Citations

Background: Aggressive non-Hodgkin lymphoma (aNHL) commonly occurs in older adults. However, geriatric assessments (GAs), which are associated with both survival and treatment-related toxicity, are infrequently performed due to logistical constraints. In this study, we examined how the brief, self-reported Vulnerable Elders Survey-13 (VES-13) screening tool and a comprehensive GA relate to clinical outcomes in this population. Methods: We conducted a longitudinal study of 105 adults aged ≥65 years who were initiating upfront chemoimmunotherapy for aNHL at 2 academic medical centers and their community affiliates between September 2020 and January 2023. Baseline frailty was evaluated using the Fondazione Italiana Linfomi (FIL) GA, which classifies patients as fit, unfit, or frail. The VES-13 was assessed both continuously and by categorizing patients as vulnerable (score ≥3). Using logistic regression controlling for covariates, we examined the association of VES-13 with grade ≥3 nonhematologic toxicity (toxicity), unplanned hospitalization within 6 months of treatment initiation, decline in quality of life (QoL) 1 year after therapy initiation, and death within 1 year of treatment initiation. In addition, we calculated the sensitivity and specificity of both the VES-13 and the FIL GA in identifying patients at risk for these outcomes. Results: The median patient age was 73 years (range, 64–99 years), with 42% aged ≥75 years. GA classified 40% of patients as fit, 10% as unfit, and 51% as frail, while 46% of patients had a VES-13 score ≥3, indicating vulnerability. VES-13 score was significantly associated with toxicity (odds ratio [OR], 1.18; P =.026), unplanned hospitalization (OR, 1.16; P =.043), QoL decline at 1 year (OR, 1.22; P =.027), and death within 1 year (OR, 1.49; P =.010). Using the VES-13 to define vulnerability (score ≥3), sensitivity and specificity were 59% and 61% for predicting toxicity, and 100% and 59% for predicting death within 1 year, respectively. FIL GA frailty had a sensitivity of 63% and specificity of 56% for predicting toxicity, and 100% sensitivity and 54% specificity for predicting death within 1 year. Conclusions: We found that VES-13 score was associated with survival, toxicity, health care utilization, and QoL decline, demonstrating similar sensitivity and specificity to the FIL GA. These findings underscore the potential utility of VES-13 as a screening instrument to identify older patients with aNHL at risk for adverse clinical outcomes.

Abstract In a real-world analysis of brexucabtagene autoleucel recipients with relapsed/refractory B-cell acute lymphoblastic leukemia (N = 278), lack of response to prior blinatumomab correlates with significantly worse following chimeric antigen receptor T-cell therapy outcomes.

Abstract Background Staphylococcus aureus is a challenging pathogen which causes a multiplicity of human diseases around the globe, acquired in the community and healthcare settings alike. Treatment of S. aureus infection can be challenging and is further compromised by antimicrobial resistance. Understanding of its pathogenesis, transmission pathways and the dynamics of its genome evolution and associated adaptations is incomplete. One strategy to assess genome evolution is to assess codon bias. Historically, studies have focused on the consequences of coding changes; assessing silent mutations in the form of synonymous codon usage bias represents a novel approach to studying genetic adaptation.Figure 1.Principal component analysis (PCA) (A) and sPLS-DA (B) separate S. aureus from complex mates.sPLS-DA component 1 relies on a hypothetical protein, identified here by its NCTC 8325 genomic coordinates, as important for separating S. aureus from complex mates (C). Methods Here, the codon bias of 2,872 genes was queried against all publicly available complete S. aureus plasmids and genomes. Using public BioSample data, genomes were sorted by disease state (e.g., bacteremia, pulmonary infection) and geographical collection site. Results Isolates from Nairobi, Kenya were most different from other isolates, suggesting that geographical isolation may shape codon bias signature. In addition, codon bias of two MSCRAMM proteins implicated in human disease - SdrC and FnBPB - varied between isolation sources, suggesting a role for codon bias in S. aureus genetic adaptation to selective pressures of infection. Additionally, codon bias out-performed coding and nucleic acid mutational metrics in separating members of the S. aureus complex, S. aureus, Staphylococcus argenteus, Staphylococcus schweitzeri, and Staphylococcus roterodami; Fig 1, sparse partial least squares analysis (sPLS-DA) mean AUC codon: 0.89762, nucleic acid: 0.82266, codon bias: 0.9996 (relying on hypothetical protein NC_007795.1:516850-517419). Conclusion This analysis suggests that codon bias in S. aureus is a source of genetic adaptation that can be used to separate strains by geographical isolation source, and separate S. aureus sensu stricto from other S. aureus complex members. Disclosures Nicholas Bedard, MD, Stryker: Advisor/Consultant Robin Patel, MD, a patent on Bordetella pertussis/parapertussis PCR issued, a patent on a device/method for sonication, a patent on PET imaging of bacterial infection: a patent on Bordetella pertussis/parapertussis PCR issued, a patent on a device/method for sonication, a patent on PET imaging of bacterial infection|MicuRx Pharmaceuticals and bioMérieux: Grant/Research Support|PhAST, Day Zero Diagnostics, DEEPULL DIAGNOSTICS, S.L., Nostics, HealthTrackRx, bioMérieux and CARB-X: Advisor/Consultant|Up-to-Date and the Infectious Diseases Board Review Course: Honoraria

