This paper introduces the application of multispectral imaging in quantitative pathology. The automated system aims to classify microscopic samples taken by needle biopsy for the purpose of prostate cancer diagnosis. The main contribution here is that instead of analysing conventional grey scale or RGB colour images, sixteen spectral bands have been used in the analysis. Four major classes have to be discriminated. To achieve that, the same feature vector, based on texture and structural measurements, was derived for each colour band. Principal component analysis has been used to reduce the dimensionality of the combination feature vector to a manageable size. Tests has been carried out using supervised Classical Linear Discrimination method and have shown that the use of multispectral information can significantly improve the classification performance when compared with the case where this information is not taken into consideration.