Direct Flow Cytometric Assessment of Cerebrospinal Fluid CAR-T Cells for ICANS Diagnosis

Direct Flow Cytometric Assessment of Cerebrospinal Fluid CAR-T Cells for ICANS Diagnosis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Direct Flow Cytometric Assessment of Cerebrospinal Fluid CAR-T Cells for ICANS Diagnosis Koji Kato, Ken Takigawa, Kohta Miyawaki, Shiho Taniguchi, Masatoshi Shimo, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9396828/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Diagnosing immune effector cell-associated neurotoxicity syndrome (ICANS) following chimeric antigen receptor (CAR)-T cell therapy remains challenging due to the lack of objective, specific biomarkers. We analyzed cerebrospinal fluid (CSF) samples collected at neurological symptom onset from 24 patients following CD19-targeted CAR-T therapy including 18 with ICANS requiring steroid treatment and 6 without ICANS who had alternative neurological diagnoses or self-limiting symptoms. Multicolor flow cytometry identified CAR-T cells as CD3⁺ FMC63⁺ lymphocytes. The proportion of CAR-T cells within CD3⁺ T cells was significantly higher in ICANS patients than in non-ICANS patients (median 32.3% vs. 7.9%, p = 0.018), with a predominance of CD4⁺ T cells. Receiver operating characteristic analysis identified a diagnostic cutoff of ≥20%, achieving a sensitivity of 0.72 and a specificity of 0.83 (AUC 0.80). Combining this CSF parameter with peripheral blood markers such as fibrinogen or CRS grade further improved diagnostic accuracy, suggesting that CSF-derived parameters reflecting the local immune environment at the site of pathology complement systemic markers in the diagnostic evaluation of ICANS. These findings indicate that flow cytometric assessment of CAR-T cells in CSF, which uniquely allows proportional quantification within T-cell subsets, provides an objective diagnostic tool for ICANS requiring therapeutic intervention. Health sciences/Medical research/Translational research Biological sciences/Cancer/Cancer therapy/Cancer immunotherapy Health sciences/Health care/Diagnosis Biological sciences/Immunology/Immunotherapy Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Chimeric antigen receptor (CAR)-T cell therapy has revolutionized the treatment of relapsed or refractory hematologic malignancies, achieving high remission rates in patients with otherwise limited options [ 1 – 4 ]. However, as CAR-T cell therapy becomes more widely adopted, immune effector cell-associated neurotoxicity syndrome (ICANS) has emerged as a critical management challenge, significantly impacting treatment outcomes and patient safety. ICANS occurs in 20% to 70% of patients receiving CAR-T cell therapy and presents with various neurological symptoms. Mild cases present with reversible symptoms, such as confusion, delirium, and aphasia, while severe cases may progress to seizures, coma, and fatal cerebral edema [ 1 – 4 ]. The unpredictable nature of ICANS and its potentially life-threatening outcomes underscore the clinical necessity of accurate, early diagnosis and reliable severity assessment to optimize therapeutic interventions and improve patient outcomes. Current diagnostic and monitoring approaches for ICANS include the Immune Effector Cell-Associated Encephalopathy (ICE) Score, electroencephalography (EEG), neuroimaging, and cerebrospinal fluid (CSF) analysis. The ICE score, while widely used as a standardized grading tool, relies on bedside cognitive assessment and is inherently subject to inter-rater variability and situational factors such as patient sedation. EEG and neuroimaging findings are nonspecific and not pathognomonic for ICANS [ 5 – 8 ]. CSF analyses, such as total cell count and protein concentration, are insufficient to differentiate ICANS from other neurological complications. Consequently, the diagnosis of ICANS currently requires exclusion of other neurological etiologies, highlighting the unmet need for objective, specific biomarkers. Although the precise pathophysiology of ICANS remains incompletely understood, accumulating evidence suggests that blood-brain barrier (BBB) disruption, immune cell trafficking into the central nervous system (CNS), and subsequent neuroinflammation mediated by cytokines play a central role. Most biomarker research to date has focused on peripheral blood (PB) markers, including serum cytokine levels and CAR-T cell kinetics [ 2 , 3 , 9 , 10 ]; however, these systemic markers may not adequately capture the immune processes within the CNS. CSF is obtained directly from the CNS compartment and may better reflect the local immune milieu at the site of pathology. This biological proximity makes CSF an attractive source for biomarkers with greater relevance to real-world diagnostic decision-making. This study, therefore, aimed to address this diagnostic gap by quantifying CAR-T cells in CSF. Prior studies have detected CAR-T cell DNA in CSF using PCR-based methods, but the quantity of transgene did not correlate with neurotoxicity severity [ 3 , 11 , 12 ], and these approaches cannot characterize the phenotypic composition of infiltrating T cells. We therefore hypothesized that multicolor flow cytometry, by enabling direct quantification of CAR-T cells and their subsets within CSF, could provide a more diagnostically informative assessment of ICANS. Subjects and Methods CSF Sample Collection From June 2021 to October 2025, CSF samples were collected from patients undergoing CAR-T cell therapy at Kyushu University Hospital when ICANS was suspected, as part of diagnostic evaluation. A lumbar puncture was performed at the onset of neurological symptoms in accordance with institutional clinical practice. The study was approved by the Institutional Review Board of Kyushu University Hospital (approval number: 22062) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients. Clinical and Laboratory Data Clinical parameters collected included age, sex, disease status prior to CAR-T cell infusion, type of CAR-T cell product administered, and grading of cytokine release syndrome (CRS) and ICANS according to the American Society for Transplantation and Cellular Therapy (ASTCT) criteria [ 5 ]. PB parameters were collected at the time of clinical suspicion of ICANS, including platelet count, lactate dehydrogenase (LDH), creatinine, C-reactive protein (CRP), ferritin, and fibrinogen. Endothelial activation was assessed using the Endothelial Activation and Stress Index (EASIX) and its modified forms [ 13 , 14 ]: EASIX = (LDH [U/L] × creatinine [mg/dL]) / platelet count [10⁹/L] modified-EASIX = (CRP [mg/dL] ×LDH [U/L]) / platelet count [10⁹/L] simplified-EASIX = LDH [U/L] / platelet count [10⁹/L] Electroencephalography Electroencephalography (EEG) was performed using the International 10–20 system with bipolar longitudinal montages. Given the clinical severity of patients, post-CAR-T EEG evaluations were primarily performed at the bedside at suspected ICANS onset. Findings were categorized as normal, mildly abnormal, or moderately abnormal by a board-certified neurophysiologist. The presence or absence of frontal intermittent rhythmic delta activity (FIRDA) was evaluated [ 15 ]. Flow Cytometry Analysis CSF samples were processed immediately after collection. After centrifugation at 500 × g for 5 minutes, the pellet was resuspended in staining medium consisting of HBSS (−) supplemented with 2% heat-inactivated fetal bovine serum and 4 mM EDTA. Cells were stained with the following fluorochrome-conjugated monoclonal antibodies: PE-conjugated FMC63 (REA1297, Miltenyi Biotec, Bergisch Gladbach, Germany), BV510-conjugated CD3 (UCHT1, BioLegend, San Diego, USA), BV785-conjugated CD4 (SK3, BD Biosciences San Diego, USA), BV570-conjugated CD8 (RPA-T8, BioLegend), BV605-conjugated CD19 (HIB19, BioLegend), and PerCP/Cy5.5-conjugated CD14 (63D3, BioLegend), CD33 (WM53, BioLegend), CD235ab (HIR2, BioLegend). After staining, samples were incubated for 1 hour at 4℃ in the dark with gentle agitation. Subsequently, staining medium (999 µL) and propidium iodide (final concentration: 1 µg/mL) were added for dead cell exclusion, and cells were washed by centrifugation at 500 × g for 5 minutes and resuspended in 200 µL of staining medium. Multicolor flow cytometry was performed on a FACSAria IIIu (BD Biosciences, San Jose, CA, USA) or MA900 (Sony Biotechnology Inc., San Jose, CA, USA). CAR-T cells were identified as CD3⁺ FMC63⁺ cells within the lymphocyte gate, after exclusion of dead cells, monocytes, granulocytes, and erythrocytes. Due to institutional equipment transition, flow cytometry was performed on a FACSAria IIIu for samples collected before June 2023 and on an MA900 thereafter. Instrument settings were optimized to ensure comparable results, and consistent gating strategies were applied across both platforms. Statistical Analysis The Wilcoxon rank-sum test was used to compare continuous or ordinal variables between groups, and Pearson’s chi-squared test was employed for nominal variables. Correlations between continuous variables were assessed using Spearman’s rank correlation coefficient. