Segregation of ovarian cancer stage exploiting spectral biomarkers derived from blood plasma or serum analysis: ATR‐FTIR spectroscopy coupled with variable selection methods

Abstract

Ovarian cancer is a solid tumor and a leading cause of mortality. Diagnostic tools for the detection of early stage (stage I) ovarian cancer are urgently needed. For this purpose, attenuated total reflection Fourier-transform infrared spectroscopy (ATR-FTIR) coupled with variable selection methods, successive projection algorithm or genetic algorithm (GA) combined with linear discriminant analysis (LDA), were employed to identify spectral biomarkers in blood plasma or serum samples for accurate diagnosis of different stages of ovarian cancer, histological type and segregation based on age. Three spectral datasets (stage I vs. stage II-IV; serous vs. non-serous carcinoma; and, ≤60 years vs. >60 years) were processed: sensitivity and specificity required for real-world diagnosis of ovarian cancer was achieved. Toward segregating stage I vs. stage II-IV, sensitivity and specificity (plasma blood) of 100% was achieved using a GA-LDA model with 33 wavenumbers. For serous vs. non-serous category (plasma blood), the sensitivity and specificity levels, using 29 wavenumbers by GA-LDA, were remarkable (up to 94%). For ≤60 years and >60 years categories (plasma blood), the sensitivity and specificity, using 42 wavenumbers by GA-LDA, gave complete accuracy (100%). For serum samples, sensitivity and specificity results gave relatively high accuracy (up to 91.6% stage I vs. stage II-IV; up to 93.0% serous vs. non-serous; and, up to 96.0% ≤60 years vs. >60 years) using several wavenumbers. These findings justify a prospective population-based assessment of biomarkers signatures using ATR-FTIR spectroscopy as a screening tool for stage of ovarian cancer.

Use of Fourier-transform infrared spectroscopy in the diagnosis of rheumatoid arthritis: a pilot study

“…Biofluids like blood, serum or plasma seems to be good specimen in regard of many protein biomarkers presence [10]. Lima et al proved that ATR-FTIR [with genetic algorithm (GA) combined with linear discriminant analysis (LDA)] may be used in early detection of ovarian cancer and differentiation of disease stages [11, 12]. Moreover, ATR-FTIR spectroscopy show higher classification rate than other (Raman) spectroscopic methods [13].…”

4-Nonylphenol effects on rat testis and sertoli cells determined by spectrochemical techniques coupled with chemometric analysis

“…In the past few years, ATR-FTIR and Raman spectroscopy have been extensively applied in toxicology studies, including in vivo (Chen et al, 2015;Li et al, 2015;Obinaju and Martin, 2016) and in vitro tests (Obinaju et al, 2015;Li et al, 2016;Strong et al, 2016). In addition, their potential for disease diagnosis has also been demonstrated in many publications (Gajjar et al, 2012;Owens et al, 2014;Lima et al, 2015;Paraskevaidi et al, 2017b). ATR-FTIR spectroscopy measures the energy absorbed by functional groups within a sample after exposure to IR radiation and generates a spectrum with peaks related to chemical structure of particular entities, e.g., lipids ∼1740 cm -1 , DNA ∼1080 cm -1 , Amide I and II ∼1650 and 1550 cm -1 , respectively.…”

Spectral classification for diagnosis involving numerous pathologies in a complex clinical setting: A neuro-oncology example

“…Some studies have also compared the use of plasma or serum, with similar results. FTIR or Raman spectroscopy methods with biofluids have been used with a wide range of cancers; from head and neck (75% classification accuracy), bladder cancer (up to 80% accuracy), ovarian and endometrial cancer (96% and 81% respectively) and brain tumours (93% high-grade, 96% low-grade) [8,[10][11][12][13][14][15][16].…”