Shaw, Monica

Abstract Introduction: CAR-T therapy has transformed cancer therapy. with its first Food and Drug Administration (FDA) approved product in August 2017. Given the expanding indications and growing experience, particularly with management of immediate toxicities requiring specialized expertise such as Cytokine Release Syndrome (CRS) and Immune Effector Cell-Associated Neurotoxicity syndrome (ICANS), FDA recently removed the Risk Evaluation and Mitigation Strategies (REMS) for CD19 and BCMA targeting CAR-T on June 26th2025. In particular, given that CRS and ICANS typically occur in the first 2 weeks post CAR-T infusion, the minimum required stay near CAR-T treatment center is reduced from one month to 2 weeks. While this could reduce treatment burden for patients (pts) who need to travel away from home to receive CAR-T, there is a need to update the transition of care from CAR-T center to pts' local oncology practice. Pts carry in their rate of recovery and have varying supportive care needs in the second half of the month post CAR-T infusion. To assist with care transition and planning, we examine the use of AI multimodal large language model (LLM) to predict patients' supportive care needs 2 weeks after CAR-T infusion based on the clinical data at the time of CAR-T evaluation, prior to leukapheresis. This timepoint was selected to also maximize travel, lodging and caregiver planning for pts and family as well. Methodology: Pts who were treated with FDA-approved CAR-T at Mayo Clinic Rochester between December 2022 and June 2025 were examined. Based on experiences with referral practices in the Midwest region, we defined patients with high monitoring needs as those who between day 14-30 post CAR-T had either hospital admission due to recurrent CRS or ICANS or required transfusion of more than 1 blood product per week (red blood cell or platelets). Others were classified as low monitoring needs. Clinical notes, labs, vital signs and electrocardiogram at the time of CAR-T evaluation were abstracted and summarized by Gemini LLM for association with patients with high versus low monitoring needs post CAR-T. Model was developed and tested for predicted needs and concordance with actual outcome of these pts. Results: Among the 362 pts who received FDA approved CAR-T in the timeframe, 84 pts had high monitoring needs. An additional 84 pts with low monitoring needs were randomly selected from 278 pts. These were divided into a training cohort and test cohort of 84 pts each, with 42 pts with high and 42 pts with low monitoring needs. Demographics of the two cohorts displayed no statistical difference. LLM analysis of the clinical data at the time of CAR-T eval identified 5 categories associated with risk for high monitoring needs: 1) Disease status (clinical notes about bulky, progressive disease or CNS involvement); 2) inflammatory and tumor burden markers (CRP, LDH); 3) hematologic status (platelet, plt; neutrophil count ANC); 4) renal function; 5) performance status. Using common clinical lab parameter convention (platelets <100, ANC< 1, CrCl< 30ml/min), and prioritizing for not missing pts with high monitoring needs, LLM model was defined to predict patients for high monitoring needs if high risk feature was identified in category 1 or any category combinations; whereas patients with no high risk feature or only 1 category that is not disease status was predicted as low monitoring needs. In this model, the sensitivity was 85.7%, specificity 23.8%, F1 score 65.5%. Applying this model to test cohort 2, the sensitivity was 83.3%, specificity 28.6%, and F1 score 65.4%. Discussion: We demonstrate proof-of-concept that LLM can be used to analyze clinical data at the time of evaluation for CAR-T eligibility to predict for supportive care needs post treatment. Not surprisingly, features identified by LLM to be associated with high monitoring needs are those previously reported by us and others at the time of CAR-T evaluation and or pre-lymphodepletion to be associated with risk for cytopenia, infection, hospitalization and non-relapse mortality. Based on the summary output generated by LLM, additional opportunities for classification methods were identified to improve specificity and balanced accuracy of the LLM model. This is a promising tool for further development to assist clinicians and patients to plan their treatment journey.

