Publications & Posters

Serum neurofilament light chain levels as biomarker of paclitaxel-induced cognitive impairment in patients with breast cancer: a prospective study

Supportive Care in Cancer | October 2, 2021

Argyriou AA, Karteri S, Bruna J, Mariotto S, Simo M, Velissaris D, Kalofonou F, Cavaletti G, Ferrari S and Kalofonos HP

Support Care Cancer. 2021



To prospectively assess the utility of serum neurofilament light chain (sNfL) levels in identifying the risk to develop chemotherapy-induced cognitive impairment (CICI) in cancer patients. We also examined if sNfL can be identified as an early biomarker of CICI development.


We longitudinally measured sNfL levels in 20 female patients with breast cancer, scheduled to receive the 12 weekly paclitaxel-based regimen. An equal number of age-matched female heathy subjects was incuded as control group. CICI was graded by means of the Montreal Cognitive Assessment scale (MOCA); peripheral neurotoxicity (PN) was graded using the neurosensory Common Criteria for Adverse Events (CTCAE)v5.0, while sNfL levels were quantified using a high-sensitive technique (Quanterix, Simoa) before the administration of chemotherapy (T0), after 3 courses (T1), and at the end of chemotherapy (T2).


Pre-treatment sNfL levels were comparable in patients and controls (p = 0.103). At T2, 5/20 patients (mean age 61.4 ± 5.0 years) developed CICI. These 5 patients also had clinically-significant PN. Patients with and without CICI had comparable sNfL values at T2 (p = 0.1). In addition, at T2, sNfL levels did not correlate significantly with MOCA score in CICI patients (p = 0.604). The difference of sNfL levels between T1 and T0 failed to predict independently the occurrence of CICI at T2.


Our findings do not support the utility of measuring sNfL levels as a biomarker of CICI. Grade 2–3 PN most strongly confounded our outcomes. Considering the small sample size, which might have prevented the results from being extrapolated, further testing in larger studies is warranted.