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic accuracy of CSF CAR-T cell proportions, with the optimal cutoff value determined by the Youden index. All statistical analyses were performed using JMP version 19. A two-sided p-value < 0.05 was considered statistically significant. Results Patient Characteristics Twenty-four patients who underwent CSF sampling at the time of clinical suspicion of ICANS were included in the analysis. Patients were categorized into two groups: ICANS (n = 18) and non-ICANS (n = 6) (Table 1). The non-ICANS group consisted of patients with an alternative confirmed cause of neurological symptoms, such as CNS infiltration (n = 2), or patients whose symptoms resolved spontaneously without dexamethasone treatment (e.g., headache, aphasia, wrist dorsiflexion weakness, or transient expressive difficulty; each in one patient). The median age of the cohort was 65 years (range, 42–73), and the majority were male (n = 18, 75%). The most common primary disease was DLBCL in 19 patients (79%), followed by high-grade B-cell lymphoma in 3 (13%) and follicular lymphoma in 2 (8%). The CAR-T cell products administered were tisagenlecleucel (tisa-cel, n = 9), lisocabtagene maraleucel (liso-cel, n = 9), and axicabtagene ciloleucel (axi-cel, n = 6). Five patients were in complete response at the time of infusion. CRS occurred in 22 patients (92%), including 3 (13%) with grade ≥ 3. ICANS developed in 18 patients (75%), including 2 (11%) with grade ≥ 3, with a median onset of 6 days (range, 3–38). Exploration of ICANS-Associated Parameters To identify potential biomarkers for diagnosing ICANS, we explored parameters associated with ICANS onset. Several risk factors for ICANS have been identified, including high tumor burden, elevated inflammatory markers, endothelial activation indices, and CRS severity; however, these parameters have predominantly been evaluated at baseline or at the time of CAR-T cell infusion [ 2 , 3 , 14 , 16 , 17 ]. In this study, we assessed previously reported parameters at the time of neurological symptom onset and additionally evaluated CAR-T cell frequencies in CSF detected by multicolor flow cytometry. Figure 1 illustrates the flow cytometry gating strategy used to detect CAR-T cells in CSF and PB. The diagnostic performance of each parameter was assessed using ROC analysis, and the area under the curve (AUC) with 95% confidence intervals (CIs) is presented in Fig. 2 . Diagnostic value of CAR-T cell frequency in CSF Among PB parameters, fibrinogen and creatinine levels showed high discriminatory ability (AUC 0.86 and 0.84, respectively) (Fig. 2 ). Patients with ICANS showed lower fibrinogen levels (median 232 mg/dL, IQR 182–286) compared with the non-ICANS group (median 359 mg/dL, IQR 280–483; p = 0.005) (Supplementary Table 1). Creatinine levels also demonstrated good discrimination with significantly higher levels in the ICANS group (median 0.85 [0.68–1.05] vs. 0.56 [0.35–0.69] mg/dL; p = 0.007). Other parameters, including platelet count, LDH, CRP, ferritin, and EASIX-related scores, showed modest AUCs that were not statistically significant. Regarding clinical parameters, the overall EEG assessment showed moderate discriminatory ability (AUC 0.76, 95% CI 0.54–0.98), although the distribution did not differ significantly between groups (p = 0.125) (Supplementary Table 1). FIRDA was equally prevalent in both groups (75%), providing no discriminatory value (AUC 0.50). CRS grade also demonstrated moderate discrimination (AUC 0.72, 95% CI 0.56–0.89), with a cutoff of grade ≥ 2 achieving a specificity of 1.00 and sensitivity of 0.56 (Fig. 2 ). None of the routine CSF parameters reached statistical significance (Fig. 2 ). Notably, there was no significant difference in the CSF leukocyte counts between the ICANS and non-ICANS groups (Fig. 3 a), suggesting that total cellularity alone is insufficient for discriminating ICANS. In contrast, flow cytometry-derived CSF CAR-T cell parameters demonstrated superior diagnostic performance with high AUCs (Fig. 2 ). The proportion of CAR-T cells within CD3⁺ T cells in the CSF was significantly higher in patients with ICANS (32.3% [15.3–53.8]) than in those without ICANS (7.9% [2.7–15.0]; p = 0.018) (Fig. 3 b). The CD4/CD8 ratio in the CSF CAR-T cells did not differ significantly between groups. CAR-T cells in CSF versus PB In a subset of 11 patients, PB was collected concurrently with CSF sampling and submitted for multicolor flow cytometry. The correlation between PB and CSF CAR-T cell frequencies showed a moderate but non-significant positive trend (r = 0.482, p = 0.133) (Supplementary Fig. 1), indicating that PB CAR-T cell levels do not reliably predict CSF CAR-T cell infiltration. We further examined the compartmental distribution of CAR-T cells by comparing the CD4/CD8 ratio between PB and CSF. The CD4/CD8 ratio of CAR-T cells was significantly higher in CSF than in PB (p = 0.002), indicating a preferential enrichment of CD4⁺ CAR-T cells in the CNS compartment (Fig. 3 c). This CD4-dominant skewing in CSF was also observed among non-CAR-T cells (p = 0.011) (Fig. 3 c). Diagnostic Performance of CSF CAR-T Cell Frequency ROC analysis for the CSF CD3⁺ CAR-T cell frequency identified an optimal cutoff of ≥ 20%, achieving a sensitivity of 0.72 and specificity of 0.83 (AUC 0.80, 95% CI 0.60–0.99) (Fig. 4 a). To explore whether CSF flow cytometry data could complement existing diagnostic parameters, we examined the effect of combining the CSF CD3⁺ CAR-T cell frequency (≥ 20%) with either a PB parameter (fibrinogen ≤ 244 mg/dL) or a clinical parameter (CRS grade ≥ 2). When CSF data were added to fibrinogen, the AUC improved from 0.86 (95% CI 0.71-1.00; sensitivity 0.72, specificity 1.00) to 0.91 (95% CI 0.77-1.00; sensitivity 0.89, specificity 0.83) (Fig. 4 b). Similarly, the addition of CSF data to CRS grade improved the AUC from 0.72 (95% CI 0.56–0.89; sensitivity 0.556, specificity 1.000) to 0.89 (95% CI 0.74-1.00; sensitivity 0.89, specificity 0.83) (Fig. 4 c). In both cases, incorporating CSF-derived information improved sensitivity while maintaining acceptable specificity, suggesting that CSF flow cytometry data reflecting the local immune environment at the site of pathology may serve as a useful complement to conventional parameters in the diagnostic evaluation of ICANS. Discussion This study demonstrates that the proportion of CAR-T cells within CD3⁺ T cells in CSF correlates with ICANS onset, with CD4⁺ CAR-T cells being predominant. A provisional cutoff value of 20% effectively distinguished ICANS with good sensitivity and specificity. Conventional diagnostic tools such as the ICE score, EEG, and neuroimaging have low specificity, and systemic inflammatory markers such as CRP and ferritin also cannot reliably distinguish ICANS from CRS or other inflammatory conditions. As no objective test for ICANS has been established to date, this approach offers a targeted diagnostic tool that can identify patients with ICANS who require steroid treatment. To our knowledge, this is the first study to systematically evaluate CSF CAR-T cell quantification by flow cytometry as a diagnostic biomarker for ICANS, extending prior technical validation of flow cytometric CAR-T detection in CSF [ 18 ]. Importantly, all CSF samples in this study were obtained as part of routine diagnostic evaluation when ICANS was suspected, rather than through research-specific procedures, underscoring the feasibility of implementing this approach in standard clinical practice. Furthermore, combining the CSF CD3⁺ CAR-T cell frequency (≥ 20%) with either fibrinogen (≤ 244 mg/dL) or CRS grade (≥ 2) improved diagnostic accuracy compared with each parameter alone, suggesting that CSF flow cytometry can complement existing parameters. The lower fibrinogen levels observed in the ICANS group may reflect consumptive coagulopathy associated with endothelial activation [ 2 ] as well as tocilizumab-induced suppression of IL-6-dependent fibrinogen synthesis [ 19 ]. As these systemic markers capture different biological dimensions from CSF-derived parameters reflecting the local immune milieu, a multiparameter approach integrating both may enhance diagnostic precision for ICANS. Previous studies using PCR-based methods have shown that CAR-T cells can be detected in CSF regardless of neurotoxicity status. Santomasso et al. reported that the quantity of CAR-T cell DNA in CSF did not correlate with neurotoxicity severity [ 3 ], and similar findings were reported by Mueller et al. using quantitative PCR [ 11 ] and Berger et al. using digital PCR [ 12 ]. These DNA-based approaches measure transgene copy number and cannot characterize the phenotype or subset composition of CAR-T cells. In contrast, multicolor flow cytometry enables direct identification of CAR-T cells at the protein level and simultaneous assessment of their proportion within T cells. Our findings indicate that it is not the absolute quantity of CAR-T DNA but rather the proportion of CAR-T cells among CD3⁺ T cells in CSF that distinguishes ICANS, as confirmed by comparison with non-ICANS patients with alternative neurological conditions. Although the sample size was limited, we did not observe a clear correlation between CSF CAR-T cell fraction and ICANS severity. Notably, the absolute CSF leukocyte counts were extremely limited in many cases (median 3 cells/µL), which limits the feasibility of PCR-based quantification. In this context, multicolor flow cytometry offers a distinct advantage by enabling direct phenotypic assessment including T-cell subset characterization even in paucicellular specimens. Autopsy studies of fatal ICANS cases have provided the primary basis for understanding its neuropathology. Gust et al. [ 2 ] demonstrated endothelial activation and BBB disruption, and Karschnia et al. [ 20 ] reported perivascular accumulation of predominantly CD8⁺ T cells, although CD4⁻/CD8⁻ double-negative T cells were also present. These autopsy findings from fatal cases have shaped a predominantly CD8-centric interpretation of ICANS. However, these observations were derived from the most severe, fatal cases and may not fully represent the immunological landscape of clinically typical, treatable ICANS. This is consistent with Shah et al., who observed CD4⁺ non-CAR T cell predominance in CSF of two patients with steroid-refractory ICANS by flow cytometry [ 21 ], although systematic evaluation was not performed. An important question is whether this CD4 predominance in CSF reflects a physiological characteristic of the CNS compartment or a process specific to ICANS. The CD4/CD8 ratio was significantly higher in CSF than in PB for both CAR-T and non-CAR-T cells, consistent with the known predominance of central memory CD4⁺ T cells in normal CSF [ 22 ]. This suggests that the CD4 skewing observed in our study is at least partly attributable to the physiological composition of the CNS immune compartment. Nevertheless, several lines of evidence suggest that CD4⁺ T cells may also be functionally relevant to the pathogenesis of ICANS. CD4⁺ T cell-dominant neuroinflammation has been well documented in other CNS inflammatory diseases; for example, in multiple sclerosis, activated memory CD4⁺ T cells migrate across the BBB and drive sustained inflammation within the CNS [ 23 ]. In the specific context of CAR-T therapy, Lu et al. recently demonstrated that the CXCL16-CXCR6 axis selectively recruits CD4⁺-dominant T cells into the CNS during ICANS [ 24 ], providing a mechanistic basis for the preferential enrichment of CD4⁺ CAR-T cells in CSF. Furthermore, Baur et al. reported that greater peripheral expansion of CD4⁺ CAR-T cells correlates with increased severity of both CRS and ICANS [ 25 ], supporting a functional role for CD4⁺ T cells in immune-mediated toxicity. Taken together, these findings suggest that the predominance of CD4⁺ T cells in CSF may reflect both the physiological CNS immune composition and active chemokine-driven recruitment of CD4⁺ CAR-T cells during ICANS. This study has several limitations. First, the sample size was small, and the study was conducted at a single center, which limits the generalizability of the findings and the statistical power to detect associations with ICANS severity. Second, the non-ICANS control group was heterogeneous, comprising patients with CNS infiltration and those with self-limiting neurological symptoms, which may have influenced the discriminative performance of the evaluated parameters. Finally, a more detailed characterization of CD4⁺ T-cell subsets in CSF, in both CAR-T and non-CAR-T cells, is needed to further elucidate the pathogenesis of ICANS. In conclusion, this study demonstrates that quantifying CAR-T cells in CSF by multicolor flow cytometry provides an objective measure for identifying ICANS that requires therapeutic intervention. The proportion of CAR-T cells among CD3⁺ T cells in CSF distinguished ICANS from other neurological conditions, and combining this measure with existing parameters further improved diagnostic accuracy. The observed CD4⁺ predominance among CSF CAR-T cells offers new insight into the immunological landscape of ICANS, complementing the CD8-centric model derived from fatal cases. Multicenter validation of the proposed diagnostic cutoff and detailed characterization of CD4⁺ T-cell subsets in the CNS compartment are warranted. Declarations Acknowledgments We thank the patients and their families for participating in this study. We are grateful to the clinical staff at Kyushu University Hospital for their dedicated care of CAR-T recipients. We thank Arisa Matsuyama and Naoko Ban for their technical assistance with flow cytometric analyses. Author Contributions K.T., K.Miyawaki and K.K designed the study, collected and analyzed data, and wrote the manuscript. S.T., M.S., K.Mori., T.T., T.Sakoda., F.J., T.Y., and T.Shima. contributed to patient care and data collection. K.Miyawaki., Y.K., Y.M., K.A., and K.K. supervised the study and critically reviewed the manuscript. All authors approved the final version. Competing interests Koji Kato; Honoraria: AbbVie, Bristol-Myers Squibb, Chugai, Janssen, Kyowa Kirin, Ono, Gilead Sciences, Novartis; Consulting or Advisory Role: AbbVie, AstraZeneca, Chugai, Daiichi Sankyo, Eisai, Janssen, Bristol-Myers Squibb, Novartis, Gilead Sciences; Research Funding: AbbVie, MSD, Bristol-Myers Squibb, Chugai, Daiichi Sankyo, Eisai, Janssen, Kyowa Kirin, Novartis, Ono, Gilead Sciences. Kohta Miyawaki; Honoraria: Bristol-Myers Squibb, Gilead Sciences, Novartis. The remaining authors declare that they have no conflict of interest. Data availability statement The datasets generated during the current study are available from the corresponding author on reasonable request. References Neelapu SS, Tummala S, Kebriaei P, Wierda W, Gutierrez C, Locke FL, et al. Chimeric antigen receptor T-cell therapy — assessment and management of toxicities. Nat Rev Clin Oncol. 2018;15:47–62. https://doi.org/10.1038/nrclinonc.2017.148 Gust J, Hay KA, Hanafi L-A, Li D, Myerson D, Gonzalez-Cuyar LF, et al. Endothelial Activation and Blood–Brain Barrier Disruption in Neurotoxicity after Adoptive Immunotherapy with CD19 CAR-T Cells. Cancer Discov. 2017;7:1404–19. https://doi.org/10.1158/2159-8290.cd-17-0698 Santomasso BD, Park JH, Salloum D, Rivière I, Flynn J, Mead E, et al. Clinical and Biologic Correlates of Neurotoxicity Associated with CAR T Cell Therapy in Patients with B-cell Acute Lymphoblastic Leukemia (B-ALL). Cancer Discov. 2018;8:CD-17-1319. https://doi.org/10.1158/2159-8290.cd-17-1319 Abramson JS. Anti-CD19 CAR T-Cell Therapy for B-Cell Non-Hodgkin Lymphoma. Transfus Med Rev. 2020;34:29–33. https://doi.org/10.1016/j.tmrv.2019.08.003 Lee DW, Santomasso BD, Locke FL, Ghobadi A, Turtle CJ, Brudno JN, et al. ASTCT Consensus Grading for Cytokine Release Syndrome and Neurologic Toxicity Associated with Immune Effector Cells. Biol Blood Marrow Transplant. 2019;25:625–38. https://doi.org/10.1016/j.bbmt.2018.12.758 Strati P, Nastoupil LJ, Westin J, Fayad LE, Ahmed S, Fowler NH, et al. Clinical and radiologic correlates of neurotoxicity after axicabtagene ciloleucel in large B-cell lymphoma. Blood Adv. 2020;4:3943–51. https://doi.org/10.1182/bloodadvances.2020002228 Hunter BD, Jacobson CA. CAR T-Cell Associated Neurotoxicity: Mechanisms, Clinicopathologic Correlates, and Future Directions. JNCI: J Natl Cancer Inst. 2019;111:646–54. https://doi.org/10.1093/jnci/djz017 Gust J, Taraseviciute A, Turtle CJ. Neurotoxicity Associated with CD19-Targeted CAR-T Cell Therapies. CNS Drugs. 2018;32:1091–101. https://doi.org/10.1007/s40263-018-0582-9 Sterner RC, Sterner RM. Immune effector cell associated neurotoxicity syndrome in chimeric antigen receptor-T cell therapy. Front Immunol. 2022;13:879608. https://doi.org/10.3389/fimmu.2022.879608 Wang X, He X, Zhang T, Liu J, Zhao M. Latest updates on pathogenesis mechanisms and management strategies for cytokine release syndrome, neurotoxicity, and hemophagocytic lymphohistiocytosis related to CAR-T cell therapies. Ann Hematol. 2025;104:3129–51. https://doi.org/10.1007/s00277-025-06467-y Mueller KT, Maude SL, Porter DL, Frey N, Wood P, Han X, et al. Cellular kinetics of CTL019 in relapsed/refractory B-cell acute lymphoblastic leukemia and chronic lymphocytic leukemia. Blood. 2017;130:2317–25. https://doi.org/10.1182/blood-2017-06-786129 Berger SC, Fehse B, Akyüz N, Geffken M, Wolschke C, Janson D, et al. Molecular monitoring of T-cell kinetics and migration in severe neurotoxicity after real-world CD19-specific chimeric antigen receptor T cell therapy. Haematologica. 2022;108:444–56. https://doi.org/10.3324/haematol.2022.281110 Luft T, Benner A, Terzer T, Jodele S, Dandoy CE, Storb R, et al. EASIX and mortality after allogeneic stem cell transplantation. Bone Marrow Transplant. 2020;55:553–61. https://doi.org/10.1038/s41409-019-0703-1 Pennisi M, Sanchez-Escamilla M, Flynn JR, Shouval R, Tomas AA, Silverberg ML, et al. Modified-EASIX predicts severe cytokine release syndrome and neurotoxicity after Chimeric Antigen Receptor (CAR) T cells. Blood Adv. 5:3397–406. https://doi.org/10.1182/bloodadvances.2020003885 Huby S, Gelisse P, Tudesq J-J, Labauge P, Duflos C, Cartron G, et al. Frontal Intermittent Rhythmic Delta Activity Is a Useful Diagnostic Tool of Neurotoxicity After CAR T-Cell Infusion. Neurol Neuroimmunol Neuroinflammation. 2023;10:e200111. https://doi.org/10.1212/nxi.0000000000200111 Grant SJ, Grimshaw AA, Silberstein J, Murdaugh D, Wildes TM, Rosko AE, et al. Clinical Presentation, Risk Factors, and Outcomes of Immune Effector Cell-Associated Neurotoxicity Syndrome Following Chimeric Antigen Receptor T Cell Therapy: A Systematic Review. Transplant Cell Ther. 2022;28:294–302. https://doi.org/10.1016/j.jtct.2022.03.006 Korell F, Penack O, Mattie M, Schreck N, Benner A, Krzykalla J, et al. EASIX and Severe Endothelial Complications After CD19-Directed CAR-T Cell Therapy—A Cohort Study. Front Immunol. 2022;13:877477. https://doi.org/10.3389/fimmu.2022.877477 Johansson U, Gallagher K, Burgoyne V, Maus MV, Casey KS, Brini GG, et al. Detection of CAR‐T19 cells in peripheral blood and cerebrospinal fluid: An assay applicable to routine diagnostic laboratories. Cytom Part B: Clin Cytom. 