Background: Management of cytokine release syndrome (CRS) with chimeric antigen receptor T cell therapy (CAR-T) has evolved in the last decade. First line treatment tocilizumab (toci) was initially given for grade 2 or higher CRS in clinical trials (Toci/CRS≥2). Earlier use of toci in grade 1 CRS was found to be safe without negative impact on clinical response in cohort 4 of ZUMA-1 study for axicabtagene ciloleucel (axi-cel) for B-cell non-Hodgkin lymphoma (NHL; Toci/CRS1). This practice was also adopted for other CAR-T therapies for other disease indications. Subsequently, FDA also updated the package insert for axi-cel in 2021 to include the use of prophylactic dexamethasone (ppx dex) to be given on days 0, 1, 2 of CAR-T infusion based on the results from cohort 6 of ZUMA-1. We examined CRS and ICANS (immune effector cell associated neurologic syndrome) outcome of patients who received FDA approved axi-cel in standard-of-care (SOC) practice with these evolving practices. Of note, adoption of ppx dex has been variable across CAR-T treatment centers in the US (Khurana et al, Blood 2022). At Mayo Clinic Rochester, patients considered high-risk for severe CRS/ICANS (those who received bridging therapy and had LDH above upper limit of normal and/or with secondary CNS involvement or other CNS pathologies) would be considered for ppx dex. Methods: Medical records of patients who received axi-cel in SOC between 2018 and 2023 at Mayo Clinic Rochester were reviewed. ASTCT guideline was used for CRS and ICANS grading. Categorical variables were compared using Chi-squared test and continuous variables were compared using Kruskal-Wallis test. Results: Among 187 pts receiving SOC axi-cel, the median age was 60 years and 121 (65%) were males, 152 (81%) had CRS, of which 73 (39%) were grade ≥2; 97 (52%) had ICANS, of which 65 (35%) were grade ≥2. Of the 152 CRS pts, 54/152 (36%) pts never received toci or ppx/dex. Of the pts who received Toci, 35/152 (23%) received Toci when CRS is at least grade 2 (Toci/CRS≥2), 49/152 (32%) received Toci/CRS1 without ppx dex (Toci/CRS1), and 14 (9%) received ppx dex with Toci given in CRS1 when pts developed CRS (PPX DEX). Baseline demographics were comparable except age (Toci/CRS≥2: 58y; Toci/ CRS1: 63y; ppx dex: 50y; p=0.001) and pre-LD CRP (Toci/CRS≥2: 34mg/L; Toci/ CRS1: 6mg/L; ppx dex: 16mg/L; p=0.002). Pre-LD LDH and ferritin were also comparable. PPX DEX or Toci/CRS1 had significantly lower rate of maximum grade CRS≥2 (PPX DEX: 4 (29%); Toci/ CRS1: 20 (41%); 35 (100%); p<0.001) and duration of CRS (PPX DEX: 2.5d; Toci/ CRS1: 3d; Toci/CRS≥2: 5d; p<0.001). Time to onset of CRS was not statistically different between the 3 groups. Not surprisingly, Toci/CRS1 group had higher overall rate of toci use, compared to . Interestingly, use rate of ≥2 doses of Toci or siltuximab was significantly higher in Toci/CRS1 or PPX DEX (PPX DEX: 8 (57%); Toci/ CRS1: 26 (53%); 9 (26%); p=0.02). Pt who received PPX DEX or Toci/CRS1 had a significantly lower rate of ICANS (PPX DEX: 6 (43%); Toci/ CRS1: 33 (67%); 32 (91%); p=0.001) and rate of max grade ICANS ≥2 (PPX DEX: 3 (21%); Toci/ CRS1: 23 (47%); 27 (77%); p<0.001). The time to onset and duration of ICANS was not different among the 3 groups. Conclusion: Our real-world study shows that earlier use of Toci in grade 1, compared to its first use in grade 2, can result in lower overall incidence of more severe CRS, duration of CRS, incidence of ICANS and more severe ICANS. While we selected patients at higher risk for having more severe CRS in our PPX DEX cohort, experience to date suggests that these patients, compared to toci/CRS>=2 cohort, could also have reduced severity of CRS and ICANS with this updated management.