Potential for prostate cancer prevention through physical activity
- First Online:
- Cite this article as:
- Young-McCaughan, S. World J Urol (2012) 30: 167. doi:10.1007/s00345-011-0812-y
- 538 Downloads
Review the most recent literature investigating exercise in the prevention of prostate cancer to determine whether or not exercise might play a role preventing the disease.
A review of the published literature available on the Medline, PubMed, and Ovid literature databases and search of relevant publications through September 2011 was conducted to identify studies that investigated the role of physical activity in the prevention of prostate cancer.
Summarizing the findings from 40 epidemiological studies, 22 suggested that physical activity reduced the risk of prostate cancer, 14 found no association between physical activity and risk of prostate cancer, and four showed an increased risk of prostate cancer. In the 22 studies reporting a protective effect, the effect sizes were predominantly small or moderate. In the four studies reporting that physical activity increased a man’s risk for prostate cancer, the effect size was also predominantly small.
There is a growing body of epidemiological research suggesting that physical activity is protective against the development of prostate cancer. This paper reviewed 22 studies published in the past 12 years updating an earlier review finding that although some studies show no benefit and a very few show increased risk for prostate cancer, the majority of studies document a significant, albeit small, protective effect of physical activity.
KeywordsPhysical activity Exercise Prostate cancer Prevention
A key component to a healthy lifestyle is exercise. The evidence is overwhelming documenting the myriad health benefits of physical activity in both preventing disease and promoting quality of life . While exercise was not initially recognized to prevent cancer, research conducted over the past 30 years has demonstrated that physical activity is a key component of reducing the risk of many cancers [2, 3]. The evidence is strongest for the reduced risk of developing breast  and colorectal [5, 6, 7] cancers, but physical activity may also be associated with the reduced risk of cancer of the pancreas  and endometrium . Researchers have also investigated the association between physical activity and the risk of lung, ovarian, and kidney cancers .
It is not as clear whether or not exercise prevents prostate cancer. Data from large epidemiological studies assessing physical activity and risk of prostate cancer are conflicting [10, 11]. In a comprehensive review published in 2001 of 24 studies investigating physical activity levels and risk of prostate cancer, 14 suggested that physical activity reduced the risk of prostate cancer, six found no association, and the remaining four showed an increased risk of prostate cancer in the more physically active men . The purpose of this paper is to update this earlier review with the literature published in the last decade investigating exercise in the prevention of prostate cancer to ascertain the role exercise might play in preventing the disease.
Summary of epidemiological studies investigating the role of physical activity in the prevention of prostate cancer
Study design and population
Sample size, number of controls and cases (% of sample)
Assessment of physical activity
Hayes et al. USA, 
Case control of men identified through cancer registries in Georgia, Detroit & New Jersey; and controls using random digit dialing with a population-based control group recruited from the same geographic area
Sample = 2,133
Controls = 1,201
Cases = 932 (44%); 449 AA & 483 C
Review of occupation
No effect. No association between occupational physical activity and risk of PC
Clarke and Whittemore, USA, 
Prospective cohort of men participating in the NHANES I established between 1971 & 1975 and followed up 4 times last in 1992
Sample = 5,377 (755 AA & 4,622 C)
Controls = 5,176
Cases = 201(4%); 47 AA & 154 C
Self-report of non-recreational activity as very active, moderately active or inactive and recreational activities as much, moderate, or little or no exercise
No effect of non-recreational physical activity on risk of PC for C men
Decreased risk of PC for active AA men. AA men reporting low levels of non-recreational physical activity had an increased risk of PC compared with very active AA men. For AA men RR = 3.7 (CI: 1.7–8.4).
