We aimed to investigate the association of a sustainable diet with a long-term reduction in waist circumference (WC) while identifying novel biomarkers for WC reduction (WCR). The participants were recruited initially during 2004–2013 in a large hospital-based cohort, and the follow-up measurements were conducted during 2012–2016. The 65,611 adults aged 45–75 were categorized into WC-loss (n = 22,290) and WC-gain (n = 43,321). Each study investigated demographic, anthropometric, biochemical, genetic, and dietary factors. The modified Healthy Eating Index (MHEI), dietary patterns, and glycemic index were calculated from a validated semi-quantitative food frequency questionnaire. Novel biomarkers influencing WC reduction were identified using machine learning approaches. A WCR was inversely associated with metabolic syndrome (MetS) risk and its components. Daily energy intake did not differ between those with and without WCR. However, MHEI, which represents diet quality, demonstrated a positive association with WCR. Among various dietary patterns, the Asian-style balanced diet (ABD), including more fermented soybeans and less restricted salt than the Diet Approach to Stop Hypertension, was positively associated with WCR. However, an inverse association was observed between the diet that was high in noodle and processed meat consumption and that which was high in rice consumption. However, the PRS for abdominal obesity did not significantly interrupt WCR. The receiver operating characteristic curve in the prediction model for WCR was about 0.86. The biomarkers in the models included MetS components, inflammation index, diet components, alcohol consumption, and smoking status, but not genetic factors. In conclusion, adopting a high-quality diet with a high MHEI like ABD leads to WCR, irrespective of genetic influences. These results could be applied to develop effective strategies for preventing and managing abdominal obesity.