Cancer Prevention: Lifestyle Factors with the Strongest Evidence

Cancer Is Not Primarily Genetic Bad Luck

A 2018 analysis published in CA: A Cancer Journal for Clinicians estimated that approximately 42% of all cancer cases and 45% of cancer deaths in the United States are attributable to potentially modifiable risk factors. The original Doll and Peto analysis from 1981 — remarkably still robust — estimated that diet, obesity, and tobacco alone account for a majority of cancer deaths.

This means the trajectory of most cancers is substantially within individual control — a fact frequently obscured by the understandable but misleading emphasis on genetic factors.

The Tier-1 Modifiable Risk Factors

Tobacco (Attributable Fraction: ~19% of cancer deaths) The most significant single modifiable cause of cancer. Beyond lung cancer, smoking is causally linked to cancers of the bladder, kidney, cervix, stomach, pancreas, and blood.

Excess Body Weight (Attributable Fraction: ~7.8%) Adipose tissue is not metabolically inert — it produces estrogen, inflammatory cytokines, and growth factors that directly promote tumor initiation and growth. Obesity is the second-largest modifiable cancer risk factor after tobacco.

Physical Inactivity Exercise reduces cancer risk through multiple pathways: reduced insulin and IGF-1 levels, lower estrogen, improved immune surveillance, reduced inflammation, and potentially direct tumor-suppressive effects. Regular exercise reduces risk of colon and breast cancer by approximately 25%.

Alcohol (Attributable Fraction: ~5.6%) As detailed above, alcohol carries a causal, linear relationship with multiple cancer types.

Ultra-Processed Food Multiple large prospective cohort studies have found that higher ultra-processed food consumption is associated with increased risk of various cancers, with each 10% increase in ultra-processed food intake associated with 12% higher cancer risk.

The Evidence on Early Detection

Colonoscopy: Reduces colorectal cancer mortality by 60–70% through polyp removal before malignant transformation. Recommended starting at 45 (or earlier with family history).

Low-dose CT for lung cancer: In high-risk smokers, LDCT screening reduces lung cancer mortality by approximately 20%.

PSA testing: Controversial — reduces prostate cancer mortality modestly but leads to significant overdiagnosis and overtreatment. Decision requires careful individual risk-benefit discussion.

Multi-cancer early detection (MCED) blood tests: Emerging liquid biopsy technologies (Galleri, etc.) can detect cell-free tumor DNA across 50+ cancer types simultaneously. Currently under active clinical validation.