2021;100:622–31. https://doi.org/10.1002/cyto.b.22005 Perl M, Herfeld K, Harrer DC, Höpting M, Schweiger M, Sterz U, et al. Tocilizumab administration in cytokine release syndrome is associated with hypofibrinogenemia after chimeric antigen receptor T-cell therapy for hematologic malignancies. Haematologica. 2024;109:2969–77. https://doi.org/10.3324/haematol.2023.284564 Karschnia P, Strübing F, Teske N, Blumenberg V, Bücklein VL, Schmidt C, et al. Clinicopathologic Findings in Fatal Neurotoxicity After Adoptive Immunotherapy With CD19‐Directed CAR T‐Cells. HemaSphere. 2021;5:e533. https://doi.org/10.1097/hs9.0000000000000533 Shah NN, Johnson BD, Fenske TS, Raj RV, Hari P. Intrathecal chemotherapy for management of steroid-refractory CAR T-cell–associated neurotoxicity syndrome. Blood Adv. 2020;4:2119–22. https://doi.org/10.1182/bloodadvances.2020001626 Graaf MT de, Smitt PAES, Luitwieler RL, Velzen C van, Broek PDM van den, Kraan J, et al. Central memory CD4+ T cells dominate the normal cerebrospinal fluid. Cytom Part B: Clin Cytom. 2011;80B:43–50. https://doi.org/10.1002/cyto.b.20542 Chitnis T. The Role of CD4 T Cells in the Pathogenesis of Multiple Sclerosis. Int Rev Neurobiol. 2007;79:43–72. https://doi.org/10.1016/s0074-7742(07)79003-7 Lu I-N, Müller-Miny L, Krekeler C, Cheung PF-Y, Antonopoulou G, Jeibmann A, et al. The CXCL16/CXCR6 axis is linked to immune effector cell-associated neurotoxicity in chimeric antigen receptor (CAR) T cell therapy. Genome Med. 2025;17:71. https://doi.org/10.1186/s13073-025-01498-6 Baur K, Buser A, Jeker LT, Khanna N, Läubli H, Heim D, et al. CD4+ CAR T-cell expansion is associated with response and therapy related toxicities in patients with B-cell lymphomas. Bone Marrow Transplant. 2023;58:1048–50. https://doi.org/10.1038/s41409-023-02016-1 Table Table 1 is available in the supplementary files section Additional Declarations Yes Supplementary Files SupplementaryMaterials.pdf Supplementary Materials BMTTable.pdf Cite Share Download PDF Status: Under Review Version 1 posted Reviewer # 1 agreed at journal 21 Apr, 2026 Reviewers invited by journal 15 Apr, 2026 Submission checks completed at journal 13 Apr, 2026 Editor assigned by journal 12 Apr, 2026 First submitted to journal 12 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9396828","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":623368101,"identity":"d0488193-273b-4733-a023-5d8810271246","order_by":0,"name":"Koji Kato","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABXUlEQVRIie2QMUvDQBSAXwnE5TSbXImmf+FCoBYq+lcuBNIlomPBooFAXYKulQr9CxHBOeEgWU7nDA7W2aEidFGKd2mFpnZwFMw3HPfu3nfvvQOoqPiDYACa+GKjKUWsoPlpPL9bEK9RCNQDqAlV2fmNIlMJkHiu7K9TStQv/HEy6H0aVraZveTdNoJtluTAmT0ahkT0edCA2sNyGR3FNIlSYjXZlh14vINAd50W5MyOnrhUHNOHR7qkGJjS5Fkl9j1DZnDUZ0LxmhgmQsHe3gcCRaRz8kOZEfsukMpMKsfTQhkNPFnlfFXRpXLbJ3akSMUvqqhYNubnhcJWlXooZrm+tCwsGht6aQepumu1KO9YUe6e1G5IZvbLs+AsdN7CqWFoV5y8e732oaY743yStndHAyeC1+5pQ8OlHwNAtBSCKhcKwSIULamYl1M2yi98c7a018K1KRUVFRX/hS9RoX7PzKerkAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-5815-4585","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":true,"prefix":"","firstName":"Koji","middleName":"","lastName":"Kato","suffix":""},{"id":623368102,"identity":"2c3c6348-d562-4918-b6e6-79067721de6e","order_by":1,"name":"Ken Takigawa","email":"","orcid":"","institution":"Hamanomachi Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ken","middleName":"","lastName":"Takigawa","suffix":""},{"id":623368103,"identity":"021c47bb-bc9b-416a-bdd6-73c29a1b2798","order_by":2,"name":"Kohta Miyawaki","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Kohta","middleName":"","lastName":"Miyawaki","suffix":""},{"id":623368104,"identity":"21b26aff-88a6-4dd5-85c3-4ff30877b141","order_by":3,"name":"Shiho Taniguchi","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Shiho","middleName":"","lastName":"Taniguchi","suffix":""},{"id":623368105,"identity":"56bb7575-00fd-4d35-9aba-9bc9562d272a","order_by":4,"name":"Masatoshi Shimo","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Sciences","correspondingAuthor":false,"prefix":"","firstName":"Masatoshi","middleName":"","lastName":"Shimo","suffix":""},{"id":623368106,"identity":"2aa04b44-14c4-49d0-b32f-f23e3683ab66","order_by":5,"name":"Kyohei Mori","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Kyohei","middleName":"","lastName":"Mori","suffix":""},{"id":623368107,"identity":"45d62a15-3c7e-4022-8cb1-4a289c99af3b","order_by":6,"name":"Tatsuya Terasaki","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Tatsuya","middleName":"","lastName":"Terasaki","suffix":""},{"id":623368108,"identity":"8e7f84ec-5ae1-4439-a862-6a4a599bbee6","order_by":7,"name":"Teppei Sakoda","email":"","orcid":"","institution":"Kyushu University Graduate School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Teppei","middleName":"","lastName":"Sakoda","suffix":""},{"id":623368109,"identity":"75a1c760-2fff-41ca-9d7b-1c3d372aad4f","order_by":8,"name":"Fumiaki Jinnouchi","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Fumiaki","middleName":"","lastName":"Jinnouchi","suffix":""},{"id":623368110,"identity":"1d031454-a6f8-4d02-9389-b97fc76b86d9","order_by":9,"name":"Takuji Yamauchi","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Takuji","middleName":"","lastName":"Yamauchi","suffix":""},{"id":623368111,"identity":"defdfc9b-d5e6-4ac2-99af-1f16ef49bb2c","order_by":10,"name":"Takahiro Shima","email":"","orcid":"","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Takahiro","middleName":"","lastName":"Shima","suffix":""},{"id":623368112,"identity":"090d90f4-53f1-4bd8-ac2c-e9557444d403","order_by":11,"name":"Yoshikane Kikushige","email":"","orcid":"https://orcid.org/0000-0002-7721-1696","institution":"Kyushu University Graduate School of Medical Science","correspondingAuthor":false,"prefix":"","firstName":"Yoshikane","middleName":"","lastName":"Kikushige","suffix":""},{"id":623368113,"identity":"59c36c97-fbaf-4777-ad45-5c71770ff0f1","order_by":12,"name":"Yasuo Mori","email":"","orcid":"https://orcid.org/0000-0001-6425-1720","institution":"Kyushu University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yasuo","middleName":"","lastName":"Mori","suffix":""},{"id":623368114,"identity":"0caa0d0a-1671-43ba-95ab-fc576fcd6028","order_by":13,"name":"Koichi Akashi","email":"","orcid":"","institution":"Kyushu University","correspondingAuthor":false,"prefix":"","firstName":"Koichi","middleName":"","lastName":"Akashi","suffix":""}],"badges":[],"createdAt":"2026-04-12 21:20:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9396828/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9396828/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107617672,"identity":"97af374a-b375-4d2d-9f33-e246d2fcf770","added_by":"auto","created_at":"2026-04-23 09:21:41","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":227207,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGating strategy for the identification of CAR-T cells in CSF by multicolor flow cytometry.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCells were first gated on FSC-A vs. SSC-A to identify leukocytes while excluding erythrocytes, and singlets were selected using FSC-H vs. SSC-W. Lymphocytes were subsequently gated using PI, CD14, CD33, and CD235 in combination with SSC-A to exclude dead cells, monocytes, granulocytes, and RBCs. T and B lymphocytes were distinguished by CD3 and CD19 expression, respectively. CAR-T cells were defined as CD3⁺ FMC63⁺ and further subdivided into CD4⁺ and CD8⁺ subsets.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/cf4e1ddab68ed0a4f272eb8c.png"},{"id":107617674,"identity":"9c68ccd3-c333-4717-8743-1288fee2d316","added_by":"auto","created_at":"2026-04-23 09:21:41","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":31281,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest plot comparing the diagnostic performance of clinical, PB, CSF, and EEG parameters for ICANS.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eArea under the curve (AUC) with 95% confidence intervals estimated by the DeLong method. Bold indicates parameters with p \u0026lt; 0.05. Cutoff values were determined by the Youden index (maximizing sensitivity + specificity − 1), except for EEG overall assessment (≥mildly abnormal, pre-specified). FIRDA: AUC = 0.50, no discriminative cutoff identified.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/63c9d5a82f34c447bf7284c3.png"},{"id":107617676,"identity":"428b13c5-a5e1-4fe0-b249-80201e1e37b7","added_by":"auto","created_at":"2026-04-23 09:21:41","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":215012,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCerebrospinal fluid findings in ICANS and non-ICANS patients.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e CSF leukocyte counts in patients with and without ICANS. \u003cstrong\u003eb\u003c/strong\u003eThe proportion of CAR-T cells within CD3⁺ T cells in CSF (p = 0.018). \u003cstrong\u003ec\u003c/strong\u003e CD4⁺/CD8⁺ ratio of CAR-T cells and non-CAR-T cells in CSF and peripheral blood (PB).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/6b520d59ee19335a24859bb5.png"},{"id":107617677,"identity":"7ff00f9b-5e02-4af2-9b18-6d0bdf1caa22","added_by":"auto","created_at":"2026-04-23 09:21:41","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":226417,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiagnostic performance of CSF CAR-T cell frequency alone and in combination with PB or clinical parameters for ICANS diagnosis.