The present research was conducted to assess the influences of starvation and refeeding on growth, nonspecific immunity and lipid metabolic adaptation in Onychostoma macrolepis. To date, there have been no similar reports in O. macrolepis. The fish were randomly assigned into two groups: control group (continuous feeding for six weeks) and starved–refed group (starvation for three weeks and then refeeding for three weeks). After three weeks of starvation, the results showed that the body weight (BW, 1.44 g), condition factor (CF, 1.17%), visceral index (VSI, 3.96%), hepatopancreas index (HSI, 0.93%) and intraperitoneal fat index (IPFI, 0.70%) of fish were significantly lower compared to the control group (BW, 5.72 g; CF, 1.85%; VSI, 6.35%; HSI, 2.04%; IPFI, 1.92%) (p < 0.05). After starvation, the serum triglyceride (TG, 0.83 mmol/L), total cholesterol (T-GHOL, 1.15 mmol/L), high-density lipoprotein (HDL, 1.13 mmol/L) and low-density lipoprotein (LDL, 0.46 mmol/L) concentrations were significantly lower than those in the control group (TG, 1.69 mmol/L; T-GHOL, 1.86 mmol/L; HDL, 1.62 mmol/L; LDL, 0.63 mmol/L) (p < 0.05). The activities of intestinal digestive enzymes (amylase, lipase and protease) in the starved-refed group were significantly lower than those in the control group after three weeks of starvation (p < 0.05). The highest activities of immune enzymes such as lysozyme (LZM), acid phosphate (ACP), alkaline phosphate (ALP), superoxide dismutase (SOD), glutathione peroxidase (GSH-PX) and catalase (CAT) in the hepatopancreas were presented in the starved–refed group at second week, and significantly higher than those in the control group (p < 0.05). Meanwhile, starvation significantly improved intestinal immune enzymes activities (p < 0.05). the lowest TG contents and the highest expression levels of lipolysis genes including hormone-sensitive lipase (HSL) and carnitine palmitoyl transferase 1 isoform A (CPT-1A) appeared in the hepatopancreas, muscle and intraperitoneal fat after starvation, indicating the mobilization of fat reserves in these tissues (p < 0.05). After refeeding, the recovery of TG content might be mediated by the upregulation of the expression levels of lipogenesis genes such as sterol regulatory element binding protein 1 (SREBP1) and fatty acid synthase (FAS). Understanding the duration of physiological and metabolic changes in O. macrolepis and their reversibility or irreversibility to supplementary feeding response could provide valuable reference for the adaptability of O. macrolepis in large-scale culturing, proliferation and release.