Putnam et al. USA, 
Retrospective cohort of men identified through the Iowa Cancer Registry compared with a population-based control group recruited from the same geographic area
Sample = 1,572
Controls = 1,471
Cases = 101 (6%)
Mailed questionnaire supplemented with a telephone interview of exercise during adult life
No effect. Neither leisure nor occupational physical activity associated with risk of PC
Lacey et al. China, 
Population-based case control of men diagnosed with PC between 1993 and 1995 identified through the Shanghai Cancer Institute and collaborating hospitals with a population-based control group recruited from the same geographic area
Sample = 1,624
Controls = 471
Cases with BPH = 206 (13%)
Cases with PC = 238 (15%)
Interviews of occupational activity and hours spent in four activity categories (sleeping, sedentary, moderate activity, and vigorous activities) used to calculate weekly METs and energy expenditure
No effect. No protective role of physical activity for either PC or BPH
Lee et al. USA, 
Prospective cohort of men participating in the Harvard Alumni Health Study
Sample = 8,922
Controls = 8,483
Cases = 439 (5%)
Self-report of walking, flights of stairs climbed, sports & recreational activities over the past week
No effect. No association between activity and risk of PC
Wannamethee et al. England, Wales and Scotland, 
Prospective cohort of men participating in the BRHS
Sample = 7,588
Controls = 7,468
Cases = 120 (2%)
40–59 at enrollment
Self-report of usual pattern of physical activity
Decreased risk. PC risk lower for those engaged in vigorous levels of activity. RR = 0.23 (CI: 0.06–0.94)
Norman et al. Sweden, 
Cohort study of men identified from Swedish census information that allowed for classification of physical activity levels at work and cross-referenced with the Swedish Cancer Registry
Sample = 2,400,043
Controls = 2,307,835
Cases = 92,208 (4%)
Classification of occupational code
No effect. No association between occupational activity and PC mortality
Friedenreich et al. Canada, 
Population-based case–control study. Cases with Stage T2 or greater disease identified from the Alberta Cancer Registry and controls using random digit dialing
Sample = 2,051
Controls = 1,063
Cases = 988 (48%)
Mean 67; controls matched within 5 years to cases
Interview using the Lifetime Total Physical Activity Questionnaire
No effect of lifetime physical activity on incident PC. No association between total lifetime physical activity and PC risk was found
Decreased risk of PC depending upon age engaged in activity. Activity done during the first 18 years of life decreased risk. OR = 0.78 (CI: 0.59–1.04)
Decreased risk of PC with vigorous activity. Vigorous activity decreased risk. OR = 0.7 (CI: 0.54–0.92)
Sanderson et al. USA, 
Population-based case control study. Cases identified through the South Carolina Central Cancer Registry and controls through Medicare beneficiary files
Sample = 845
Controls = 429
Cases = 416 (49%)
Self-report of strenuous or moderate leisure-time physical activity one or more hours/week
No effect. Neither strenuous nor moderate physical activity was associated with PC risk for either AA or C men
Patel et al. USA, 
Prospective cohort of men in the American Cancer Society CPS-II followed over 9 years
Sample = 72,174
Controls = 66,671
Cases = 5,503 (8%)
50–74 at enrollment
Self-administered questionnaire of recreational activities
No effect on incident PC of recreational activity. No difference in overall risk of PC comparing men who engaged in >35 MET-hours/week of recreational physical activity to men who reported no recreational physical activity
Decreased risk of aggressive PC with higher levels of recreational activity. Reduced risk of aggressive PC in men who engaged in >35 MET-hours/week of recreational physical activity as compared with men who reported no recreational physical activity. RR = 0.65 (CI: 0.52–0.92)
Nilsen et al. Norway, 
Prospective cohort of men participating in the HUNT study and followed over 17 years
Sample = 29,110
Controls = 28,153
Cases = 957 (3%);
266 with advanced metastatic disease at diagnosis; 354 of the cases died from PC
Mean 52.3 at baseline
Self-administered questionnaire of recreational physical activities. A summary score labeled low, medium or high was computed based upon frequency, duration and intensity of activities
No effect of activity on overall risk of PC. No association between physical activity and overall risk of PC
Decreased risk of advanced PC with higher activity levels. Compared with men who reported no activity, men reporting the highest physical activity had lower risk for advanced PC
RR = 0.64 (CI: 0.43–0.95)