Abstract Background Motor disturbances are common in neurodegenerative disorders. Digital health technologies can be a tool for remote, accessible, and low‐burden motor assessments. The ALLFTD Mobile App includes a finger‐tapping test (mApp‐FTT) to assess dexterity. We evaluated the reliability and construct validity of smartphone‐based measures of finger tapping in an in‐person sample. Methods Participants ( n = 33; mean age=68.2 [SD=8.6]) completed the mobile finger‐tapping test (mApp‐FTT) on iOS and Android devices (two counterbalanced 20‐second trials per hand) and the mechanical finger‐tapping test (FTT). The sample included cognitively unimpaired participants (22.2%) and those diagnosed with frontotemporal dementia‐spectrum disorders (41.7%), Alzheimer's dementia (22.2%), and progressive supranuclear palsy or corticobasal syndrome (13.9%). A motor domain was calculated based on neurological exam and added to the Clinical Dementia Rating® plus National Alzheimer's Coordinating Center–Frontotemporal Lobar Degeneration (CDR®+NACC‐FTLD). Pearson's correlations assessed associations between mApp‐FTT performance and FTT performance, clinical severity (CDR®+NACC‐FTLD Box Score), motor rating scales (Progressive Supranuclear Palsy Rating Scale [PSPRS] and Unified Parkinson's Disease Rating Scale [UPDRS]), and cognitive function. Due to small sample size, PSPRS subdomains were dichotomized, and t‐tests were performed to evaluate group differences based on symptom presence. Results Total mApp‐FTT taps on iOS and Android were strongly correlated ( r = 0.86, 95% CI [0.76, 0.92], p < .001). mApp‐FTT was associated with manual FTT on iOS ( r = 0.70, 95% CI [0.52, 0.82], p < .001) and Android ( r = 0.62, 95% CI [0.41, 0.77], p < .001). Among iOS users, total PSPRS score was negatively correlated with mApp‐FTT ( r = ‐0.60, 95% CI [‐0.75, ‐0.39], p < .001). Worse mApp‐FTT performance was observed in participants with higher limb rigidity (mean difference=15.46, 95% CI [3.40, 27.52], p < .001), limb dystonia (mean difference=19.22, 95% CI [5.67, 32.77], p < .001), and finger tapping (mean difference=12.74, 95% CI [3.42, 22.06], p < .001). Total UPDRS score correlated with lower mApp‐FTT ( r = ‐0.48, 95% CI [‐0.66, ‐0.28], p < .001), as did disease severity (CDR®+NACC‐FTLD Box Score; r=‐0.46, 95% CI [‐0.66, ‐0.22], p < .001). Conclusion Smartphone‐based finger‐tapping assessments effectively detect motor and cognitive impairment, with stronger associations for motor function. Findings in iOS users highlight the need to assess operating system‐related biases and validate in Android. Remote smartphone assessments offer a scalable tool for tracking disease progression, warranting larger longitudinal studies.

PURPOSE To compare patient-reported and clinical outcomes between radical cystectomy (RC) and bladder-sparing therapy (BST) in patients with recurrent high-grade non–muscle-invasive bladder cancer (NMIBC). PATIENTS AND METHODS This pragmatic, prospective observational cohort study was designed with patients, who selected and prioritized outcomes. Eligible adults were candidates for both RC or BST, had previous induction Bacillus Calmette-Guérin (BCG), and received their last treatment within 12 months. The primary outcome was the EORTC-QLQ-C30 physical function scale at 12 months. Secondary outcomes included other EORTC-QLQ-C30 scales, depression, anxiety, bladder cancer–specific quality of life (QOL), financial burden, and cancer-specific outcomes. Targeted maximum likelihood estimation (TMLE) was used to calculate average treatment effect (ATE) estimates between arms. Inverse probability weighted risk ratios (wRR) were calculated using quasi-Poisson regression. RESULTS Of 570 participants (mean age 71.4 years; 21% female), 371 selected BST and 199 selected RC. Physical function was significantly worse in the RC arm at 3 months; by 9 months, there was no difference between arms, and at 12 months, physical function did not differ (ATE, 0.9; 95% CI, –0.6 to 2.4; P = .22). RC was associated with better emotional function, generic health-related QOL, and financial burden, and lower depression and anxiety, while BST was associated with better bowel and sexual health. Cancer-specific survival was 99% for BST versus 96% for RC (wRR, 0.99; 95% CI, 0.97 to 1.01). RC was associated with a higher risk of adverse events and serious adverse events, including a 90-day mortality rate of 2.5%. CONCLUSION Most patient-prioritized outcomes were similar or better among participants who chose RC compared with BST. These findings support the continued role of RC in managing recurrent high-grade NMIBC.

