Our hypothesis is that carbon-13 Nuclear Magnetic Resonance (13C NMR) spectroscopy defines distinctive and individual molecular tissue parameters which differentiate normal, reactive, and malignant tissues of a given organ. Our preliminary results indicate that conditions can be established for NMR study of non- living tissues in vitro without the occurrence of spectral changes during analysis and demonstrate consistant molecular characteristics which differentiate prostatic, colonic and gastric adenocarcinomas from matched normal or hyperplastic tissues. We propose to extend thee initial observations to a much wider range and number of lesions within prostatic and gastrointestinal tissues. We will use high resolution, natural abundance 13C NMR spectroscopy to study human malignant and non-malignant tissues through a systematic analysis of histologically well-characterized pathology specimens. The following specific approaches will be used: 1) We will further refine the study parameters in order to define the optimum conditions for in vitro tissue analysis by 13C NMR, using a rat animal model. 2) The 13C spectra of tissues which are normal, reactive, premalignant or malignant will be characterized. Molecular markers which differentiate neoplasms from surrounding non-malignant tissue and which correlate with known clinical-pathological characteristics of a given tumor will be identified. 3) A Bruker 400 MHz wide bore NMR spectrometer will be used with proton decoupling. Resonances will be identified using published NMR data, computerized NMR databases, known histochemical and biochemical characteristics of a given tissue type, tissue extracts, proton NMR spectroscopy and non-NMR techniques. NMR data which differentiates specific malignant from non- malignant lesions will provide important diagnostic information which the pathologist can utilize in vitro. Current systems of histological grading and of tumor staging have improved the recognition of more homogeneous subgroups among heterogeneous tumor types, but variations remain in the prognosis for individual tumors within a given histological classification. The proposed studies to characterize the NMR spectral parameters which differentiate neoplastic lesions will be a prelude to the determination of the role these NMR-observable compounds can play in defining prognostically different subdivisions for histologically similar prostatic and gastrointestinal tumors. In addition, the proposed in vitro studies will provide a guide for the development of high resolution NMR techniques that can ultimately provide lesion-specific metabolic information to aid in the non-invasive, in vivo diagnosis of lesions identified through Magnetic Resonance Imaging. Acquisition of this data will also lead to future tumor studies with specific 13C-enriched molecules in order to detail oncogenesis-related metabolic alterations in these malignancies.