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea\u003c/strong\u003e Receiver operating characteristic (ROC) curve for CAR-T cell frequency in CD3⁺ T cells in CSF. A cutoff of ≥20% yielded a sensitivity of 0.72 and specificity of 0.83 (AUC 0.80). \u003cstrong\u003eb\u003c/strong\u003e ROC curves for fibrinogen alone and in combination with CSF CAR-T cell frequency. \u003cstrong\u003ec\u003c/strong\u003e ROC curves for CRS grade alone and in combination with CSF CAR-T cell frequency.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/8599fcf48d15911b421138c4.png"},{"id":107708922,"identity":"6c779470-f58d-47f6-a148-c55b230bd150","added_by":"auto","created_at":"2026-04-24 09:33:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":794186,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/56792df0-542c-463b-9788-034d403a8a6d.pdf"},{"id":107617673,"identity":"97e52ed4-8c07-4269-8eb6-7696657d690a","added_by":"auto","created_at":"2026-04-23 09:21:41","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":161088,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Materials\u003c/p\u003e","description":"","filename":"SupplementaryMaterials.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/5e2d36a493d26e67e97f77c9.pdf"},{"id":107705843,"identity":"8c8e05f4-ab16-45d2-b66e-dd47115dc904","added_by":"auto","created_at":"2026-04-24 09:15:31","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":131194,"visible":true,"origin":"","legend":"","description":"","filename":"BMTTable.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9396828/v1/f8fd53ecc4efac3ddf7b9d7c.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e","formattedTitle":"Direct Flow Cytometric Assessment of Cerebrospinal Fluid CAR-T Cells for ICANS Diagnosis","fulltext":[{"header":"Introduction","content":"\u003cp\u003eChimeric antigen receptor (CAR)-T cell therapy has revolutionized the treatment of relapsed or refractory hematologic malignancies, achieving high remission rates in patients with otherwise limited options [\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, as CAR-T cell therapy becomes more widely adopted, immune effector cell-associated neurotoxicity syndrome (ICANS) has emerged as a critical management challenge, significantly impacting treatment outcomes and patient safety. ICANS occurs in 20% to 70% of patients receiving CAR-T cell therapy and presents with various neurological symptoms. Mild cases present with reversible symptoms, such as confusion, delirium, and aphasia, while severe cases may progress to seizures, coma, and fatal cerebral edema [\u003cspan additionalcitationids=\"CR2 CR3\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. The unpredictable nature of ICANS and its potentially life-threatening outcomes underscore the clinical necessity of accurate, early diagnosis and reliable severity assessment to optimize therapeutic interventions and improve patient outcomes.\u003c/p\u003e \u003cp\u003eCurrent diagnostic and monitoring approaches for ICANS include the Immune Effector Cell-Associated Encephalopathy (ICE) Score, electroencephalography (EEG), neuroimaging, and cerebrospinal fluid (CSF) analysis. The ICE score, while widely used as a standardized grading tool, relies on bedside cognitive assessment and is inherently subject to inter-rater variability and situational factors such as patient sedation. EEG and neuroimaging findings are nonspecific and not pathognomonic for ICANS [\u003cspan additionalcitationids=\"CR6 CR7\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. CSF analyses, such as total cell count and protein concentration, are insufficient to differentiate ICANS from other neurological complications. Consequently, the diagnosis of ICANS currently requires exclusion of other neurological etiologies, highlighting the unmet need for objective, specific biomarkers.\u003c/p\u003e \u003cp\u003eAlthough the precise pathophysiology of ICANS remains incompletely understood, accumulating evidence suggests that blood-brain barrier (BBB) disruption, immune cell trafficking into the central nervous system (CNS), and subsequent neuroinflammation mediated by cytokines play a central role. Most biomarker research to date has focused on peripheral blood (PB) markers, including serum cytokine levels and CAR-T cell kinetics [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]; however, these systemic markers may not adequately capture the immune processes within the CNS. CSF is obtained directly from the CNS compartment and may better reflect the local immune milieu at the site of pathology. This biological proximity makes CSF an attractive source for biomarkers with greater relevance to real-world diagnostic decision-making. This study, therefore, aimed to address this diagnostic gap by quantifying CAR-T cells in CSF. Prior studies have detected CAR-T cell DNA in CSF using PCR-based methods, but the quantity of transgene did not correlate with neurotoxicity severity [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], and these approaches cannot characterize the phenotypic composition of infiltrating T cells. We therefore hypothesized that multicolor flow cytometry, by enabling direct quantification of CAR-T cells and their subsets within CSF, could provide a more diagnostically informative assessment of ICANS.\u003c/p\u003e"},{"header":"Subjects and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCSF Sample Collection\u003c/h2\u003e \u003cp\u003eFrom June 2021 to October 2025, CSF samples were collected from patients undergoing CAR-T cell therapy at Kyushu University Hospital when ICANS was suspected, as part of diagnostic evaluation. A lumbar puncture was performed at the onset of neurological symptoms in accordance with institutional clinical practice. The study was approved by the Institutional Review Board of Kyushu University Hospital (approval number: 22062) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eClinical and Laboratory Data\u003c/h3\u003e\n\u003cp\u003eClinical parameters collected included age, sex, disease status prior to CAR-T cell infusion, type of CAR-T cell product administered, and grading of cytokine release syndrome (CRS) and ICANS according to the American Society for Transplantation and Cellular Therapy (ASTCT) criteria [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. PB parameters were collected at the time of clinical suspicion of ICANS, including platelet count, lactate dehydrogenase (LDH), creatinine, C-reactive protein (CRP), ferritin, and fibrinogen. Endothelial activation was assessed using the Endothelial Activation and Stress Index (EASIX) and its modified forms [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eEASIX = (LDH [U/L] \u0026times; creatinine [mg/dL]) / platelet count [10⁹/L]\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003emodified-EASIX = (CRP [mg/dL] \u0026times;LDH [U/L]) / platelet count [10⁹/L]\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003esimplified-EASIX\u0026thinsp;=\u0026thinsp;LDH [U/L] / platelet count [10⁹/L]\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e\n\u003ch3\u003eElectroencephalography\u003c/h3\u003e\n\u003cp\u003eElectroencephalography (EEG) was performed using the International 10\u0026ndash;20 system with bipolar longitudinal montages. Given the clinical severity of patients, post-CAR-T EEG evaluations were primarily performed at the bedside at suspected ICANS onset. Findings were categorized as normal, mildly abnormal, or moderately abnormal by a board-certified neurophysiologist. The presence or absence of frontal intermittent rhythmic delta activity (FIRDA) was evaluated [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eFlow Cytometry Analysis\u003c/h3\u003e\n\u003cp\u003eCSF samples were processed immediately after collection. After centrifugation at 500 \u0026times; g for 5 minutes, the pellet was resuspended in staining medium consisting of HBSS (\u0026minus;) supplemented with 2% heat-inactivated fetal bovine serum and 4 mM EDTA. Cells were stained with the following fluorochrome-conjugated monoclonal antibodies: PE-conjugated FMC63 (REA1297, Miltenyi Biotec, Bergisch Gladbach, Germany), BV510-conjugated CD3 (UCHT1, BioLegend, San Diego, USA), BV785-conjugated CD4 (SK3, BD Biosciences San Diego, USA), BV570-conjugated CD8 (RPA-T8, BioLegend), BV605-conjugated CD19 (HIB19, BioLegend), and PerCP/Cy5.5-conjugated CD14 (63D3, BioLegend), CD33 (WM53, BioLegend), CD235ab (HIR2, BioLegend). After staining, samples were incubated for 1 hour at 4℃ in the dark with gentle agitation. Subsequently, staining medium (999 \u0026micro;L) and propidium iodide (final concentration: 1 \u0026micro;g/mL) were added for dead cell exclusion, and cells were washed by centrifugation at 500 \u0026times; g for 5 minutes and resuspended in 200 \u0026micro;L of staining medium. Multicolor flow cytometry was performed on a FACSAria IIIu (BD Biosciences, San Jose, CA, USA) or MA900 (Sony Biotechnology Inc., San Jose, CA, USA). CAR-T cells were identified as CD3⁺ FMC63⁺ cells within the lymphocyte gate, after exclusion of dead cells, monocytes, granulocytes, and erythrocytes. Due to institutional equipment transition, flow cytometry was performed on a FACSAria IIIu for samples collected before June 2023 and on an MA900 thereafter. Instrument settings were optimized to ensure comparable results, and consistent gating strategies were applied across both platforms.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eThe Wilcoxon rank-sum test was used to compare continuous or ordinal variables between groups, and Pearson\u0026rsquo;s chi-squared test was employed for nominal variables. Correlations between continuous variables were assessed using Spearman\u0026rsquo;s rank correlation coefficient. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic accuracy of CSF CAR-T cell proportions, with the optimal cutoff value determined by the Youden index. All statistical analyses were performed using JMP version 19. A two-sided p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003ePatient Characteristics\u003c/h2\u003e \u003cp\u003eTwenty-four patients who underwent CSF sampling at the time of clinical suspicion of ICANS were included in the analysis. Patients were categorized into two groups: ICANS (n\u0026thinsp;=\u0026thinsp;18) and non-ICANS (n\u0026thinsp;=\u0026thinsp;6) (Table\u0026nbsp;1). The non-ICANS group consisted of patients with an alternative confirmed cause of neurological symptoms, such as CNS infiltration (n\u0026thinsp;=\u0026thinsp;2), or patients whose symptoms resolved spontaneously without dexamethasone treatment (e.g., headache, aphasia, wrist dorsiflexion weakness, or transient expressive difficulty; each in one patient). The median age of the cohort was 65 years (range, 42\u0026ndash;73), and the majority were male (n\u0026thinsp;=\u0026thinsp;18, 75%). The most common primary disease was DLBCL in 19 patients (79%), followed by high-grade B-cell lymphoma in 3 (13%) and follicular lymphoma in 2 (8%). The CAR-T cell products administered were tisagenlecleucel (tisa-cel, n\u0026thinsp;=\u0026thinsp;9), lisocabtagene maraleucel (liso-cel, n\u0026thinsp;=\u0026thinsp;9), and axicabtagene ciloleucel (axi-cel, n\u0026thinsp;=\u0026thinsp;6). Five patients were in complete response at the time of infusion. CRS occurred in 22 patients (92%), including 3 (13%) with grade\u0026thinsp;\u0026ge;\u0026thinsp;3. ICANS developed in 18 patients (75%), including 2 (11%) with grade\u0026thinsp;\u0026ge;\u0026thinsp;3, with a median onset of 6 days (range, 3\u0026ndash;38).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eExploration of ICANS-Associated Parameters\u003c/h3\u003e\n\u003cp\u003eTo identify potential biomarkers for diagnosing ICANS, we explored parameters associated with ICANS onset. Several risk factors for ICANS have been identified, including high tumor burden, elevated inflammatory markers, endothelial activation indices, and CRS severity; however, these parameters have predominantly been evaluated at baseline or at the time of CAR-T cell infusion [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. In this study, we assessed previously reported parameters at the time of neurological symptom onset and additionally evaluated CAR-T cell frequencies in CSF detected by multicolor flow cytometry. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e illustrates the flow cytometry gating strategy used to detect CAR-T cells in CSF and PB. The diagnostic performance of each parameter was assessed using ROC analysis, and the area under the curve (AUC) with 95% confidence intervals (CIs) is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eDiagnostic value of CAR-T cell frequency in CSF\u003c/h2\u003e \u003cp\u003eAmong PB parameters, fibrinogen and creatinine levels showed high discriminatory ability (AUC 0.86 and 0.84, respectively) (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Patients with ICANS showed lower fibrinogen levels (median 232 mg/dL, IQR 182\u0026ndash;286) compared with the non-ICANS group (median 359 mg/dL, IQR 280\u0026ndash;483; p\u0026thinsp;=\u0026thinsp;0.005) (Supplementary Table\u0026nbsp;1). Creatinine levels also demonstrated good discrimination with significantly higher levels in the ICANS group (median 0.85 [0.68\u0026ndash;1.05] vs. 0.56 [0.35\u0026ndash;0.69] mg/dL; p\u0026thinsp;=\u0026thinsp;0.007). Other parameters, including platelet count, LDH, CRP, ferritin, and EASIX-related scores, showed modest AUCs that were not statistically significant.\u003c/p\u003e \u003cp\u003eRegarding clinical parameters, the overall EEG assessment showed moderate discriminatory ability (AUC 0.76, 95% CI 0.54\u0026ndash;0.98), although the distribution did not differ significantly between groups (p\u0026thinsp;=\u0026thinsp;0.125) (Supplementary Table\u0026nbsp;1). FIRDA was equally prevalent in both groups (75%), providing no discriminatory value (AUC 0.50). CRS grade also demonstrated moderate discrimination (AUC 0.72, 95% CI 0.56\u0026ndash;0.89), with a cutoff of grade\u0026thinsp;\u0026ge;\u0026thinsp;2 achieving a specificity of 1.00 and sensitivity of 0.56 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNone of the routine CSF parameters reached statistical significance (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Notably, there was no significant difference in the CSF leukocyte counts between the ICANS and non-ICANS groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea), suggesting that total cellularity alone is insufficient for discriminating ICANS. In contrast, flow cytometry-derived CSF CAR-T cell parameters demonstrated superior diagnostic performance with high AUCs (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The proportion of CAR-T cells within CD3⁺ T cells in the CSF was significantly higher in patients with ICANS (32.3% [15.3\u0026ndash;53.8]) than in those without ICANS (7.9% [2.7\u0026ndash;15.0]; p\u0026thinsp;=\u0026thinsp;0.018) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). The CD4/CD8 ratio in the CSF CAR-T cells did not differ significantly between groups.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eCAR-T cells in CSF versus PB\u003c/h2\u003e \u003cp\u003eIn a subset of 11 patients, PB was collected concurrently with CSF sampling and submitted for multicolor flow cytometry. The correlation between PB and CSF CAR-T cell frequencies showed a moderate but non-significant positive trend (r\u0026thinsp;=\u0026thinsp;0.482, p\u0026thinsp;=\u0026thinsp;0.133) (Supplementary Fig.\u0026nbsp;1), indicating that PB CAR-T cell levels do not reliably predict CSF CAR-T cell infiltration. We further examined the compartmental distribution of CAR-T cells by comparing the CD4/CD8 ratio between PB and CSF. The CD4/CD8 ratio of CAR-T cells was significantly higher in CSF than in PB (p\u0026thinsp;=\u0026thinsp;0.002), indicating a preferential enrichment of CD4⁺ CAR-T cells in the CNS compartment (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec). This CD4-dominant skewing in CSF was also observed among non-CAR-T cells (p\u0026thinsp;=\u0026thinsp;0.011) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eDiagnostic Performance of CSF CAR-T Cell Frequency\u003c/h2\u003e \u003cp\u003eROC analysis for the CSF CD3⁺ CAR-T cell frequency identified an optimal cutoff of \u0026ge;\u0026thinsp;20%, achieving a sensitivity of 0.72 and specificity of 0.83 (AUC 0.80, 95% CI 0.60\u0026ndash;0.99) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). To explore whether CSF flow cytometry data could complement existing diagnostic parameters, we examined the effect of combining the CSF CD3⁺ CAR-T cell frequency (\u0026ge;\u0026thinsp;20%) with either a PB parameter (fibrinogen\u0026thinsp;\u0026le;\u0026thinsp;244 mg/dL) or a clinical parameter (CRS grade\u0026thinsp;\u0026ge;\u0026thinsp;2). When CSF data were added to fibrinogen, the AUC improved from 0.86 (95% CI 0.71-1.00; sensitivity 0.72, specificity 1.00) to 0.91 (95% CI 0.77-1.00; sensitivity 0.89, specificity 0.83) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Similarly, the addition of CSF data to CRS grade improved the AUC from 0.72 (95% CI 0.56\u0026ndash;0.89; sensitivity 0.556, specificity 1.000) to 0.89 (95% CI 0.74-1.00; sensitivity 0.89, specificity 0.83) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). In both cases, incorporating CSF-derived information improved sensitivity while maintaining acceptable specificity, suggesting that CSF flow cytometry data reflecting the local immune environment at the site of pathology may serve as a useful complement to conventional parameters in the diagnostic evaluation of ICANS.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study demonstrates that the proportion of CAR-T cells within CD3⁺ T cells in CSF correlates with ICANS onset, with CD4⁺ CAR-T cells being predominant. A provisional cutoff value of 20% effectively distinguished ICANS with good sensitivity and specificity. Conventional diagnostic tools such as the ICE score, EEG, and neuroimaging have low specificity, and systemic inflammatory markers such as CRP and ferritin also cannot reliably distinguish ICANS from CRS or other inflammatory conditions. As no objective test for ICANS has been established to date, this approach offers a targeted diagnostic tool that can identify patients with ICANS who require steroid treatment. To our knowledge, this is the first study to systematically evaluate CSF CAR-T cell quantification by flow cytometry as a diagnostic biomarker for ICANS, extending prior technical validation of flow cytometric CAR-T detection in CSF [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Importantly, all CSF samples in this study were obtained as part of routine diagnostic evaluation when ICANS was suspected, rather than through research-specific procedures, underscoring the feasibility of implementing this approach in standard clinical practice. Furthermore, combining the CSF CD3⁺ CAR-T cell frequency (\u0026ge;\u0026thinsp;20%) with either fibrinogen (\u0026le;\u0026thinsp;244 mg/dL) or CRS grade (\u0026ge;\u0026thinsp;2) improved diagnostic accuracy compared with each parameter alone, suggesting that CSF flow cytometry can complement existing parameters. The lower fibrinogen levels observed in the ICANS group may reflect consumptive coagulopathy associated with endothelial activation [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] as well as tocilizumab-induced suppression of IL-6-dependent fibrinogen synthesis [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. As these systemic markers capture different biological dimensions from CSF-derived parameters reflecting the local immune milieu, a multiparameter approach integrating both may enhance diagnostic precision for ICANS.