Decreased risk of death from PC with higher activity levels. Compared with men who reported no activity, men reporting the highest physical activity had lower risk of death from PC. RR = 0.67 (CI: 0.48–0.94)
Krishnadasan et al. USA, 
Nested case–control of men working at a nuclear and rocket engine-testing facility in Southern California
Sample = 2,167
Controls = 1,805
Cases = 362 (17%)
Age not reported, but cases and controls matched for age
Review of company records of occupational physical activity and self-report of recreational activities
Decreased risk of incident PC with occupations requiring higher physical activity for aerospace workers. High activity levels at work were associated with lower incidence of PC for aerospace workers. OR = 0.55 (CI: 0.32–0.95)
No effect of occupational activity on incident PC for radiation workers. No association between activity levels at work and PC for radiation workers
Laguiou et al. Greece, 
Case–control of men identified from hospital admission records of six hospitals in Athens, Greece and part of one of two case–control studies conducted between 1994 and 1997
Sample = 750
Controls = 246
Cases with BPH = 184 (25%)
Cases with PC = 320 (43%)
5% <60, 41% between 60 & 69, 37% between 70 and 79, 17% >80
Occupations before retirement classified as high, medium, or low physical activity levels
No effect. No significant association between physical activity and PC
Moore et al. USA, 
Prospective cohort of men in the NIH AARP Diet and Health Survey followed over 8.2 years
Sample = 293,902
Controls = 276,030
Cases = 17,872 (6%); including 1,942 advanced disease & 513 fatal
50–71 at baseline
Self-administered questionnaire of frequency of vigorous exercise at least 20 min in duration currently and during adolescence
No association between current exercise and incident, advanced, or fatal PC
Decreased risk with activity during adolescence. An increased level of physical activity during adolescence was associated with a 3% reduction in total PC risk. All comparisons are to no or rare exercise; RR for exercise 1–3 times/month = 1.01 (CI: 0.92–1.10), for exercise 1–2 times/week = 1.02 (CI: 0.95–1.10), for exercise 3–4 times/week = 0.99 (CI: 0.92–1.05), for exercise ≥5 times/week = 0.97 (CI: 0.91–1.03). Although all RRs crossed 0, the significance for the trend with increasing activity, Ptrend = 0.03
Wiklund et al. Sweden, 
Population-based case control of men participating in the CAPS study enrolled in Jan 2001 and Oct 2003
Sample = 2,567
Controls = 1,118
Cases = 1,449 (56%)
Mean 66.8 (SD = 7.3) for cases, 67.7 (SD = 7.5) for controls
Self-report of lifetime occupational, household, and recreational activities
Increased risk of PC in the most active men as compared with the least active. For household activity OR 1.44 (CI: 1.08–1.92); for recreational activity OR = 1.56 (CI: 1.16–2.10)
Antonelli et al. USA, 
Prospective case–control of men undergoing prostate needle biopsy at Duke Veterans Administration Medical Center in Durham, North Carolina
Sample = 190
Controls = 111
Cases = 79 (43%); 35 of which had high-grade disease with a Gleason score ≥7
59–70, median 62
Self-report using the Grodin Leisure-time exercise questionnaire
Decreased risk of PC with higher activity levels. Men who reported ≥9 MET-hours/week were significantly less likely to have cancer on biopsy. OR = 0.35 (CI: 0.17–0.75)
Decreased risk of high-grade PC with higher activity levels. Men with cancer who reported ≥9 MET-hours/week were significantly less likely to have a high-grade disease. OR = 0.14 (CI: 0.02–0.94)
Johnsen et al. Europe, 
Prospective cohort of men from eight European countries participating in the EPIC study followed for 8.5 years
Sample = 153,455
Controls = 150,997
Cases = 2,458 (2%)
Occupational and leisure-time activities calculated in METs
No effect of leisure-time activities on incident PC. No association between PC risk and leisure-time activities
Decreased risk of advanced PC with increasing occupational activities. Progressively higher levels of occupational physical activity associated with lower risk of advanced PC. Comparisons are to sedentary occupations; IRR for standing occupations = 0.79 (CI: 0.58–1.07), manual occupations = 0.75 (CI: 0.53–1.05). Although all IRR crossed 0, the significance for the trend with increasing occupational activity, Ptrend = 0.024
Moore et al. USA, 
Prospective cohort of men enrolled in the NIH AARP Diet and Health Survey followed over 7 years
Sample = 163,677; 160,006 C & 3,671 AA
Controls = 153,682; 150,382 C & 6,600 AA
Cases = 9,995 (6%); 9,624 C & 371 AA
51–72 at the start of follow-up
Questionnaire of time/week spent engaging in different types of physical activity over 4 time periods (i. e., ages 15–18, 19–29, 35–39, and past 10 years)
