and PA, and advise men to achieve at least moderate CRF (≥5 to 10 METs) along with participating in at least the minimum amount of PA recommended by many
AC
health organizations in order to reduce the risk for cancer mortality (American College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Kushi et al., 2012; Pate et al., 1995; World Health Organization, 2010). The current results are consistent with previous reports generally showing an inverse association between CRF, PA and cancer mortality (Li et al., 2015; Schmid and Leitzmann, 2015). The findings also add novel insights to existing knowledge, particularly in terms of the impact of
8
ACCEPTED MANUSCRIPT moderate levels of CRF and their potential protective benefits against cancer mortality. In addition, to the best of our knowledge, the present study is the first to demonstrate population attributable risks of CRF and PA related to cancer mortality in men. The latter results are important with respect to resource allocation and public health decisions for primary cancer
T
prevention programs. Previous studies have reported some inconsistency regarding moderate
IP
CRF levels and cancer mortality (Evenson et al., 2003; Farrell et al., 2007; Kampert et al., 1996;
CR
Laukkanen et al., 2010; Sawada et al., 2003; Schmid and Leitzmann, 2015). While several studies have reported reduced cancer mortality risk in both high and moderate CRF levels
US
(Farrell et al., 2007; Kampert et al., 1996; Sawada et al., 2003), others have demonstrated this
AN
association only with relatively high CRF (Evenson et al., 2003; Laukkanen et al., 2010). Importantly, the current and previous data are consistent in that most low fit individuals are at
M
higher risk for cancer-related outcomes (Evenson et al., 2003; Farrell et al., 2007; Kampert et al.,
ED
1996; Sawada et al., 2003; Schmid and Leitzmann, 2015). However, most of these individuals have only a small chance of reaching high levels of CRF (highest quartile or quintile) even with
PT
supervised exercise training programs. It is well-documented that untrained subjects can
CE
typically improve CRF by 10-20% after a structured exercise training intervention using directly measured maximal oxygen uptake (VO2max) (Kenney et al., 2012). This degree of improvement
AC
is unlikely to move most individuals from the lowest to highest quartile or quintile of CRF; thus the cancer protective benefits are unlikely to be achieved. Our study sheds light on this issue by demonstrating 26% and 40% reductions in cancer mortality with moderate CRF levels (≥5 to <8 and 8 to<10 METS, respectively) (Figure 2). Achieving a modestly higher CRF level (moving from the first to the second quartile) was associated with a substantial risk reduction in cancer
9
ACCEPTED MANUSCRIPT mortality. These results underscore the concept that being moderately fit, a level achievable by most individuals, provides considerable benefit for prevention of cancer mortality. We also observed a 20% reduced risk of cancer mortality among subjects considered to be physically active, defined as meeting the minimal guidelines on PA (Figure 2) (American
IP
T
College of Sports Medicine., 2014; Garber et al., 2011; Haskell et al., 2007; Kushi et al., 2012;
CR
Pate et al., 1995; World Health Organization, 2010). This finding is consistent with a recent meta-analysis (Li et al., 2015), showing a 13% reduced cancer mortality risk among those who
US
met the minimal PA recommendations. Taken together, the current and previous observations provide further support for the widely-held public-health message of meeting the 150 min/week
AN
moderate intensity PA (American College of Sports Medicine., 2014; Garber et al., 2011;
M
Haskell et al., 2007; Kushi et al., 2012; Pate et al., 1995; World Health Organization, 2010),
conditions, including cancer.