\u003c/p\u003e \u003cp\u003ePrevious studies using PCR-based methods have shown that CAR-T cells can be detected in CSF regardless of neurotoxicity status. Santomasso et al. reported that the quantity of CAR-T cell DNA in CSF did not correlate with neurotoxicity severity [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], and similar findings were reported by Mueller et al. using quantitative PCR [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] and Berger et al. using digital PCR [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. These DNA-based approaches measure transgene copy number and cannot characterize the phenotype or subset composition of CAR-T cells. In contrast, multicolor flow cytometry enables direct identification of CAR-T cells at the protein level and simultaneous assessment of their proportion within T cells. Our findings indicate that it is not the absolute quantity of CAR-T DNA but rather the proportion of CAR-T cells among CD3⁺ T cells in CSF that distinguishes ICANS, as confirmed by comparison with non-ICANS patients with alternative neurological conditions. Although the sample size was limited, we did not observe a clear correlation between CSF CAR-T cell fraction and ICANS severity. Notably, the absolute CSF leukocyte counts were extremely limited in many cases (median 3 cells/\u0026micro;L), which limits the feasibility of PCR-based quantification. In this context, multicolor flow cytometry offers a distinct advantage by enabling direct phenotypic assessment including T-cell subset characterization even in paucicellular specimens.\u003c/p\u003e \u003cp\u003eAutopsy studies of fatal ICANS cases have provided the primary basis for understanding its neuropathology. Gust et al. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] demonstrated endothelial activation and BBB disruption, and Karschnia et al. [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] reported perivascular accumulation of predominantly CD8⁺ T cells, although CD4⁻/CD8⁻ double-negative T cells were also present. These autopsy findings from fatal cases have shaped a predominantly CD8-centric interpretation of ICANS. However, these observations were derived from the most severe, fatal cases and may not fully represent the immunological landscape of clinically typical, treatable ICANS. This is consistent with Shah et al., who observed CD4⁺ non-CAR T cell predominance in CSF of two patients with steroid-refractory ICANS by flow cytometry [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], although systematic evaluation was not performed. An important question is whether this CD4 predominance in CSF reflects a physiological characteristic of the CNS compartment or a process specific to ICANS. The CD4/CD8 ratio was significantly higher in CSF than in PB for both CAR-T and non-CAR-T cells, consistent with the known predominance of central memory CD4⁺ T cells in normal CSF [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. This suggests that the CD4 skewing observed in our study is at least partly attributable to the physiological composition of the CNS immune compartment. Nevertheless, several lines of evidence suggest that CD4⁺ T cells may also be functionally relevant to the pathogenesis of ICANS. CD4⁺ T cell-dominant neuroinflammation has been well documented in other CNS inflammatory diseases; for example, in multiple sclerosis, activated memory CD4⁺ T cells migrate across the BBB and drive sustained inflammation within the CNS [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. In the specific context of CAR-T therapy, Lu et al. recently demonstrated that the CXCL16-CXCR6 axis selectively recruits CD4⁺-dominant T cells into the CNS during ICANS [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], providing a mechanistic basis for the preferential enrichment of CD4⁺ CAR-T cells in CSF. Furthermore, Baur et al. reported that greater peripheral expansion of CD4⁺ CAR-T cells correlates with increased severity of both CRS and ICANS [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], supporting a functional role for CD4⁺ T cells in immune-mediated toxicity. Taken together, these findings suggest that the predominance of CD4⁺ T cells in CSF may reflect both the physiological CNS immune composition and active chemokine-driven recruitment of CD4⁺ CAR-T cells during ICANS.\u003c/p\u003e \u003cp\u003eThis study has several limitations. First, the sample size was small, and the study was conducted at a single center, which limits the generalizability of the findings and the statistical power to detect associations with ICANS severity. Second, the non-ICANS control group was heterogeneous, comprising patients with CNS infiltration and those with self-limiting neurological symptoms, which may have influenced the discriminative performance of the evaluated parameters. Finally, a more detailed characterization of CD4⁺ T-cell subsets in CSF, in both CAR-T and non-CAR-T cells, is needed to further elucidate the pathogenesis of ICANS.\u003c/p\u003e \u003cp\u003eIn conclusion, this study demonstrates that quantifying CAR-T cells in CSF by multicolor flow cytometry provides an objective measure for identifying ICANS that requires therapeutic intervention. The proportion of CAR-T cells among CD3⁺ T cells in CSF distinguished ICANS from other neurological conditions, and combining this measure with existing parameters further improved diagnostic accuracy. The observed CD4⁺ predominance among CSF CAR-T cells offers new insight into the immunological landscape of ICANS, complementing the CD8-centric model derived from fatal cases. Multicenter validation of the proposed diagnostic cutoff and detailed characterization of CD4⁺ T-cell subsets in the CNS compartment are warranted.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the patients and their families for participating in this study. We are grateful to the clinical staff at Kyushu University Hospital for their dedicated care of CAR-T recipients. We thank Arisa Matsuyama and Naoko Ban for their technical assistance with flow cytometric analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eK.T., K.Miyawaki and K.K designed the study, collected and analyzed data, and wrote the manuscript. S.T., M.S., K.Mori., T.T., T.Sakoda., F.J., T.Y., and T.Shima. contributed to patient care and data collection. K.Miyawaki., Y.K., Y.M., K.A., and K.K. supervised the study and critically reviewed the manuscript. All authors approved the final version.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKoji Kato; Honoraria: AbbVie, Bristol-Myers Squibb, Chugai, Janssen, Kyowa Kirin, Ono, Gilead Sciences, Novartis; Consulting or Advisory Role: AbbVie, AstraZeneca, Chugai, Daiichi Sankyo, Eisai, Janssen, Bristol-Myers Squibb, Novartis, Gilead Sciences; Research Funding: AbbVie, MSD, Bristol-Myers Squibb, Chugai, Daiichi Sankyo, Eisai, Janssen, Kyowa Kirin, Novartis, Ono, Gilead Sciences. Kohta Miyawaki; Honoraria: Bristol-Myers Squibb, Gilead Sciences, Novartis. The remaining authors declare that they have no conflict of interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eNeelapu SS, Tummala S, Kebriaei P, Wierda W, Gutierrez C, Locke FL, et al. Chimeric antigen receptor T-cell therapy \u0026mdash; assessment and management of toxicities. Nat Rev Clin Oncol. 2018;15:47\u0026ndash;62. https://doi.org/10.1038/nrclinonc.2017.148\u003c/li\u003e\n \u003cli\u003eGust J, Hay KA, Hanafi L-A, Li D, Myerson D, Gonzalez-Cuyar LF, et al. Endothelial Activation and Blood\u0026ndash;Brain Barrier Disruption in Neurotoxicity after Adoptive Immunotherapy with CD19 CAR-T Cells. Cancer Discov. 2017;7:1404\u0026ndash;19. https://doi.org/10.1158/2159-8290.cd-17-0698\u003c/li\u003e\n \u003cli\u003eSantomasso BD, Park JH, Salloum D, Rivi\u0026egrave;re I, Flynn J, Mead E, et al. Clinical and Biologic Correlates of Neurotoxicity Associated with CAR T Cell Therapy in Patients with B-cell Acute Lymphoblastic Leukemia (B-ALL). Cancer Discov. 2018;8:CD-17-1319. https://doi.org/10.1158/2159-8290.cd-17-1319\u003c/li\u003e\n \u003cli\u003eAbramson JS. Anti-CD19 CAR T-Cell Therapy for B-Cell Non-Hodgkin Lymphoma. Transfus Med Rev. 2020;34:29\u0026ndash;33. https://doi.org/10.1016/j.tmrv.2019.08.003\u003c/li\u003e\n \u003cli\u003eLee DW, Santomasso BD, Locke FL, Ghobadi A, Turtle CJ, Brudno JN, et al. ASTCT Consensus Grading for Cytokine Release Syndrome and Neurologic Toxicity Associated with Immune Effector Cells. Biol Blood Marrow Transplant. 2019;25:625\u0026ndash;38. https://doi.org/10.1016/j.bbmt.2018.12.758\u003c/li\u003e\n \u003cli\u003eStrati P, Nastoupil LJ, Westin J, Fayad LE, Ahmed S, Fowler NH, et al. Clinical and radiologic correlates of neurotoxicity after axicabtagene ciloleucel in large B-cell lymphoma. Blood Adv. 2020;4:3943\u0026ndash;51. https://doi.org/10.1182/bloodadvances.2020002228\u003c/li\u003e\n \u003cli\u003eHunter BD, Jacobson CA. CAR T-Cell Associated Neurotoxicity: Mechanisms, Clinicopathologic Correlates, and Future Directions. JNCI: J Natl Cancer Inst. 2019;111:646\u0026ndash;54. https://doi.org/10.1093/jnci/djz017\u003c/li\u003e\n \u003cli\u003eGust J, Taraseviciute A, Turtle CJ. Neurotoxicity Associated with CD19-Targeted CAR-T Cell Therapies. CNS Drugs. 