No effect of activity on incident PC for C men. No association between physical activity and PC in C men regardless of age when exercised
Decreased risk of PC in more active AA men. AA men engaging in ≥ 4 h of moderate to vigorous activity between the ages of 19 and 29 had lower risk of PC as compared to AA men engaging in physical activity infrequently. RR = 0.65 (CI: 0.43–0.99)
Orsini et al. Sweden, 
Prospective cohort of Swedish men residing in one of two counties in central Sweden followed for 10 years
Sample = 45,887
Controls = 43,152
Cases = 2,735 (6%)
45–79 at enrollment between 1997 and 1998
Self-report of occupational activity and walking or bicycling
Decreased risk of PC with more active occupations. Compared with men who mostly sit during work, men who sit only half of the time experienced 20% lower risk for PC (7–31%)
Decreased risk of PC with increased recreational activity. As walking or bicycling increased 30 min/day in the range of 30–120 min/day, the risk decreased 7% for all PC (0–16%), 8% for localized PC (1–12%), and 12% for advanced PC (2–20%)
Batty et al. England, 
Prospective cohort of male, non-industrial, government employees participating in the Whitehall cohort study followed over 40 years
Sample = 17,934
Controls = 17,356
Cases = 578 (3%)
40–69 at enrollment between 1967 & 1970
Self-report at enrollment into study of number of minutes walked or bicycled to work and classification as inactive, moderately active, or active based upon report of hobbies or sports
No effect. No evidence that physical activity predicted the risk of death from PC
Joshu et al. USA, 
Retrospective cohort. men with localized PC treated with prostatectomy between 1993 & 2006 by the same surgeon at Johns Hopkins Hospital in Baltimore, Maryland; mean follow-up 7.3 years
Sample = 1,337 Cases = 1,337 (100%)
Self-report of physical activity 5 years before surgery and 1 year after surgery
No effect. Physical activity ≥5 h/week did not attenuate the risk of recurrence in men who gained more than 2.2 kg
Kenfield et al. USA, 
Cohort study of male health professionals, part of the health professionals follow-up study followed for at least 4 years
Sample = 51,529
Controls = 48,824
Cases = 2,705 (5%) diagnosed with non-metastatic disease; of these, 110 eventually died of PC
Mean between 68 and 70 depending upon activity levels
Self-report of walking, jogging, running, bicycling, swimming, tennis, calisthenics, number of flights of stairs climbed, heavy outdoor work, and weight training interpreted as vigorous or non-vigorous depending upon the analysis of MET-hours/week
Decreased risk. Men who engaged in ≥3 h/week of vigorous activity had a lower risk of death from PC as compared with men who engaged in <1 h/week of vigorous activity. HR = 0.39 (CI: 0.18–0.84)
A total of 23 studies were identified [12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34]. Five reports published since 2000 updated studies reviewed in the Friedenreich and Thune  review. Using the Harvard Alumni Health Study data, a 2001 report by Lee et al.  updated a 1976 report by Polednak et al.  and a 1994 report by Lee et al. . Using the Shanghai Cancer Registry, a 2001 report by Lacey et al.  updated a 1994 report by Hsing et al. . Using the National Health and Nutrition Examination Survey I (NHANES I) data, a 2000 report by Clarke and Whittemoore  updated a 1995 report by Steenland et al. . Using the Health Professionals Follow-Up Study data, a 2011 report by Kenfield et al.  updated a 1998 report by Giovannucci et al. . And using the Nord-Trøndelag Health Study (HUNT) data, a 2006 report by Nilsen et al.  updated a 2000 report by Lund Nilsen et al. . All five more recent reports were included in this review. One study by Hayes et al.  was not included in the Friedenreich and Thune  review and is included in this review. Two reports published since 2000 using the Health Professionals Follow-Up Study data to investigate the relationship between physical activity and prostate cancer were identified [21, 34]. Only the more current analysis by Kenfield et al.  was included in this review. Moore published two reviews using the National Institutes of Health (NIH) American Association of Retired Persons (AARP) Diet and Health Survey [26, 30]. Both reports are included in this review as the earlier publication focuses on the relationship between physical activity and the aggressiveness of prostate cancer  while the later publication focuses on the relationship between physical activity and ethnicity in men with prostate cancer . Thus, a total of 22 studies will be reviewed here to update the 2001 Freidenreich and Thune  review.