ED
which appears to have a considerable impact on reducing mortality from many chronic
PT
A variety of physiological mechanisms associated with CRF and PA may be involved in
CE
mediating cancer mortality (Brown et al., 2012; Li et al., 2015; Schmid and Leitzmann, 2015). Cancer is a broad and varied group of diseases, and therefore the biological pathways by which
AC
CRF and PA might influence cancer likely differ based on cancer type and site (Brown et al., 2012; Li et al., 2015). Potential protective mechanisms might include improved insulin sensitivity, reduced chronic inflammation, enhanced regulation of sex steroid hormones and growth-related hormones, decreased adipose tissue, optimized immune function, elevated antioxidant capacity, enhanced DNA repair, cell proliferation and apoptosis (Brown et al., 2012; Li et al., 2015; Schmid and Leitzmann, 2015). These mechanisms likely interact in a complex manner by blocking cancer cell initiation and countering cancer cell replication among fit and 10
ACCEPTED MANUSCRIPT regularly active individuals (Brown et al., 2012; Kiningham, 1998; Laukkanen et al., 2010; Sawada et al., 2003). However, despite growing observational evidence supporting the hypothesis that CRF and PA have a role in cancer prevention, prospective controlled studies addressing the mechanisms associated with exercise intervention on cancer genesis are lacking.
IP
T
The strengths of this study with respect to internal and external validity include its relatively
CR
large sample size (n=5,876), extended follow-up time [mean of approximately 10 years (range 0.11 to 26.8)] and a prospective assessment of cancer outcomes. In addition, cancer mortality
US
was verified through the Veterans Affairs Computerized Patient Record System which has been demonstrated to be comparatively accurate and complete (Page et al., 1996). CRF was measured
AN
objectively from treadmill speed and grade using an established technique (American College of
M
Sports Medicine., 2014; Fletcher et al., 2013). This method has been widely used in epidemiologic studies and has been shown to be strongly predictive for all-cause and
ED
cardiovascular mortality and other outcomes (Barry et al., 2014; Bouchard et al., 2015; Kodama
PT
et al., 2009; Williams, 2001). Our study also has several limitations. First, we used a brief selfreport for assessing PA status; while this has obvious limitations, self-report PA questionnaires
CE
are commonly used in epidemiologic studies, including those assessing cancer outcomes, and
AC
have been well-accepted as a tool for PA evaluation (Arem et al., 2014; Li et al., 2015; Li et al., 2016). Second, we considered active individuals as those who met the minimal recommendations for PA (American College of Sports Medicine., 2014; Haskell et al., 2007; Pate et al., 1995). While this approach cannot precisely quantify the minimal threshold of 150 min/week of moderate intensity PA, it has been a widely utilized by leading medical organizations (Haskell et al., 2007). Third, our multivariate hazards model, although adjusted for established covariates similar to those in previous studies of exercise and cancer, dietary habits were not collected in
11
ACCEPTED MANUSCRIPT the current cohort and were not included in the risk model; this is also consistent with previous studies (Evenson et al., 2003; Farrell et al., 2007; Sawada et al., 2003). Fourth, Veteran subjects are a unique population with a rich mixture of co-morbidities that may have influenced the results; future studies should be undertaken to confirm our findings. Finally, as in the case with
T
all epidemiological studies, our findings provide an association between CRF, PA, and cancer
CR
IP
mortality, but do not show cause and effect relationship.
In summary, achieving moderate and high CRF levels and being physically active appear to have
US
protective benefits against cancer-related death in men. These findings underscore the
AN
importance of CRF and PA as modifiable risk factors in primary prevention of cancer mortality. Funding
M
No funding was received for this study
ED
Conflict of Interest
AC
CE
PT
All authors declare there is no conflict of interest
12
ACCEPTED MANUSCRIPT References American Cancer Society, 2015. Cancer Facts & Figures 2015. Atlanta: American Cancer Society. American College of Sports Medicine., 2014. ACSM's guidelines for exercise testing and
IP
T
prescription, 9th ed. Wolters Kluwer/Lippincott Williams & Wilkins Health, Philadelphia. Arem, H., Moore, S.C., Park, Y., Ballard-Barbash, R., Hollenbeck, A., Leitzmann, M.,
CR
Matthews, C.E., 2014. Physical activity and cancer-specific mortality in the NIH-AARP Diet and
US
Health Study cohort. Int J Cancer 135:423-31.
Barry, V.W., Baruth, M., Beets, M.W., Durstine, J.L., Liu, J., Blair, S.N., 2014. Fitness vs.
AN
fatness on all-cause mortality: a meta-analysis. Progress in cardiovascular diseases 56:382-90.
