Our goal is to investigate the serum neurofilament light chain (NfLs) dynamics related to the presence of ICANS and its severity.

A nested, matched case-control study within a prospective cohort of all anti-CD19 CAR-T-treated patients with high-grade B-cell lymphoma, previously treated with other hematologic therapies, comparing those who developed ICANS (cases) with those who did not (controls). Only patients with available blood samples were included. Using single-molecule array assay, we assessed NfLs levels (normalized Z-scores) at baseline, day 7, and day 14 post CAR-T administration. Comparative risk analyses, linear mixed analysis with a split-plot design, and ROC analysis for optimal NfLs cut-off to differentiate ICANs presence and grade≥2 were performed.

From a cohort of 159 patients treated with anti-CD19 CAR-T cell therapy, 54 patients (34%) presented ICANS. We identified 32 cases who met the inclusion criteria, and we included 22 age- and sex-matched patients without ICANS who also met the inclusion criteria and had successive biological samples available. The main risk factors for ICANS were Axi-cel use (OR 6.6;95% CI 2.0-21.7) and CRS occurrence (OR 4.8;95% CI 1.1-20.8). Baseline NfL concentrations were elevated in cases and controls vs healthy population (p=0.025, p=0.032, respectively; Figure 1). In the linear mixed analysis, day 7 NfLs were significantly higher in ICANS grade≥2 (means Z-score 2.19 and 1.60, respectively;p=0.022), with no differences at baseline, day 14 nor depending on the presence of ICANS. Patients with day 7 NfLs above Z-score>2.144 had more frequent ICANS grade≥2 (73.3%vs.29.7%, p=0.004).

Patients with ICANS of higher grade (2 or more) exhibit an increase in NfLs compared to milder cases or patients without ICANS, being a Z-score ≥ 2.144 the best cut-off value to differentiate them.