Of the 22 studies reviewed, 14 used a cohort design [13, 14, 16, 17, 18, 22, 23, 26, 29, 30, 31, 32, 33, 34] and eight used a case–control design [12, 15, 19, 20, 24, 25, 27, 28]. Thirteen studies were conducted in North America (Canada and the United States) [12, 13, 14, 16, 19, 20, 22, 24, 26, 28, 30, 33, 34], eight were conducted in Europe (England, Greece, Norway and Sweden) [17, 18, 23, 25, 27, 29, 31, 32], and one was conducted in Asia (China) . The number of men included in the analyses ranged from 190  to over 2 million . Physical activity was assessed primarily using mailed questionnaires, sometimes with an interview to clarify information. Several studies determined the level of physical activity based upon the man’s occupation [12, 15, 18, 24, 25, 27, 29, 31]. Of the 22 studies, 12 suggested that physical activity reduced the risk of prostate cancer [13, 17, 19, 22, 23, 24, 26, 28, 29, 30, 31, 34], nine found no association between physical activity and the risk of prostate cancer [12, 14, 15, 16, 18, 20, 25, 32, 33], and one study showed an increased risk of prostate cancer . The odds and risk ratios reflected a reduction in risk and so were expressed as numbers less than 1. The effect of risk reduction was considered large if the odds or risk ratio was between 0.00 and 0.20, considered moderate if between 0.21 and 0.50, and considered small if between 0.51 and 0.99. Using these definitions, eight of the 12 studies suggesting a reduced risk of prostate cancer reported a small effect size [19, 22, 23, 24, 26, 29, 30, 31], three reported a moderate effect size [17, 28, 34], and two reported a large effect size [13, 28]. One study  reported a moderate effect protective against incident prostate cancer and a large effect protective against advanced prostate cancer and so is referenced twice in the previous sentence summarizing the findings. The effect size for the one study showing an increased risk of prostate cancer with exercise  was small (i.e., ≤2). Overall, the findings from these studies suggest that the higher levels of exercise are protective against the development of prostate cancer.
Reported occupational activity, presumably the most consistent daily activity engaged in on a regular basis, might have the greatest impact on the risk of prostate cancer. However, only three [24, 29, 31] of the eight [12, 15, 18, 24, 25, 27, 29, 31] studies reviewed that specifically assessed occupational activity found a decreased risk of prostate cancer associated with occupations requiring a greater level of physical activity.
Six of the reviewed studies looked specifically at advanced, aggressive, and/or fatal prostate cancers [22, 23, 26, 28, 31, 34], all but one  found an inverse relationship between physical activity and the risk of developing advanced, aggressive, or fatal disease. The effect of physical activity on reducing the risk of developing advanced, aggressive, and/or fatal prostate cancers was small in three of the studies [22, 23, 31] and moderate in two of the studies [28, 34].
Ethnicity has been identified as a risk factor for prostate cancer with African-American men being at the greatest risk . Four studies specifically considered ethnicity in assessing the impact of physical activity in the risk of prostate cancer [12, 13, 20, 30]. Inactivity was found to be a significant risk factor for African-American but not Caucasian men in one study . Inactive African-American men had almost four times the risk of prostate cancer as compared with very active African-American men (relative risk = 3.7, 95% confidence interval 1.7–8.4). In this prospective cohort study, only 755 African-American men were enrolled and only 47 developed prostate cancer. A larger cohort study enrolling 3,671 African-American men, 371 of whom developed prostate cancer, found that African-American men engaging in at least 4 h of moderate to vigorous activity per week had a lower risk of prostate cancer as compared with African-American men engaging in physical activity infrequently . But the effect was much smaller with the confidence interval approaching 1. And two studies [12, 20] found no association between physical activity and risk of prostate cancer in either African-Americans or Caucasians.
Three studies evaluated physical activity across the lifespan [19, 26, 30]. All three found that activity performed earlier in life reduced the risk of prostate cancer; two studies found exercise performed before age 18 was beneficial [19, 26]. One study found exercise performed between the ages of 19 and 29 protective, but only in African-American men . The effect of physical activity, even at a younger age, was small.