M
Borg, G.A., 1982. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14:377-81. Bouchard, C., Blair, S.N., Katzmarzyk, P.T., 2015. Less Sitting, More Physical Activity, or
ED
Higher Fitness? Mayo Clin Proc 90:1533-40.
PT
Brown, J.C., Winters-Stone, K., Lee, A., Schmitz, K.H., 2012. Cancer, physical activity, and exercise. Comprehensive Physiology 2:2775-809.
CE
Evenson, K.R., Stevens, J., Cai, J., Thomas, R., Thomas, O., 2003. The effect of
AC
cardiorespiratory fitness and obesity on cancer mortality in women and men. Med Sci Sports Exerc 35:270-7.
Farrell, S.W., Cortese, G.M., LaMonte, M.J., Blair, S.N., 2007. Cardiorespiratory fitness, different measures of adiposity, and cancer mortality in men. Obesity (Silver Spring) 15:3140-9. Ferlay, J., Soerjomataram, I., Dikshit, R., Eser, S., Mathers, C., Rebelo, M., Parkin, D.M., Forman, D., Bray, F., 2015. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 136:E359-86.
13
ACCEPTED MANUSCRIPT Fletcher, G.F., Ades, P.A., Kligfield, P., Arena, R., Balady, G.J., Bittner, V.A., Coke, L.A., Fleg, J.L., Forman, D.E., et al., 2013. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation 128:873-934. Garber, C.E., Blissmer, B., Deschenes, M.R., Franklin, B.A., Lamonte, M.J., Lee, I.M., Nieman,
T
D.C., Swain, D.P., American College of Sports, M., 2011. American College of Sports Medicine
IP
position stand. Quantity and quality of exercise for developing and maintaining
CR
cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc 43:1334-59.
US
Greenland, S., 2004. Interval estimation by simulation as an alternative to and extension of
AN
confidence intervals. International journal of epidemiology 33:1389-97. Haskell, W.L., Lee, I.M., Pate, R.R., Powell, K.E., Blair, S.N., Franklin, B.A., Macera, C.A.,
M
Heath, G.W., Thompson, P.D., et al., 2007. Physical activity and public health: updated
ED
recommendation for adults from the American College of Sports Medicine and the American Heart Association. Circulation 116:1081-93.
PT
Kampert, J.B., Blair, S.N., Barlow, C.E., Kohl, H.W., 3rd, 1996. Physical activity, physical
CE
fitness, and all-cause and cancer mortality: a prospective study of men and women. Annals of epidemiology 6:452-7.
AC
Kenney, W.L., Wilmore, J.H., Costill, D.L., 2012. Physiology of sport and exercise, 5th ed. Human Kinetics, Champaign, IL. Kiningham, R.B., 1998. Physical activity and the primary prevention of cancer. Primary care 25:515-36. Kodama, S., Saito, K., Tanaka, S., Maki, M., Yachi, Y., Asumi, M., Sugawara, A., Totsuka, K., Shimano, H., et al., 2009. Cardiorespiratory fitness as a quantitative predictor of all-cause
14
ACCEPTED MANUSCRIPT mortality and cardiovascular events in healthy men and women: a meta-analysis. JAMA 301:2024-35. Kushi, L.H., Doyle, C., McCullough, M., Rock, C.L., Demark-Wahnefried, W., Bandera, E.V., Gapstur, S., Patel, A.V., Andrews, K., et al., 2012. American Cancer Society Guidelines on
IP
food choices and physical activity. CA Cancer J Clin 62:30-67.
T
nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy
CR
Laukkanen, J.A., Pukkala, E., Rauramaa, R., Makikallio, T.H., Toriola, A.T., Kurl, S., 2010. Cardiorespiratory fitness, lifestyle factors and cancer risk and mortality in Finnish men. Eur J
US
Cancer 46:355-63.
AN
Li, T., Wei, S., Shi, Y., Pang, S., Qin, Q., Yin, J., Deng, Y., Chen, Q., Wei, S., et al., 2015. The
cohort studies. Br J Sports Med.
