Development of an ligase chain reaction–fluorescence–SERS multimodal biosensing workflow for ultrasensitive detection and single-base discrimination of KRAS ctDNA

A dual-modality biosensing platform integrating ligase chain reaction (LCR), fluorescence monitoring, and surface-enhanced Raman scattering (SERS) was developed. The LCR step ensures high-fidelity enzymatic amplification, enabling precise discrimination of single-base mismatches at the KRAS locus. Real-time fluorescence using 20 nm gold magnetic nanoparticles provides rapid quantitative evaluation, while SERS analysis on nanostructured OnSpec-Lite bimetallic chips offers molecular-fingerprint confirmation. Under optimized conditions (30 cycles, 60 °C), the biosensor achieved a calculated limit of detection (LOD) of 1.03 fM for SERS and 11.94 fM for fluorescence detection, with corresponding limits of quantification (LOQ) of 3.12 fM and 36.17 fM, respectively. Derived from the linear regression of the SERS calibration curve (y = 73.77x + 871.1; R2 = 0.9617), a linear concentration range of 1 pM to 1000 nM was obtained. Meanwhile, the fluorescence calibration yielded a linear range of 100 pM to 1000 nM (y = 0.069x + 4.848; R2 = 0.8705). Experimentally, the lowest validated ctDNA concentrations detected were 1 pM for SERS and 100 pM for fluorescence. Both detection systems demonstrated high reproducibility, with relative standard deviations (RSDs) below 10%. Furthermore, the platforms maintained analytical robustness in 1/10 and 1/5 serum dilutions, successfully mitigating matrix interference while preserving distinctive Raman signatures. This study introduces a proof-of-concept analytical framework that bridges the high specificity of enzymatic ligation with the ultra-sensitive transduction of nanoplasmonic detection. By combining real-time optical quantification with confirmatory vibrational spectroscopy, the system minimizes false positives and provides cross-validated results essential for high-stakes diagnostics.