Risk Factors for Prostate Cancer
The established non-modifiable risk factors for prostate cancer are age (>70% diagnosed after age 65), African ancestry (64% higher incidence and 2.3-fold mortality increase), and family history (2.5-fold risk with one first-degree relative, escalating to 11-fold with three affected relatives). 1
Non-Modifiable Risk Factors
Age
Age represents the strongest risk factor for prostate cancer. More than 70% of all patients are older than 65 years at diagnosis, with a median age of 71 years 2. A 50-year-old man has a 42% lifetime chance of developing histological prostate cancer, though only 9.5% will develop clinically important disease and 2.9% will die from it 2. The risk increases stepwise with each age increment when accounting for biopsy verification 3.
Race and Ethnicity
African American men face substantially elevated risk, with 64% higher incidence and 2.3-fold increased mortality compared to white men 4. They present with higher-grade prostatic intraepithelial neoplasia, higher Gleason scores, and more advanced pathologic stage 2. However, stage- and grade-adjusted mortality rates are similar between races, suggesting differences may relate to access to care rather than tumor biology 2. Geographic variation is striking—Northern and Western Europe show age-standardized rates of 111.6 and 97.2 per 100,000 men annually, while Japan and China have markedly lower rates (2-20 deaths per 100,000) 2, 1.
Family History and Genetics
Family history creates dose-dependent risk escalation: 2.5-fold with one first-degree relative, 5-fold with two relatives, and 11-fold with three affected relatives 2. An estimated 9% of prostate malignancies stem from inherited predisposition 2. When a first-degree relative is diagnosed before age 60, risk increases 2.1- to 2.5-fold 4.
Specific genetic syndromes warrant heightened surveillance:
- BRCA2 mutations: 2- to 6-fold increased risk with more aggressive phenotype and reduced survival 4
- BRCA1 mutations: Less consistent but elevated risk 4
- Lynch syndrome (MLH1, MSH2, MSH6, PMS2, EPCAM): 2- to 5-fold increased risk 4
The NCCN recommends BRCA2 carriers begin screening at age 40 4. Approximately 42% of early-onset cancers (diagnosed before age 55) may be inherited 2.
Modifiable and Environmental Factors
Physical Activity and Anthropometric Measures
Height shows consistent association with prostate cancer risk across both observational and genetic studies (meta-analysis OR=1.09; Mendelian randomization OR=1.07 for aggressive disease) 5. Physical activity demonstrates protective effects, with occupational physical activity showing OR=0.87 in meta-analyses and accelerometer-measured activity showing OR=0.49 in Mendelian randomization studies 5.
Body mass index shows complex associations that vary by study design. Higher BMI associates with increased risk in community-based studies but this relationship attenuates when accounting for biopsy bias 3. For fatal prostate cancer specifically, higher BMI increases risk 6.
Smoking and Alcohol
Current smoking paradoxically shows reduced incidence (OR=0.74 for current smokers ≥20 cigarettes/day) but increased mortality risk 2, 5. This heterogeneity suggests smoking may influence late-stage progression rather than cancer initiation 7. Smoking initiation in Mendelian randomization shows OR=0.91 5. Heavy alcohol consumption (≥150g/day for ≥10 years) increases risk 2.83-fold 8, though moderate alcohol intake shows no consistent association 7.
Diet and Nutrition
Diet represents one of the most cited environmental factors, supported by increased cancer rates in immigrants from low-incidence Asian countries to the United States 2. However, no definitive dietary interventions have been established for prevention 1.
Specific dietary factors with suggestive evidence include:
- Tomato sauce/lycopene: Inverse association with risk 6
- Calcium: High intake associates with increased risk, particularly for advanced and fatal disease 6
- Alpha-linolenic acid: Positive association with both incidence and fatal disease 6
- Dairy products and meat: Possible increased risk 7
- Cruciferous vegetables and soy: Possible protective effects 7
Clinical Conditions
Diabetes shows protective association (OR=0.78) 9. This inverse relationship persists across multiple studies. Testosterone supplementation in hypogonadal men does not increase prostate cancer risk 1.
Important Clinical Caveats
Detection bias significantly distorts risk factor associations. Men with family history or elevated PSA undergo biopsy more frequently, artificially inflating apparent risk associations 3. Studies without mandated biopsies overestimate associations for family history (56% relative difference), PSA level (69% relative difference), and other factors 3.
Risk factors for incidence differ from those for mortality. Smoking, BMI, height, and certain dietary factors show weak or no association with incident cancer but strong associations with fatal disease 6. Advanced stage at diagnosis serves as a better surrogate for fatal prostate cancer than high Gleason grade 6.
PSA screening era has altered risk factor relationships. The proportion of advanced-stage cancers decreased substantially after PSA introduction, though risk factors for advanced disease remained similar across eras 6.
Risk Assessment for Clinical Practice
For men considering screening or risk stratification, prioritize:
- Age >65 years (strongest predictor)
- African ancestry (2.3-fold mortality increase)
- Multiple affected first-degree relatives (especially if diagnosed <60 years)
- BRCA2, BRCA1, or Lynch syndrome carriers (consider genetic testing if family history suggests hereditary pattern)
- Baseline PSA value (stronger predictor than family history or race alone) 4
Despite extensive research, no convincing modifiable risk factors have been definitively established for primary prevention 5. The evidence remains suggestive rather than conclusive for lifestyle interventions, though physical activity, maintaining healthy weight, and avoiding heavy alcohol consumption represent reasonable recommendations based on available data.