To the Editor, We read with great interest the prospective multicenter study by Auriti et al. (published online in Clinical Chemistry and Laboratory Medicine, https://doi.org/10.1515/cclm2025-0128) evaluating the diagnostic accuracy of presepsin (P-SEP) for late-onset sepsis (LOS) in critically ill neonates and defining its optimal cut-off value. This work directly addresses a critical clinical gap: neonatal LOS diagnosis is often confounded by non-infectious inflammatory states (e.g., perinatal asphyxia, surgical trauma), and traditional biomarkers like C-reactive protein (CRP) and procalcitonin (PCT) have limitations in specificity and susceptibility to non-infectious interference. The authors’ rigorous design – including standardized P-SEP measurement via the PATHFAST™ point-of-care system, longitudinal monitoring of P-SEP kinetics (T0 to T3), and clear stratification of sepsis into confirmed and suspected cases – provides robust evidence for clinical application [1]. Notably, the finding that baseline P-SEP values (median 518.0 ng/L) are unaffected by underlying non-infectious pathologies (e.g., congenital heart disease, surgical anomalies) is particularly valuable, as it validates P-SEP’s potential as an infection-specific marker in this vulnerable population. We greatly appreciate the authors’ efforts to advance evidence-based neonatal sepsis diagnosis [1].