Methods of optical science and engineering have been developed for cancer detection and diagnosis and more recently to assess response to therapy in a variety of tissue sites for applications in both pre-clinical and clinical studies [19
]. The interaction of light with complex media such as biological tissues, is characterized by processes that depend on the physical nature of the light and the specific tissue morphology and composition [28
]. The incident light can be scattered (elastically or inelastically) multiple times due to microscopic differences in the index of refraction of cells and subcellular organelles within the tissues, and may be non-radiatively absorbed by chromophores present in the medium or by fluorophores, which release their excess energy by radiative decay, producing fluorescence. The remitted fluorescent light can, in turn, be multiply scattered or absorbed. Although complex, these optical responses can be measured by a variety of spectroscopic techniques and processed through rigorous computational or theoretical models to obtain quantitative biochemical and morphological information about the tissues [30
In optical spectroscopy, the wavelengths of illumination span the ultraviolet (UV) through the near-infrared (NIR) wavelengths. In steady-state reflectance spectroscopy, a broadband light source is used for illumination and a spectrum of the reflected light is collected [33
], while in steady-state fluorescence spectroscopy a narrow spectral-band of incident light (obtained via filtering a broadband source or from a narrowband laser) is used to excite fluorophores and the emerging fluorescence spectrum at each excitation wavelength is detected [34
The diffuse reflectance spectrum is a function of the optical absorption and scattering coefficient spectra [28
]. The absorption and scattering coefficients are wavelength dependent and their value at each wavelength reflects the probability that a photon (of given wavelength) will be absorbed or scattered by the tissue when it traverses an infinitesimal step within the medium. The shape and magnitude of the absorption coefficient depends on the extinction coefficient and concentrations, respectively of dominant tissue chromophores which include oxygenated hemoglobin (HbO2
) and deoxygenated hemoglobin (Hb), beta carotene, water and lipids in the UV-NIR spectrum [35
]. Since diffuse reflectance spectroscopy can measure both HbO2
and Hb one can estimate both the total hemoglobin concentration (THb = HbO2
+ Hb) and the oxygenation saturation in tumors (SO2
/THb). The optical scattering coefficient is known to be sensitive to the spatial architecture and organization of the tissue and therefore can be used as a means to quantify cellular morphology and structure [37
]. The shape and magnitude of the intrinsic fluorescence spectrum depends on the concentrations of the tissue fluorophores, which include several important biochemical molecules such as nicotinamide adenine dinucletide (NADH) and flavin adenine dinucleotide (FAD) or structural proteins such as collagen, elastin and keratin [40
]. The ratio of NADH to FAD is called the optical reduction-oxidation (redox) ratio and can provide information about the reduction-oxidation state in the electron transport chain within the mitochondria. The optical redox ratio has been shown to increase with a decrease in cellular oxygenation [43
]. Measurement of endogenous fluorescence could also provide a means to sense changes in the extracellular matrix composition. However, the measured fluorescence spectrum can be significantly distorted by tissue absorption and scattering and needs to be corrected to get the intrinsic, turbidity-free fluorescence spectrum which can then be used to quantify either absolute or relative tissue fluorophore contributions [34
The sensing depth of light varies from several millimeters in the UV-visible spectrum to several centimeters in the NIR region [45
]. In the UV-visible region, tissues are absorption dominant, which restricts the penetration depth. With increasing wavelength, the overall absorption coefficient decreases and the ratio of scattering to absorption coefficients increase. Thus, in the red and NIR wavelengths tissues are more transparent and photons can migrate through several centimeters of tissue which allows NIR spectroscopy to interrogate sub-surface solid tumors such as those in the breast and neck nodes. UV-visible spectroscopy complementarily has a superficial sensing depth and can directly interrogate pre-cancerous growth and primary invasive carcinomas in the head and neck, anus, cervix, and recurrent chest wall disease in breast cancer. Optical spectroscopic probes can also be guided through endoscopes and biopsy needles to access tumors within body cavities as in breast cancers. This technology is also well-suited for drug discovery in pre-clinical tumors in rodent models.
shows the primary sources of optical contrast in the UV-visible-NIR range. As indicated in the table, not all optical sources of contrast are probed by all spectral regions of light and clearly the total numbers of intrinsic biomarkers that can be interrogated optically increases in the UV-visible spectral range.
Sources of optical contrast and features they correspond to in tissues
The benefits of optical spectroscopy is that it is (1) quantitative, (2) fast, (3) can be used at the bedside and (4) has exquisite sensitivity to intrinsic biomarkers already present in the tissue. The “optical biomarkers” can be measured more frequently than conventional imaging methods such as contrast-enhanced PET, CT and MRI. The synthesis of some of these contrast agents is expensive and requires specialized facilities (for example, cyclotron for PET). Further, multiple biomarkers can be measured with light. With traditional imaging approaches, patients would need to be imaged by several different scanners to fully capture biomarkers of hypoxia and angiogenesis. Because of the frequency with which optical biomarkers can be measured, these technologies could conceivably be used to identify optimized temporal windows of opportunity for when more sophisticated functional imaging techniques could be used to get complete tumor coverage. Although the optical biomarker technology does not provide the tumor coverage that CT, PET and MRI provide, it yields data from tissue sensing depths that are “on par” with that evaluated via IHC, which is the current gold standard.