Rapid diagnosis of cerebrospinal fluid (CSF) leaks is critical as endoscopic endonasal skull base surgery gains global prominence. Current clinical methods such as endoscopic examination with and without intrathecal injection of fluorescent dye are invasive and rely on subjective judgment by physicians, highlighting the clinical need for label-free point-of-care (POC). However, a viable solution remains undeveloped due to the molecular complexity of CSF rhinorrhea mixed with nasal discharge and the scarcity of specific biomarkers, delaying sensor development. In this study, we accelerated the development of a label-free CSF detection method for clinical use using a nanoparticle corona. We engineered corona nanointerfaces on near-infrared (nIR) fluorescent single-walled carbon nanotubes (SWCNTs) through noncovalent functionalization with 12 custom-designed poly(ethylene glycol) (PEG) lipids. By high-throughput screening of the corona library for the CSF biomarker β-trace protein (βTP), we selected the optimal corona, achieving a limit of detection (LOD) down to 1.46 mg/L, maintaining its selectivity even in human nasal discharge. Using molecular dynamics and docking simulations, we characterized the 3D morphology and βTP binding energy of the optimal corona in a quantified way. The corona nanosensor accurately diagnosed CSF leakages from eight patients having lumbar drainage and one patient with CSF leakage due to diverse diseases without any sample preparations. By integrating the nanosensor with custom-designed in vivo and in vitro form factors such as a camera and endoscope, we showed its potential for versatile and practical use in clinical settings. This accelerated sensor development platform can meet future urgent clinical demands for various diseases and conditions.
Keywords: SWCNT; beta-trace protein; cerebrospinal fluid; corona; molecular recognition; nanoparticle; nasal discharge.