After reviewing 24 studies, Friedenreich and Thune concluded that “the epidemiologic evidence is currently inconsistent and the magnitude of the risk reduction observed is small” [, p. 461]. Neither do the 22 studies published since their review have consistent findings and any risk reduction found is still small; however, the trend toward physical activity being protective against the development of prostate cancer persists. Summarizing the findings from the 18 studies reviewed by Friedenreich and Thune , not including the six studies with more recent findings published after their analysis, as well as the 22 published since then (40 studies in total), 22 (55%) suggested that physical activity reduced the risk of prostate cancer, 14 (35%) found no association between physical activity and risk of prostate cancer, and four (10%) showed an increased risk of prostate cancer. In the 22 studies reporting a protective effect, the effect sizes were predominantly small (n = 14; 64%) or moderate (n = 6; 27%). In the four studies reporting that physical activity increased a man’s risk for prostate cancer, the effect size was small in three studies (75%) and moderate in one study (25%). In summary, the published research conducted over the past 35 years suggests that the predominant effect of physical activity in reducing the risk of prostate cancer is most likely vigorous activity engaged in as an adolescent and young adult, and the risk reduction is for advanced and aggressive disease. This may be more important for African-American men as compared with Caucasians.
Physical activity as a moderator in prostate cancer prevention
The potential for prostate cancer prevention through exercise might be in the control of weight. Although not conclusive, research has suggested that obesity is a risk factor for prostate cancer recurrence, high-grade disease at the time of diagnosis, and the development of advanced or fatal prostate cancer . Physical activity without weight control did not confer any protection in one study . Men who gained more than 2.2 kg (almost 5 pounds) following a prostatectomy for prostate cancer but who also engaged in physical activity (≥5 h/week) did not modify their risk of prostate cancer recurrence . Still, effective weight control programs generally include an exercise component, and so using exercise to control weight may also reduce the risk of prostate cancer and or prostate cancer recurrence. Additional investigation is warranted to ascertain if physical activity independently affects risk for prostate cancer, or if physical activity moderates the risk of prostate cancer through weight control. Generally small effect sizes found in the epidemiological studies reviewed might be reflective of the moderating effect of exercise.
Physical activity in the prevention of prostate cancer recurrence
There is just beginning to be some evidence that physical activity may reduce the recurrence of prostate cancer, another approach to prevention. A prospective observational study of 1,455 men diagnosed with localized prostate cancer found that men walking briskly for 3 h/week or more had a 57% lower rate of progression as compared with men who walked at an easy pace for less than 3 h/week . And, a randomized clinical trial tested a combined exercise and diet intervention randomizing 93 men with biopsy-proven prostate cancer, serum prostate-specific antigen (PSA) between 4 and 10 ng/ml and Gleason score ≤7 who had opted not to undergo conventional treatment to either an intensive lifestyle program or usual care . The intensive lifestyle program included both diet and exercise recommendations. The diet was vegan supplemented with soy, fish oil, vitamins E and C, and selenium. The exercise prescription included moderate aerobic exercise. Stress management and support group participation were also part of the intervention. After 1 year, none of the men in the intervention group but six in the usual care group underwent conventional treatment after an increase in PSA and/or progression of disease on magnetic resonance imaging (P = 0.016). These are important findings; however, because changes in both diet and exercise were made simultaneously, the effects may have been related to either a change in activity or diet or the change in both. Additional research is warranted to determine the unique role physical activity might play in the prevention of prostate cancer recurrence.
Exercise in men rehabilitating from prostate cancer and treatment
Exercise is now widely recognized as an important component of survivorship following the diagnosis and treatment of cancer. At least 15 meta-analyses have documented the safety and tolerability as well as the benefits of exercise. While women with breast cancer have been the focus of many of these studies, the role of exercise specifically in men with prostate cancer is now being investigated. One review of eight studies focused on the effects of physical activity on health outcomes following prostate cancer diagnosis and identified improved muscular fitness, physical functioning, body composition, health-related quality of life, and sexual functioning, as well as decreased fatigue . Research testing exercise interventions in men following the diagnosis and treatment of prostate cancer continue to be published [46, 47, 48] providing a better understanding of the role of exercise in prostate cancer survivors.
