The big gap between efficacy of population level prevention and expectations due to heterogeneity andcomplexity of cancer etiologic factors calls for selective yet personalized interventions based on effective riskassessment. This paper documents our research protocol aimed at refining and validating a two-stage and webbasedcancer risk assessment tool, from a tentative one in use by an ongoing project, capable of identifyingindividuals at elevated risk for one or more types of the 80% leading cancers in rural China with adequatesensitivity and specificity and featuring low cost, easy application and cultural and technical sensitivity forfarmers and village doctors. The protocol adopted a modified population-based case control design using 72,000 non-patients as controls, 2, 200 cancer patients as cases, and another 600 patients as cases for externalvalidation. Factors taken into account comprised 8 domains including diet and nutrition, risk behaviors, familyhistory, precancerous diseases, related medical procedures, exposure to environment hazards, mood and feelings,physical activities and anthropologic and biologic factors. Modeling stresses explored various methodologies likeempirical analysis, logistic regression, neuro-network analysis, decision theory and both internal and externalvalidation using concordance statistics, predictive values, etc..