Association Between Progranulin (PGRN) Levels in Serum and Cerebrospinal Fluid with Integrated Clinical Indices in Patients with Idiopathic Normal Pressure Hydrocephalus

Background/Objectives: Idiopathic normal pressure hydrocephalus (iNPH) is a potentially treatable syndrome, but biologically informative biomarkers remain limited. Progranulin (PGRN) constitutes a pleiotropic growth factor involved in neuroinflammation, lysosomal function, and tissue repair, which has not been adequately studied in iNPH. The purpose of this study was to examine the serum and cerebrospinal fluid (CSF) levels of PGRN in corresponding patients with suspected iNPH and its correlation with integrated clinical, functional, and neuroradiological parameters. Methods: Thirteen patients with probable iNPH underwent an evaluation protocol, including clinical assessment, neuroradiological evaluation, Tap-test with concomitant gait analysis, and paired serum/CSF sampling. PGRN concentrations in biofluids were measured by ELISA. Correlation analyses were performed. Composite Tap-test response variable derived from quantitative gait-improvement indices was modeled using ridge-logistic regression with leave-one-out cross-validation. Results: In the between-group analyses, serum and CSF concentrations of PGRN were not correlated (r = −0.10, p = 0.74), suggesting that peripheral and intrathecal PGRN behave as non-redundant, compartment-specific readouts rather than as interchangeable measures of the same biological process. Higher CSF concentration of PGRN was nominally associated with older age (r = 0.69, p = 0.009) and with poorer turning-time improvement after the Tap-test (r = −0.62, p = 0.025), while serum concentration of PGRN showed no meaningful associations with clinical or neuroradiological variables. In the model of logistic regression, inclusion of CSF concentration of PGRN substantially improved discrimination of Tap-test response. The full ridge-logistic regression model, including serum and CSF concentration of PGRN, symptom duration, and Kiefer score, achieved an accuracy of 0.923 and an AUC of 0.881. The CSF concentration of PGRN coefficient remained consistently negative across bootstrap resamples (penalized OR 0.434; 95% CI: 0.354–0.697), indicating that higher baseline CSF concentration of PGRN was associated with a lower probability of significant short-term Tap-test response, whereas serum PGRN contributed negligibly to the model. Conclusions: The observed changes in PGRN in CSF may reflect compartment-specific intrathecal inflammatory or tissue-stress processes and may help identify patients with lower short-term responsiveness to CSF drainage. These findings support further longitudinal evaluation of CSF concentration of PGRN for biological stratification and prognostic refinement in iNPH.