Objective：To investigated the predictive value of the lipid accumulation product（LAP） for prediabetes development of diabetes and compared its predictive power with various anthropometric indices. Methods：The participants in this study were selected from the epidemiological study of cancer risk of Chinese patients with type 2 diabetes mellitus in Luzhou area. A total of 1 125 participants with prediabetes were selected from the baseline survey. The subjects were divided into four groups（Q1，Q2，Q3，Q4） according to the quartiles of the LAP and followed for five years. We assessed the predictive power of adiposity indices by performing univariate and multivariate binary logistic regression analysis and calculating the area under the receiver operating characteristic（ROC） curve according to their quartiles. Results：Of the 1 064 participants who completed the follow-up，237 were diabetic. The comparison of the four groups of subjects into diabetes showed that Q4 group（92 cases）>Q3 group（69 cases）>Q2 group（48 cases）>Q1 group（28 cases），the difference was statistically significant（χ2=49.308，P=0.000）. Logistic regression analysis showed that after adjusting for any variables and adjusting for age，gender，height，systolic blood pressure，TG，HDL-C，smoking and tea consumption，individuals in higher LAP quartiles at baseline were more likely to develop diabetes than those in lower LAP quartiles. The area under the ROC curve for LAP was significantly greater than that of other adiposity indices，indicating that LAP，compared with BMI，waist circumference（WC） and waist-to-hip ratio（WHR），had a better value in predicting the outcome of predia－betic subjects to diabetes. Conclusion：LAP was strongly asso－ciated with an increased risk of diabetes in prediabetes，and the association was numerical-dependent. LAP may be a predictor of the pro-gression from prediabetes to diabetes. Our data demonstrate that the LAP was a better predictor of diabetes than the BMI，waist circumference and waist-to-hip ratio in prediabetes.