We studied the effect of two independent variables, the pectin/calcium chloride weight ratio and the overall matrix weight in HPMC/pectin/calcium matrix tablet, on the release of indomethacin. A two-factor 5-level central composite experimental design was employed. Responses of the Peppas correlation parameters n and K and the 10% release time （T0.1） were optimized by response surface methodology. Significant effect of the independent variables on the biphasic release parameters, n and K, was observed. N, K and T0.1 were well fitted with the second-order quadratic equations rather than linear equations. Moreover, the mathematic fitting and the response surfaces showed significant cross-interaction between the pectin/calcium chloride ratio and the overall matrix weight. The optimal formulation with larger n, longer T0.1 and smaller K consisted of medium pectin/calcium chloride ratio around 1.0 and medium matrix weight around 200 mg. Validation studies on the optimal formulations showed good predictability of the n, K and T0.1 values with biases within the range of-7.33% and 6.26%. Our results support that central composite design can be used to optimize drug release from HPMC/pectin/calcium matrix tablet with high predictability.
