Development and Validation of High-Sensitivity Analytical Frameworks for Low-Abundance Pathological Cargo in Human Biofluids.dissertationsPollock, Jennifer MargaretcreatorHurt, RobertReaderKreiling, JillReaderTripathi, AnubhavAdvisorDesai, TejalReaderBrown University. Biology and Medicine: Biomedical Engineeringsponsor2026xxvii, 189 p.born digitalThesis (Ph. D.)--Brown University, 2026thesesLiquid biopsies offer immense potential for non-invasive disease monitoring, yet their clinical adoption is hindered by the analytical difficulty of detecting scarce biomarkers within complex biofluids. While extracellular vesicles (EVs) and proteopathic aggregates like amyloid-beta (Aβ) provide a window into systemic and central nervous system (CNS) pathologies, traditional isolation-based workflows suffer from low recovery and poor reproducibility. This dissertation describes the development of high-sensitivity analytical frameworks designed to bypass these hurdles through direct, digital quantification of pathological cargo. The first phase of this research addresses ultra-low-abundance protein monomers. By developing a dual-functionalized microparticle interface—utilizing magnetic capture beads and DNA-tagged detection beads coupled with qPCR amplification—we achieved a femtomolar limit of detection (0.05 pg/mL) for Aβ (1-42). This provides a highly sensitive tool for the early detection of Alzheimer’s pathology. Building on this, we established a robust digital Single Molecule Array (SiMoA) platform for the direct detection of intact EVs in human plasma and serum. By implementing a continuous piecewise mathematical model for Average Enzymes per Bead (AEB) determination, we successfully addressed EV polyvalency. This framework maintained a dynamic range over four orders of magnitude and achieved a limit of detection of 1.61 x 107 particles/mL, outperforming commercial assays in both reproducibility and sensitivity. Finally, we extended this SiMoA framework to neuro-diagnostics to quantify brain-derived EVs (bdEVs). By targeting CNS-specific markers, our findings identified ATP1A3 and NCAM as superior, stable alternatives to L1CAM for longitudinal monitoring. Proteomic characterization revealed that NCAM+ populations exhibit a significantly higher loading density for neurodegenerative biomarkers, such as NfL and GFAP, compared to canonical populations. This suggests that subtype-specific targeting can enhance the resolution of CNS signals in peripheral biofluids. Collectively, this research provides a scalable, high-throughput toolbox that shifts the liquid biopsy paradigm from tedious sample preparation to direct, high-precision molecular profiling. These frameworks offer a standardized path forward for early disease detection and treatment monitoring across oncology and neurology.Oncologyextracellular vesiclesNeurologyImmunoassayEnglishRPB20260516