BEST IN PHYSICS (JOINT IMAGING-THERAPY)- Application of Raman Spectroscopy for Diagnosis of Adult and Pediatric Cancers
S Devpura1*, R Naik2, J Thakur2, V Naik3, S Sethi4, F Sarkar4, W Sakr4, J Poulik5, R Rabah6, M Klein5, K Barton1, F Siddiqui1, I Chetty1, (1) Henry Ford Health System, Detroit, MI, (2) Wayne State University, Detroit, MI, (3) University of Michigan-Dearborn, Dearborn, MI, (4) Karmanos Cancer Institute, Detroit, MI, (5) Childrens Hospital of Michigan, Detroit, MI, (6) University of Michigan, Ann Arbor, MI,TU-A-WAB-3 Tuesday 8:00AM - 9:55AM Room: Wabash Ballroom
Purpose: To investigate the feasibility of Raman spectroscopy for diagnosis of pre-invasive and invasive head and neck squamous cell carcinoma (HNSCC), prostate cancer, and pediatric cancers.
Methods: Human tissues were used for each tumor type and normal tissues corresponding to the respective anatomic site were included for baseline analysis. Raman spectroscopic data was compared against pathological diagnosis, for assessment of accuracy. Formalin fixed paraffin processed (FFPP) tissues were used in HNSCC and prostate cancer investigations, while both frozen and FFPP tissues were used in pediatric research. A total of 2550 Raman spectra were collected; 401 from HNSCC tissues; 1220 from prostate tissues, and 929 from pediatric tissues (normal, neuroblastoma, and ganglioneuroma). Principal component analysis (PCA) and discriminant function analysis (DFA) were carried out to classify the data.
Results: In HNSCC, the Raman bands assignable to L-tryptophan showed an increase while keratin bands showed a reduction relative to corresponding normal tissues signals. PCA/DFA analysis predicted 89% accuracy for invasive SCC relative to pathological diagnosis. Accuracy for predicting in carcinoma-in-situ and differentiating normal tongue tissues was 91%. For prostate cancer, distinguishable Raman spectral changes were noted for each of the following pathological states: benign epithelia, prostatic intraepithelial neoplasia, prostate cancer, and the stroma surrounding these stages. In pediatric cancers, β-carotene, fat, and cholesterol are key biochemicals differentiating normal tissues from neuroblastoma and ganglioneuroma and, although these chemicals were impacted by the paraffin-fixing process, Raman spectra were able to classify each category with greater than 90% accuracy in the FFPP tissues.
Conclusion: These initial results demonstrate the promise for Raman spectroscopy in the diagnosis and staging of cancer at various stages of progression. We are currently investigating the potential for the use of Raman spectroscopy in the assessment of tumor and normal tissue response following radiation therapy, using animal models and human tissues.