2018;32:1091\u0026ndash;101. https://doi.org/10.1007/s40263-018-0582-9\u003c/li\u003e\n \u003cli\u003eSterner RC, Sterner RM. Immune effector cell associated neurotoxicity syndrome in chimeric antigen receptor-T cell therapy. Front Immunol. 2022;13:879608. https://doi.org/10.3389/fimmu.2022.879608\u003c/li\u003e\n \u003cli\u003eWang X, He X, Zhang T, Liu J, Zhao M. Latest updates on pathogenesis mechanisms and management strategies for cytokine release syndrome, neurotoxicity, and hemophagocytic lymphohistiocytosis related to CAR-T cell therapies. Ann Hematol. 2025;104:3129\u0026ndash;51. https://doi.org/10.1007/s00277-025-06467-y\u003c/li\u003e\n \u003cli\u003eMueller KT, Maude SL, Porter DL, Frey N, Wood P, Han X, et al. Cellular kinetics of CTL019 in relapsed/refractory B-cell acute lymphoblastic leukemia and chronic lymphocytic leukemia. Blood. 2017;130:2317\u0026ndash;25. https://doi.org/10.1182/blood-2017-06-786129\u003c/li\u003e\n \u003cli\u003eBerger SC, Fehse B, Aky\u0026uuml;z N, Geffken M, Wolschke C, Janson D, et al. Molecular monitoring of T-cell kinetics and migration in severe neurotoxicity after real-world CD19-specific chimeric antigen receptor T cell therapy. Haematologica. 2022;108:444\u0026ndash;56. https://doi.org/10.3324/haematol.2022.281110\u003c/li\u003e\n \u003cli\u003eLuft T, Benner A, Terzer T, Jodele S, Dandoy CE, Storb R, et al. EASIX and mortality after allogeneic stem cell transplantation. Bone Marrow Transplant. 2020;55:553\u0026ndash;61. https://doi.org/10.1038/s41409-019-0703-1\u003c/li\u003e\n \u003cli\u003ePennisi M, Sanchez-Escamilla M, Flynn JR, Shouval R, Tomas AA, Silverberg ML, et al. Modified-EASIX predicts severe cytokine release syndrome and neurotoxicity after Chimeric Antigen Receptor (CAR) T cells. Blood Adv. 5:3397\u0026ndash;406. https://doi.org/10.1182/bloodadvances.2020003885\u003c/li\u003e\n \u003cli\u003eHuby S, Gelisse P, Tudesq J-J, Labauge P, Duflos C, Cartron G, et al. Frontal Intermittent Rhythmic Delta Activity Is a Useful Diagnostic Tool of Neurotoxicity After CAR T-Cell Infusion. Neurol Neuroimmunol Neuroinflammation. 2023;10:e200111. https://doi.org/10.1212/nxi.0000000000200111\u003c/li\u003e\n \u003cli\u003eGrant SJ, Grimshaw AA, Silberstein J, Murdaugh D, Wildes TM, Rosko AE, et al. Clinical Presentation, Risk Factors, and Outcomes of Immune Effector Cell-Associated Neurotoxicity Syndrome Following Chimeric Antigen Receptor T Cell Therapy: A Systematic Review. Transplant Cell Ther. 2022;28:294\u0026ndash;302. https://doi.org/10.1016/j.jtct.2022.03.006\u003c/li\u003e\n \u003cli\u003eKorell F, Penack O, Mattie M, Schreck N, Benner A, Krzykalla J, et al. EASIX and Severe Endothelial Complications After CD19-Directed CAR-T Cell Therapy\u0026mdash;A Cohort Study. Front Immunol. 2022;13:877477. https://doi.org/10.3389/fimmu.2022.877477\u003c/li\u003e\n \u003cli\u003eJohansson U, Gallagher K, Burgoyne V, Maus MV, Casey KS, Brini GG, et al. Detection of CAR‐T19 cells in peripheral blood and cerebrospinal fluid: An assay applicable to routine diagnostic laboratories. Cytom Part B: Clin Cytom. 2021;100:622\u0026ndash;31. https://doi.org/10.1002/cyto.b.22005\u003c/li\u003e\n \u003cli\u003ePerl M, Herfeld K, Harrer DC, H\u0026ouml;pting M, Schweiger M, Sterz U, et al. Tocilizumab administration in cytokine release syndrome is associated with hypofibrinogenemia after chimeric antigen receptor T-cell therapy for hematologic malignancies. Haematologica. 2024;109:2969\u0026ndash;77. https://doi.org/10.3324/haematol.2023.284564\u003c/li\u003e\n \u003cli\u003eKarschnia P, Str\u0026uuml;bing F, Teske N, Blumenberg V, B\u0026uuml;cklein VL, Schmidt C, et al. Clinicopathologic Findings in Fatal Neurotoxicity After Adoptive Immunotherapy With CD19‐Directed CAR T‐Cells. HemaSphere. 2021;5:e533. https://doi.org/10.1097/hs9.0000000000000533\u003c/li\u003e\n \u003cli\u003eShah NN, Johnson BD, Fenske TS, Raj RV, Hari P. Intrathecal chemotherapy for management of steroid-refractory CAR T-cell\u0026ndash;associated neurotoxicity syndrome. Blood Adv. 2020;4:2119\u0026ndash;22. https://doi.org/10.1182/bloodadvances.2020001626\u003c/li\u003e\n \u003cli\u003eGraaf MT de, Smitt PAES, Luitwieler RL, Velzen C van, Broek PDM van den, Kraan J, et al. Central memory CD4+ T cells dominate the normal cerebrospinal fluid. Cytom Part B: Clin Cytom. 2011;80B:43\u0026ndash;50. https://doi.org/10.1002/cyto.b.20542\u003c/li\u003e\n \u003cli\u003eChitnis T. The Role of CD4 T Cells in the Pathogenesis of Multiple Sclerosis. Int Rev Neurobiol. 2007;79:43\u0026ndash;72. https://doi.org/10.1016/s0074-7742(07)79003-7\u003c/li\u003e\n \u003cli\u003eLu I-N, M\u0026uuml;ller-Miny L, Krekeler C, Cheung PF-Y, Antonopoulou G, Jeibmann A, et al. The CXCL16/CXCR6 axis is linked to immune effector cell-associated neurotoxicity in chimeric antigen receptor (CAR) T cell therapy. Genome Med. 2025;17:71. https://doi.org/10.1186/s13073-025-01498-6\u003c/li\u003e\n \u003cli\u003eBaur K, Buser A, Jeker LT, Khanna N, L\u0026auml;ubli H, Heim D, et al. CD4+ CAR T-cell expansion is associated with response and therapy related toxicities in patients with B-cell lymphomas. Bone Marrow Transplant. 2023;58:1048\u0026ndash;50. https://doi.org/10.1038/s41409-023-02016-1\u003cstrong\u003e\u003c/strong\u003e\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table","content":"\u003cp\u003eTable 1 is available in the supplementary files section\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bone-marrow-transplantation","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"bmt","sideBox":"Learn more about [Bone Marrow Transplantation](http://www.nature.com/bmt/)","snPcode":"41409","submissionUrl":"https://mts-bmt.nature.com/cgi-bin/main.plex","title":"Bone Marrow Transplantation","twitterHandle":"@bmtjournal","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-9396828/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9396828/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDiagnosing immune effector cell-associated neurotoxicity syndrome (ICANS) following chimeric antigen receptor (CAR)-T cell therapy remains challenging due to the lack of objective, specific biomarkers. We analyzed cerebrospinal fluid (CSF) samples collected at neurological symptom onset from 24 patients following CD19-targeted CAR-T therapy including 18 with ICANS requiring steroid treatment and 6 without ICANS who had alternative neurological diagnoses or self-limiting symptoms. Multicolor flow cytometry identified CAR-T cells as CD3⁺ FMC63⁺ lymphocytes. The proportion of CAR-T cells within CD3⁺ T cells was significantly higher in ICANS patients than in non-ICANS patients (median 32.3% vs. 7.9%, p = 0.018), with a predominance of CD4⁺ T cells. Receiver operating characteristic analysis identified a diagnostic cutoff of ≥20%, achieving a sensitivity of 0.72 and a specificity of 0.83 (AUC 0.80). Combining this CSF parameter with peripheral blood markers such as fibrinogen or CRS grade further improved diagnostic accuracy, suggesting that CSF-derived parameters reflecting the local immune environment at the site of pathology complement systemic markers in the diagnostic evaluation of ICANS. These findings indicate that flow cytometric assessment of CAR-T cells in CSF, which uniquely allows proportional quantification within T-cell subsets, provides an objective diagnostic tool for ICANS requiring therapeutic intervention.\u003c/p\u003e","manuscriptTitle":"Direct Flow Cytometric Assessment of Cerebrospinal Fluid CAR-T Cells for ICANS Diagnosis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-23 09:21:29","doi":"10.21203/rs.3.rs-9396828/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-21T14:32:33+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2026-04-15T06:15:57+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-13T11:16:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-12T21:16:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Bone Marrow Transplantation","date":"2026-04-12T21:16:39+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bone-marrow-transplantation","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"bmt","sideBox":"Learn more about [Bone Marrow Transplantation](http://www.nature.com/bmt/)","snPcode":"41409","submissionUrl":"https://mts-bmt.nature.com/cgi-bin/main.plex","title":"Bone Marrow Transplantation","twitterHandle":"@bmtjournal","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"b941ee01-d711-4f98-a22a-fa1ab75c9b51","owner":[],"postedDate":"April 23rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66340229,"name":"Health sciences/Medical research/Translational research"},{"id":66340230,"name":"Biological sciences/Cancer/Cancer therapy/Cancer immunotherapy"},{"id":66340231,"name":"Health sciences/Health care/Diagnosis"},{"id":66340232,"name":"Biological sciences/Immunology/Immunotherapy"}],"tags":[],"updatedAt":"2026-04-23T09:21:29+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-23 09:21:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9396828","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9396828","identity":"rs-9396828","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}