Challenges and promises of nutritional epidemiology to investigate cancer etiology.

The relationship between diet and cancer is complex and has been the subject of scientific controversy. Recent estimates of the fraction of cancer incidence attributable to (poor) diet ranged from 5% to 13% across different study populations. Despite growing evidence linking dietary exposures to cancer risk, nutritional epidemiology faces major challenges, including exposure misclassification, confounding, and difficulties in assessing long-term and changing dietary habits. Advances such as use of biomarker and other -omics measurements, complex statistical framework, and longitudinal web-based exposure assessments have created novel opportunities for examining the diet and cancer relationship with improved accuracy. Increasing attention is also given to composite dietary indices, including ultraprocessed food, and comprehensive lifestyle scores, which better capture the complex nature of nutrition and the multifactorial nature of cancer. Methods like Mendelian randomization and molecular profiling support causal inference and elucidate biological mechanisms. Triangulation, integrating observational, biomarker, experimental, and genetic data, strengthens the evidence for dietary risk factors. Furthermore, the role of obesity and metabolic health in cancer etiology underscores the need for integrative approaches. This manuscript reviews the current state of evidence of research on diet and cancer, current methodological challenges, and promising avenues for advancing prevention strategies and scientific understanding.