Need for biomarkers in prevention research
One of the challenges of prevention research has been identifying biomarkers responsive to a change in health behavior such as exercise . In prostate cancer, insulin, insulin-like growth factor-I (IGF-I), insulin-like growth factor binding protein-1 (IGFBP-1), sex hormone-binding globulin (SHBG), expression of p53, and growth of prostate cancer cells in vitro have been suggested as possible markers of change in diet and/or exercise [50, 51, 52, 53, 54, 55, 56, 57]. Each of these markers probes the signaling pathway at different junctures. Measuring growth of prostate cancer cells incubated in serum from men exposed to an experimental condition assesses a later step in the formation of prostate cancer hypothesizing that something in the man’s serum produced by the intervention inhibits or promotes cancer. One type of prostate cancer cells commonly used in these studies is lymph node cancer prostate (LNCaP). LNCaP cells are androgen-sensitive human prostate adenocarcinoma originally derived from the lymph node of a man with an aggressive metastatic prostate cancer. Ornish et al.  reported that LNCaP cell growth in serum from men randomized to a diet and exercise intervention was inhibited almost eight times more than when the cells were grown in serum from the men randomized to usual care. But as previously cautioned, because changes in both diet and exercise were made simultaneously, the effects may have been related to either a change in activity or diet or the change in both. Also, it is not known whether more indolent prostate cancer cells respond in the same way the highly aggressive LNCaP cells respond. However, biological markers such as these should be tested further as possible outcome measures of effective prevention strategies.
Considerations for future research
Prevention research is difficult to conduct and easily confounded by the near impossibility of accurately documenting, much less controlling for, health habits such as physical activity and diet over time. While most study participants intend to faithfully report activity and diet, memory is just not that reliable instrument. Further confusing this picture is the interaction between physical activity and diet to affect metabolism. Often activity and dietary habits co-vary in that people who exercise regularly tend to eat a healthier diet and are more successful at maintaining a healthy weight. Potentially researchers should be studying clusters of behaviors rather than focusing on one health behavior independent of other lifestyle activities. Sophisticated multi-dimensional biological models are needed to fully appreciate the complexities of health and how factors such as physical activity at different doses and at different times impact the risk of diseases, such as prostate cancer.
Future research should employ prospective randomized clinical trial designs. Prevention trials are generally expensive requiring the recruitment of large numbers of participants and close tracking and follow-up of these individuals over many years. Compliance with long-term interventions, especially a behavioral intervention such as exercise, is also problematic. Studies testing an exercise intervention are further challenged in that randomization to a treatment arm prohibiting exercise would be considered unethical by most clinical investigators because exercise is beneficial to so many aspects of health. Many exercise studies use a usual care arm that may or may not include exercise, but it is difficult to make definitive conclusions about an intervention without control of the independent variable.
In addition to research using stronger study designs, research is needed to evaluate the differential effects of physical activity performed early in life, at different intensity levels, and by individuals of different ethnicities. Also, additional research is needed to more fully understand the role of physical activity in aggressive and advanced disease as well as prostate cancer recurrence.
More researchers could take advantage of the natural history of prostate cancer to design prospective intervention trials as was done by Ornish et al. . Because many prostate cancers are slow growing and the risk of possible side effects of treatment such as incontinence and impotence untenable to some men, close observation can be an alternative to surgery or radiation therapy and an opportune time to conduct prospective research. These men could be recruited into cleverly designed clinical trials to test preventive strategies such as exercise, assessing for progression of disease.
Future research would also benefit from more extensive use of biomarkers and understanding of the biology they reflect. Normally, the physiological benefits of starting an exercise program take 8–12 weeks to manifest. But, if change in LNCaP cell growth can be observed after only 11 days of exercise, as suggested by some research [49, 52, 53], this measure would be a very useful tool in larger prevention studies. Use of biomarkers may help sort out the possible moderating and mediating effects of physical activity on prostate cancer. Could it be that physical activity helps men control their weight thereby moderating the risk of prostate cancer, or does physical activity directly impact the risk for prostate cancer? Use of biomarkers could inform this discussion.
There is a growing body of epidemiological research suggesting that physical activity is protective against the development of prostate cancer. This paper reviewed 22 studies published in the past 12 years updating an earlier review published by Friedenreich and Thune  finding that although some studies show no benefit and a very few show increased risk for prostate cancer, the majority of studies document a significant, albeit small, protective effect of physical activity.
Physical activity is a powerful intervention playing a key role in the prevention of myriad diseases, improving quality and quantity health in general, and reducing the burden of chronic disease and illness. The role of physical activity in the prevention of prostate cancer has still not been clearly elucidated, but research conducted over the past 35 years continues to suggest that there is a link.
University of Texas Health Science Center at San Antonio Cancer Therapy and Research Center National Cancer Institute CCSG Grant P30CA054174 support.
Conflict of interest
I certify that there is no actual or potential conflict of interest in relation to this article.