Development of SERS based methods for early cancer detection

The applicability of SERS in early cancer detection by means of differentiating cancerous tissue from healthy tissues was systematically investigated by employing three different sample preparation strategies. In the first approach, a 5-µm thick cryosectioned tissue specimen was placed on a PDMS layer coated glass slide before adding a volume of 20-µl 16× concentrated AgNPs containing colloidal suspension onto the tissue. Then, the colloidal suspension was dried at the suspended position to achieve optimal distribution of the AgNPs in the droplet area. In the second approach, in situ synthesis of AgNPs in the homogenized tissue sample was aimed. A 10-µl from each of 4.36x10-1 M AgNO3, 3.55x10-2 M HONH2·HCl and 3x10-2 M NaOH was successively added into the 10-µl of homogenized tissue suspension to reduce Ag+ ions into AgNPs in the homogenized tissue. Then, a 2µl of this mixture was placed onto a CaF2 slide and dried before the SERS measurements. The first approach was optimized through parameters including substrate used to place tissue specimen (PDMS, Al-foil or CaF2), signal collection type (random selection versus mapping), spectral range and acquired-mapping size while the latter approach optimized through the acquisition type and concentrations of AgNO3, HONH2·HCl and NaOH solutions. Then, the data acquired with each sampling method was analyzed with PC-LDA classification models and the coefficients by comparing intra- and inter-method reproducibility, accuracy, and SERS performance indicators of spectral richness and SNR. Finally, the data acquired from these two methods were compared with our previously developed approach of Crashed-liquefied consisting of mixing a 5-µl 32× concentrated colloidal AgNP suspension with homogenized tissue sample by utilizing human thyroid biopsies (n=64). The results showed that in situ approach indicated a higher classification accuracy compared to other approaches for malignant vs. healthy tissue and benign vs. healthy tissue diagnostic combinations by using full spectra region of spectra while the best classification performance for benign vs. malignant tumors was obtained by using Cryosectioned-PDMS and Crashed-liquefied approaches. The findings of this study clearly indicate that SERS is a suitable technique to employ in cancer diagnosis.