Background and aims <p>Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system, characterized by inflammation, demyelination, gliosis, and neuronal loss. Cognitive disorders are part of the clinical presentation, as well as in the early stages, with a prevalence rate ranging from 40 to 65% and are strictly associated with poor quality of life. In recent years, advances in the field of molecular biomarkers have contributed to improving the accuracy of cognitive diagnosis and to identifying subjects at risk of developing cognitive disorders. By focusing on longitudinal studies, this review aims to elucidate the temporal dynamics and potential predictive value of CSF and serum biomarkers in identifying cognitive decline and monitoring disease progression in MS.</p> Methods <p>A systematic review was conducted using the PRISMA guidelines. 827 studies were identified by searching PubMed, Scopus, and Web of Science between July and August 2025. After screening, 8 studies fulfilled the eligibility criteria and were included.</p> Results <p>NfL showed heterogeneous results: some studies identified it as a feasible biomarker able to predict cognitive decline, while others did not. GFAP did not show significant correlations. Parvalbumin was longitudinally associated with poorer cognition and greater fatigue. Specific microRNAs (miR-126.3p, miR-9p) were associated with processing speed decline. Higher BDNF levels were linked to cognitive improvement. Other biomarkers (OPN, IL-6, CHI3L1, CXCL13) provided insufficient evidence.</p> Conclusions <p>Findings add interesting contributions to the complex picture of the pathophysiology underpinning cognitive decline in MS. The most promising direction is the shift toward blood-based biomarkers, due to their minimally invasive nature and potential for clinical applicability, which may enable clinicians to anticipate, monitor, and potentially modify the cognitive trajectory of people living with MS.</p>