A model for Calcitonin purification from an industrial peptide mixture using polymer-based reversed phase columns was developed. Regressed competitive bi-Langmuir isotherm parameters for pure calcitonin are strong functions of the eluent composition and were correlated to the overall Henry coefficient only. It is shown that, by keeping constant these correlations, the adsorption behavior of impurities can be predicted simply by estimating their Henry coefficient from the raw mixture. The same can be repeated with very good results when changing the stationary phase or the organic solvent in the eluent. In both cases, only the Henry coefficient must be re-estimated from pulse injections.