Keywords: 
• human serum albumin /
• injectable gel /
• PLGA /
• central composite design /
• response surface

Abstract: Objective :To prepare human serum albumin(HSA) injectable gel.To optimize the formulations of injectable gel by a central composite design/response surface methodology. Methods : Injectable gel loaded HSA were prepared with poly(lactic-co-glycolic acid)(PLGA) as carrier material and N-methyl-2-pyrrolidone(NMP) and benzyl benzoate(BB)as co-solvent.Independent variables were PLGA content and the content of co-solvent,and in vitro release was dependent variable.Linear or nonlinear mathematic models were used to estimate the relationship between independent and dependent variables.Response surfaces were delineated according to best-fit mathematic models,and optimum formulations were selected thereform.Prediction was carried out through comparing the observed and predicted values. Results :The trinomial fitting equations of indexes were all superior to their multi-linear regression equations,and the predictive values in the established mathematical model were in good conformity with the experimental values. Conclusion :Central composite design/response surface method shows a good predictability in optimizing the formula technology of BDNF-loaded PLGA nanoparticles.