Microfluidic Nanosensor for Label-Free Multiplexed Detection of Breast Cancer Biomarkers via Surface-Enhanced Reflective FTIR Spectroscopy Using Thin Gold Films and Antibody-Oriented Gold Nanourchin: Feasibility Study.

The simultaneous detection of multiple cancer biomarkers using microfluidic multiplexed immunosensors is gaining significant interest in the field of Point-of-Care diagnostics. This study highlights integrating surface-enhanced infrared Fourier transform (SE-FTIR) with a plasmonic-active nanostructure thin film (PANTF) on a printed circuit board (PCB), housed within a microfluidic device for rapid, non-destructive detection of breast cancer (BC). Detection uses monoclonal antibody (mAb)-functionalized gold nanourchins (GNUs) on dual sensing regions. A total of 12 serum samples (24 data points) were tested for HER-II and CA 15-3. The system demonstrated a SE-FTIR enhancement factor (EF) of ~0.18 × 105 using Rhodamine 6G (R6G). Calibration with HER-II (1-100 ng/mL) and CA 15-3 (10-100 U/mL) showed linear responses (R2 = 0.8 and 0.76, respectively). Measurements of unknowns were performed at 1 µL/min over 68 min, with 43 min for biomarker interaction. SE-FTIR spectra were recorded at active zones and analyzed using SpectraView (SV), a custom Python 3.12-based tool. Data preprocessing included filtering (SciPy's filtfilt) and baseline correction using the Improved Asymmetric Least Squares (IASLS) algorithm (pybaselines.Whittaker). Fourier cross-correlation (FCC) showed stronger signal consistency for HER-II. Partial Least Squares (PLS) regression, a dimensionality reduction technique, enabled clear discrimination between the samples and types, with classification accuracy reaching 1.0. Cancer staging based on these biomarkers yielded an overall accuracy of 0.54, indicating that classification regardless of biomarker type. Further studies involving larger and more diverse sample sets are critical before any definitive conclusions can be drawn.