M
dose-response effect of physical activity on cancer mortality: findings from 71 prospective
ED
Li, Y., Gu, M., Jing, F., Cai, S., Bao, C., Wang, J., Jin, M., Chen, K., 2016. Association between
Cancer 138:818-32.
PT
physical activity and all cancer mortality: Dose-response meta-analysis of cohort studies. Int J
CE
LiveStrong and American Cancer Society, 2010. The Global Economic Cost of Cancer. Myers, J., Buchanan, N., Walsh, D., Kraemer, M., McAuley, P., Hamilton-Wessler, M.,
AC
Froelicher, V.F., 1991. Comparison of the ramp versus standard exercise protocols. J Am Coll Cardiol 17:1334-42.
Myers, J., Kaykha, A., George, S., Abella, J., Zaheer, N., Lear, S., Yamazaki, T., Froelicher, V., 2004. Fitness versus physical activity patterns in predicting mortality in men. Am J Med 117:912-8.
15
ACCEPTED MANUSCRIPT Myers, J., Nead, K.T., Chang, P., Abella, J., Kokkinos, P., Leeper, N.J., 2015. Improved reclassification of mortality risk by assessment of physical activity in patients referred for exercise testing. Am J Med 128:396-402. Myers, J., Prakash, M., Froelicher, V., Do, D., Partington, S., Atwood, J.E., 2002. Exercise
T
capacity and mortality among men referred for exercise testing. N Engl J Med 346:793-801.
CR
public health action. Am J Public Health 85:1202-4.
IP
Northridge, M.E., 1995. Public health methods--attributable risk as a link between causality and
Page, W.F., Mahan, C.M., Kang, H.K., 1996. Vital status ascertainment through the files of the
US
Department of Veterans Affairs and the Social Security Administration. Annals of epidemiology
AN
6:102-9.
Pate, R.R., Pratt, M., Blair, S.N., Haskell, W.L., Macera, C.A., Bouchard, C., Buchner, D.,
M
Ettinger, W., Heath, G.W., et al., 1995. Physical activity and public health. A recommendation
Medicine. JAMA 273:402-7.
ED
from the Centers for Disease Control and Prevention and the American College of Sports
PT
Sawada, S.S., Muto, T., Tanaka, H., Lee, I.M., Paffenbarger, R.S., Jr., Shindo, M., Blair, S.N.,
CE
2003. Cardiorespiratory fitness and cancer mortality in Japanese men: a prospective study. Med Sci Sports Exerc 35:1546-50.
AC
Schmid, D., Leitzmann, M.F., 2015. Cardiorespiratory fitness as predictor of cancer mortality: a systematic review and meta-analysis. Ann Oncol 26:272-8. Song, M., Giovannucci, E., 2016. Preventable Incidence and Mortality of Carcinoma Associated With Lifestyle Factors Among White Adults in the United States. JAMA oncology. Williams, P.T., 2001. Physical fitness and activity as separate heart disease risk factors: a metaanalysis. Med Sci Sports Exerc 33:754-61.
16
ACCEPTED MANUSCRIPT Wiseman, M., 2008. The second World Cancer Research Fund/American Institute for Cancer Research expert report. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. The Proceedings of the Nutrition Society 67:253-6. World Cancer Research Fund / American Institute for Cancer Research, 2007. Food, Nutrition,
T
Physical Activity, and the Prevention of Cancer: a Global Perspective. , Washington, DC.
IP
World Health Organization, 2010. Global Recommendations on Physical Activity for Health.,
CR
Geneva.
World Health Organization, 2016. Global Health Observatory, NCD mortality and morbidity
AC
CE
PT
ED
M
AN
US
2012.
