The FTIR spectrum of prostate cancer cells allows the classification of anticancer drugs according to their mode of action

Abstract

The number of anticancer agents that fail in the clinic far outweighs those considered effective, suggesting that the selection procedure for progression of drug molecules into the clinic requires improvement. Traditionally, new drugs are evaluated for their potential to kill cancer cell lines. This approach is obviously not sufficient, and molecules with new modes of action are required. We suggest here that the infrared spectrum of cells exposed to anticancer drugs could offer an opportunity to obtain a fingerprint of the metabolic changes induced by the drugs. Because the infrared spectrum of cells yields a precise image of all the chemical bonds present in the sample, different drug targets are likely to yield different infrared fingerprints characteristic of the ‘mode of action’ of the therapeutic agent under investigation. In turn, drug-induced metabolic disorders should be amenable to classification in the same way that bacteria gender, species, and strains can be classified. We examined here a human prostate cancer PC-3 cell line exposed to 7 well described antimitotics. In a first step the IC₅₀ values were determined. For FTIR imaging, PC-3 cells were exposed to the IC₅₀ concentration of each drug for 48 h. About one hundred images of 4096 IR spectra at 8 cm⁻¹ spectral resolution were acquired. We show with a Student t-test that the different molecules tested induced different infrared spectral modifications. Furthermore, drugs known to induce similar types of metabolic disturbances appear to cluster when spectrum shapes are analyzed. Finally, supervised statistical methods allowed the building of an efficient and discriminant model. When the discriminant model was applied to a full infrared image a good sorting was generally obtained and misclassified spectra generally belonged to a small number of specific cells. Taken all together these data suggest that FTIR could be used for the classification of drug action.