Comparative analysis of the sensitivity of two surface plasmon resonance (SPR) biosensors was conducted on a single-metallic Au sensor and bimetallic Ag–Au sensor, using a cathepsin S sensor as an example. Numerically modeled resonance curves of Au and Ag–Au layers, with parameters verified by the results of experimental reflectance measurement of real-life systems, were used for the analysis of these sensors. Mutual relationships were determined between ∂Y/∂n components of sensitivity of the Y signal in the SPR measurement to change the refractive index n of the near-surface sensing layer and ∂n/∂c sensitivity of refractive index n to change the analyte’s concentration, c, for both types of sensors. Obtained results were related to experimentally determined calibration curves of both sensors. A characteristic feature arising from the comparison of calibration curves is the similar level of Au and Ag–Au biosensors’ sensitivity in the linear range, where the signal of the AgAu sensor is at a level several times greater. It was shown that the influence of sensing surface morphology on the ∂n/∂c sensitivity component had to be incorporated to explain the features of calibration curves of sensors. The shape of the sensory surface relief was proposed to increase the sensor sensitivity at low analyte concentrations.

The array SPR imaging (SPRi) technique is well suited to the determination of biomarkers in body fluids, called liquid biopsy. No signal enhancement or analyte preconcentration is required. With the aim of achieving signal enhancement and lowering the cost of a single determination, the replacement of gold-covered chips by silver–gold chips was investigated. The aim of this work was to investigate the analytical characteristics of a biosensor formed on a Ag/Au chip and to compare them with those of a biosensor formed on a gold chip. A biosensor for the determination of cathepsin S (Cath S) was chosen as an example. The biosensor consisted of the linker cysteamine and an immobilized rat monoclonal antibody specific for cathepsin S. Both biosensors exhibited a Langmuirian response to Cath S concentration, with linear response ranging from LOQ to 1.5 ng mL⁻¹. The LOQ is 0.1 ng mL⁻¹ for the biosensor formed on the Ag/Au chip, and 0.22 ng mL⁻¹ for that formed on the gold chip. Recoveries and precision for medium and high Cath S concentrations were acceptable for both biosensors, i.e., precision better than 10% and recoveries within the range 102–105%. However, the results for the lowest Cath S concentration were better for the biosensor formed on the Ag/Au chip (9.4 and 106% for precision and recovery, respectively). Generally, no significant differences in analytical characteristics were observed between the Ag/Au and Au chips. The two biosensors were also compared in the determination of Cath S in real samples. Nine plasma samples from healthy donors and nine from patients with ovarian cancer were analyzed for Cath S concentration with the biosensors formed on Ag/Au and Au chips. The results obtained with the two biosensors were very similar and show no significant differences on the Bland–Altman plot. The Cath S concentration in the blood plasma of ovarian cancer patients was elevated by one order of magnitude as compared with the control (12.6 ± 3.6 vs. 1.6 ± 1.2 ng mL⁻¹).

Non-fluidic array SPR imaging (SPRi) with appropriate biosensors is a new tool for the determination of various biomarkers in body fluids. Numerous biomarkers can be determined without signal enhancement or preliminarily preconcentration. The introduction of a new material solution of the chip may increase the scope of the application of this technique. Solutions with adhesive separating foil and an Ag/Au chip were compared with the previously used two-paint separating polymer and pure gold chip. These solutions were tested using the example of a biosensor for cathepsin D (Cath D), which consisted of pepstatin A (a Cath D inhibitor) immobilized via a cysteamine linker using the NHS/EDC protocol. Four material versions of the Cath D biosensor proved adequate in terms of range of linearity, LOQ, precision and recovery. All four versions of the biosensor were used for the determination of Cath D in the blood serum patients with glioblastoma and control samples, producing very similar results and showing an elevated biomarker concentration in the case of cancer. Therefore, the problem of determining the correct level of Cath D in the serum of healthy individuals has been resolved, correcting literature data which ranged over three orders of magnitude.