Liquid biopsy (LB) has gained attention as a valuable approach for cancer diagnostics, providing a minimally invasive option compared to conventional tissue biopsies and helping to overcome issues related to patient discomfort and procedural invasiveness. Recent advances in biosensor technologies, particularly photonic sensors, have improved the accuracy, speed, and real-time capabilities for detecting circulating biomarkers in biological fluids. Incorporating these tools into clinical practice facilitates more informed therapeutic choices and contributes to tailoring treatments to individual patient profiles. This review highlights the clinical potential of LB, examines technological limitations, and outlines future research directions. Departing from traditional biosensor focused reviews, it adopts a reverse-mapping approach grounded in clinically relevant tumor biomarkers. Specifically, biomarkers associated with prevalent cancers, such as breast, prostate, and lung cancers, serve as the starting point for identifying the most suitable photonic sensing platforms. The analysis underscores the need to align sensor design with the physicochemical properties of each biomarker and the operational requirements of the application. No photonic platform is universally optimal; rather, each exhibits specific strengths depending on performance metrics such as sensitivity, limit of detection, and easy system integration. Within this framework, the review provides a comprehensive assessment of emerging photonic biosensors and outlines key priorities to support their effective clinical translation in cancer diagnostics.
In this work, we present a numerical proof-of-concept study of a device for nanoparticle sorting, targeting size ranges relevant to exosome-like dimensions (typically 40–200 nm), which remains challenging for passive sorting techniques. The system consists of three silicon waveguides embedded in a CYTOP layer and arranged in a two-step directional-coupler configuration, integrated with a microchannel that carries a water-based buffer as the carrier fluid, transporting the suspended nanoparticles. Three-dimensional Finite Element Method (3D-FEM) simulations were performed, incorporating both optical and hydrodynamic forces to track particle dynamics within the microchannel and demonstrate controlled, size-selective particle deflection. First, numerical simulations show that nanospheres with diameters ranging from 500 nm to 700 nm can be effectively separated by the transverse trapping force at a 4:1 power-splitting ratio. Then, to extend the concept toward smaller size ranges, a bifurcated microchannel is introduced, enabling fluid-assisted transport in low-optical-field regions and allowing reliable separation of particles with smaller diameters (between 200 nm and 400 nm), accompanied by an 8:1 power-splitting ratio. These results demonstrate, within a numerical framework, the feasibility of an integrated photonic–microfluidic approach for size-selective nanoparticle sorting. The proposed strategy may support future pre-processing steps in liquid biopsy workflows, particularly for enriching nanoscale components such as exosome-sized vesicles, rather than constituting a direct diagnostic tool.