Quantitative Determination of Clindamycin Phosphate in Gel Preparation Using PLSR Model

Quantitative determination of clindamycin by UV spectrophotometry is limited due to the lack of any UV chromophore of clindamycin structure. In this study, UV spectrophotometry employing partial least square regression (PLSR) model was developed with respect to highperformance liquid chromatography (HPLC) as the reference method for the quantitative determination of clindamycin phosphate in gel preparation. The successful PLSR model used UV spectral data in the region of 190-400 nm without data preprocessing. The leave-one-out cross-validation was utilized for model construction. The latent factor 2 was used in the final model with the R2 model of 0.9972, root mean square error of calibration (RMSEC), and root mean square error of cross-validation (RMSECV) of 0.0044 and 0.0048, respectively. Model ability was approved by quantitative determination of 20 validation samples that were not contributed in the model-building step. The accuracy of determination results expressed in terms of root mean square error of prediction (RMSEP), the relative standard error of prediction (RSEP%), and bias were 0.0072, 0.0447%, and -0.0025, respectively. This study demonstrated that PLSR-assisted UV spectrophotometry could be used as an alternative method for the quantitative determination of non-UV chromophore active pharmaceutical ingredients in the pharmaceutical dosage form.