17
ACCEPTED MANUSCRIPT Table 1 Demographic and Clinical Characteristics of the Population (n=5,876). Clinical History and Demographics
Value
Age (years)
60.5±12
Race 56%
African-American
22%
IP
T
Caucasian
Hispanic
9%
CR
Asian Other Body Mass Index (kg/m2)
8.8% 4.2%
US
27.7±5.5
Clinical History
AN
Any Cardiometabolic Risk Factors
89% 22%
Hypertension
54%
M
Family History of Coronary Artery Disease
Dyslipidemia Obesity (Body Mass Index≥30)
12.8% 7.7%
PT
Any Pulmonary Diseases
33%
ED
Coronary Artery Disease
35%
History of Alcohol Abuse
3% 1.5%
Diabetes
14.7%
CE
History of Drugs Abuse
38%
Current Smokers
28%
Previous Smokers
34%
AC
Non Smokers
Pack/Years
16.8±27.3
Medications Anti-Hypertensive Drugs
22.5%
Anti-Hyperlipidemias Drug
9.9%
Cardiorespiratory Fitness and Physical Activity
18
ACCEPTED MANUSCRIPT Exercise Capacity (METs)
7.3±3.3
Physical Activity Status (n=4,037) Physically Active
49%
Prostate
27%
Skin
26%
Colon and Rectum
T
Cancer Diagnosis (n=1,013)
IP
16%
Lung and Bronchus
8%
CR
Head & Neck, Oral Cavity and Pharynx Liver & Intrahepatic Bile Duct and Pancreas
US
Kidney & Renal Pelvis Other Cancers
AN
Cancer Death (n/% of total cohort)
AC
CE
PT
ED
M
Data presented as mean ± standard deviation and % for categorical variables.
19
5% 4% 3%
11% 447/7.6
ACCEPTED MANUSCRIPT Table 2 Comparison of Characteristics between Cancer Decedents and Subjects free from Cancer Death. Variables
Free from Cancer Death
Cancer Decedents
p-value
n=447 n=5429
67.6±10.1
<.001
60.6%
78.1%
<.001
31.4±39
.003
28±5.2
.058
96%
<.001
Current and Previous smokers Pack/year
16.1±26.5
Body Mass Index (kg/m2)
28.8±8.9 89%
US
Any Cardiometabolic Risk Factors
T
59.9±11
CR
Age(yr)
IP
Demographics
12.6%
15%
.142
Pulmonary Diseases
7.4%
12.4%
<.001
15.3%
18.1%
.120
7.4±3.3
5.8±2.4
<.001
49.9%
42.2%
.002
AN
Coronary Artery Disease
Diabetes
M
Exercise Capacity (METs)
ED
Physically Active
AC
CE
PT
Data presented as mean and SD or categorical variables as %.
20
ACCEPTED MANUSCRIPT Table 3 Prevalence, Relative Risks and Population Attributable Risks for Cancer Mortality
Cardiorespiratory fitness < 5 METs
Inactivity
Relative risk
PAR%
95% (CI)
95% (CI)
1.43
6.6%
(1.17-1.75)
(3.2- 9.3)
1.2
8.5%
(1.03-1.33)
(1.5-12.7)
21.8%
51%
T
Prevalence
IP
Risk Factor/Variables
p-value
<.001
.024
AC
CE
PT
ED
M
AN
US
CR
PAR; population attributable risk. CI; confidence interval. Relative risk of cardiorespiratory fitness was calculated using Cox proportional hazard model adjusted for age, smoking, drug and alcohol abuse, and body mass index. Relative risk of physical inactivity was calculated using bivariate Cox proportional hazard model.
21
ACCEPTED MANUSCRIPT Legends for Figures. Figure 1 Flowchart of Study Design Figure 2 Relative Risks of Cardiorespiratory Fitness, Physical Activity and Cancer Mortality in Men. MET; metabolic equivalent
IP
T
Figure 3 Prevalence of Cancer Mortality and Inactivity in different Cardiorespiratory Fitness Categories in Men.
AC
CE
PT
ED
M
AN
US
CR
P value was derived from comparison between cardiorespiratory fitness categories, activity status and cancer death using Chi square test for categorical variables.
22
PT
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
Figure 1
23
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 2
24
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 3
25
ACCEPTED MANUSCRIPT Highlights
T IP CR US AN M ED PT
CE
CRF and PA status are inversely associated with cancer mortality in men. Higher CRF and being physically active were associated with reduced risk for cancer death. Moderate and high CRF and being active have protective benefits against cancer mortality. Low CRF and inactivity have considerable attributable risks for cancer mortality.
AC
26