We report a newly developed method for reagent-free tissue analysis, and for discovering therapeutic targets in highly heterogeneous populations. This method, referred to as infrared “spectral histopathology” (SHP), is based on measurements of the biochemical composition and compositional changes, rather than on staining patterns, tissue architecture and morphology of the specimen. This biochemical information is captured via a microscopic spectral measurement that relies on the well-known principle that every biochemical compounds exhibits a specific infrared spectrum which may be considered a fingerprint molecular signature. SHP can distinguish normal tissue types (connective tissue, fibroblasts, erythrocytes, plasma cells, macrophages, inflammatory response, etc.) by the differences in the spectral signatures, and can classify cancer types (adenocarcinoma, squamous cell carcinoma and small cell carcinoma) with high accuracy. In SHP, about 25,000 individual infrared spectra are collected for each square millimeter of tissue from pixels ca. 6 μm on edge. These pixel spectra contain an encoded snapshot of the entire biochemical composition of the pixel. The resulting spectral datasets are subsequently decoded by machine learning algorithms that reveal changes in the biochemical composition between tissue types, and between various stages and states of disease with high spatial resolution. Thus, SHP offers the clinician novel information that complements morphological and immunohistochemical data. Between a pilot study (80 cases), a large-scale follow-up study (ca. 480 cases) and a presently ongoing validation study with a National Cancer Center (420 cases), the lung cancer database at Cireca, LLC now is sufficiently large to allow the following conclusions to be made: • SHP distinguishes normal and abnormal lung tissue with an accuracy of ca. 94%. • Benign and malignant abnormal tissue can be classified with similar accuracy. • Small cell lung cancer (SCLC), squamous cell carcinomas (SqCC) and adenocarcinomas (ADC) can be classified with accuracies ranging from ca. 94% to 88%. • Inflammatory response can be detected and distal and cancer adjacent normal tissue can be distinguished. • Regions identified by immunohistochemistry to overexpress given cancer markers, e.g. TTF-1 or PD-L1, often show correspondent regions of distinct spectral signatures in SHP. An exciting aspect of SHP is its ability to distinguish truly normal tissue types from cancer adjacent normal tissue. Detection of this cancer “field effect” may have implication in defining a tumor’s propensity to metastasize. In addition, a spectral assessment of the immunoscore [5] may be possible. Computational and data mining efforts are presently underway to exploit further applications of SHP.