Carbohydrate is a major composition in oral nutrition supplement. Modification of carbohydrate composition using tapioca resistant maltodextrin (TRM) may improve the glycemic response. This study aimed to investigate the effect of TRM and TRM containing-ONS on glycemic response, satiety, and gastrointestinal (GI) tolerability in healthy individuals. Physicochemical properties and sensory evaluation of developed ONS were also investigated. In Phase I, the peak of plasma glucose and insulin response after TRM (50 g) ingestion was lowest compared to 50 g glucose (GL) and tapioca maltodextrin (TM) in 16 healthy participants. Peak of plasma glucose of TRM, GL, and TM were 104.60±2.63, 135.87±4.88, and 127.93±4.05 mg/dl, respectively, while peak of serum insulin was 13.001±2.12, 47.90±11.93, and 52.96±17.68 µIU/ml. No significant effects were observed on subjective appetite during 180 min of study. Fifty grams of TRM increased flatulence in healthy participants. In phase II, three formulas were developed: original (0 g TRM), RMD15 (2.7 g TRM), and RMD30 (5.4 g TRM). The RMD30 significantly had lowest viscosity (viscosity of original, RMD15, RMD30 were 36.37±0.25, 34.60±0.06, 34.07±0.09 cP, respectively) and highest water activity (0.35±0.01, 0.33±0.01, 0.37±0.01 respectively) compared to original formula. The overall acceptability scores of developed formula were more than 7. In phase III, seventeen healthy participants were included to examine the acute effect of 1 serving of three developed formulas on glycemic response and tolerability. The RMD30 formula showed lowest postprandial blood glucose compared to original and RMD 15 (113.33±4.44 vs 119.25±4.67 and 114.42±6.43 mg/dl, respectively). Meanwhile, the Area Under Curve of serum insulin in RMD30 group was lower by 33.1% compared to original group (2,320±570.76 vs 3,470.12±531.86 µIU/ml). Subjective appetite responses were not significantly different among three formulas. All formulas were well tolerated by healthy participants. In phase IV, RMD30 formula was chosen to be supplemented in normoglycemic and prediabetic participants. Bodyweight did not significantly change after ingestion of 1 serving/d of RMD30 for 3 weeks (66.56±3.66 vs 66.99±3.69 kg, p=0.069). Food intake significantly increased at week 3 following supplementation compared to baseline (1,144.87±87.10 vs 1,477.76±94.16 kcal, p=0.001), without affecting habitual food intake. There were no significant changes in gastrointestinal symptoms. In conclusion, TRM lowers the postprandial plasma glucose and insulin response in healthy participants. Incorporation of TRM in nutritionally complete ONS decreased viscosity and increased water activity, while increased the sensory acceptability score. The replacement of TM using TRM by 30% in ONS decreased the insulin response, without significantly affected the glucose and satiety. Supplementation of RMD30 daily for 3 weeks increased total food intake in healthy and prediabetic participants. Developed formula was well-tolerated in acute and three-weeks study.