Abstract Introduction Obecabtagene autoleucel (obe-cel) and brexucabtagene autoleucel (brexu-cel) are CD19 targeted chimeric antigen receptor T cell (CAR-T) therapies, approved for the treatment of adults with relapsed/refractory (r/r) B-cell acute lymphoblastic leukemia (ALL). Mechanistic differences between obe-cel and brexu-cel including differing costimulatory domains (4-1BB vs. CD28), CD19 binding domains (intermediate vs. high affinity) and split dose (Days 1 and 10) vs. single infusion may impact in-vivo cellular kinetics that translate into variant clinical outcomes. In pivotal trials, obe-cel resulted in lower rates of severe cytokine release syndrome (CRS) and immune effector cell associated neurotoxicity syndrome (ICANS) compared to brexu-cel with comparable efficacy (Roddie et al., NEJM 2025; Shah et al., Lancet 2021). Real-world utilization and outcomes with obe-cel are unknown given the relatively recent approval. Methods The ROCCA database, comprising real world data from patients (pts) with r/r ALL treated at 40 North American institutions was utilized in this analysis. Pts with r/r ALL were eligible if they were apheresed for obe-cel since its approval (11/8/2024) or brexu-cel over a comparable period (since 8/1/24) and had at least 30 days of follow up. Data cut off was 7/15/2025. CRS/ICANS were graded per ASTCT criteria. Measurable residual disease (MRD) was assessed by flow cytometry and/or next generation sequencing per institutional standards. Results 38 pts have undergone apheresis for obe-cel (36 infused, all received both infusions) and 54 (53 infused) for brexu-cel over the study period. Pts in the obe-cel and brexu-cel cohorts had the following baseline characteristics: median age, 47 vs. 39 years; ECOG 2+, 15% vs. 26%; Ph+ disease, 24% vs 13%; KMT2Ar, 0 vs. 4%; TP53m, 11% vs 11%, and hypodiploid karyotype, 5% vs 6%. The following prior treatment patterns were observed: median lines of therapy before CAR-T 3 vs. 3; prior SCT 24% vs. 30%, prior blinatumomab 68% vs. 46%, prior inotuzumab 45% vs. 30%. At apheresis, active disease (&gt;5% bone marrow blasts) was present in 47% vs. 55% of obe-cel and brexu-cel pts, respectively; the remainder were in CR (with 18% vs. 22% in MRD- CR, respectively). Among obe-cel (n= 35 [92% of cohort]) and brexu-cel (21 [38% of cohort] pts with disease re-assessment prior to lymphodepletion (LD), 31% vs. 43% had active BM disease, respectively; few pts (24% vs. 17%) had &gt;20% marrow disease burden prior to LD. CAR-mediated toxicity differed significantly between the cohorts. CRS occurred in 56% of obe-cel pts compared to 94% of brexu-cel pts (p &lt; 0.0001). There were no Gr3+ CRS events among the obe-cel pts; 3 (6%) brexu-cel pts had Gr3+ CRS (p = 0.27). ICANS occurred in 17% of obe-cel pts vs. 51% of brexu-cel pts (p = 0.001). Gr3+ ICANS occurred in 6% of obe-cel vs. 32% of brexu-cel pts (p = 0.0027). Among the obe-cel pts, CRS occurred in 31% after the first infusion and 46% after the second; ICANS occurred in 3% after the first infusion and 15% after the second. Prolonged Gr4 neutropenia (ANC &lt; 500 cells/uL beyond day 30 from infusion) occurred in 24% of obe-cel vs. 28% of brexu-cel pts (p = 0.73). Deaths within the first 28 days of infusion occurred in 0 obe-cel pts and 4 brexu-cel pts (2 of infection, 1 of infection/brain bleed, and 1 of liver failure in the setting of Gr4 CRS and HLH). Response rates were high and did not significantly differ between cohorts (p = 0.85). Among the 31 obe-cel pts with available response data, 25 (81%) achieved an MRD- CR/CRi, 3 (10%) had an MRD+ CR/CRi, 2 (6%) had a CR/CRi with unknown MRD, and 1 (3%) did not respond. Among 41 brexu-cel pts with available response data, 33 (80%) had an MRD- CR/CRi, 4 (10%) an MRD+ CR/CRi, and 4 (10%) CR/CRi with unknown MRD. Conclusion Pts selected for obe-cel apheresis were similar to those for brexu-cel over the study period (noting that not all centers had access to obe-cel during this time). Similar to clinical trial results, obe-cel was associated with lower rates of CRS/ICANS. Rates of MRD-negative CR were high and did not differ between cohorts. A larger sample and longer follow up are required for further analyses; we anticipate a cohort of ~75 obe-cel treated pts by the annual meeting and will provide updated data.

Abstract Introduction: Brexucabtagene autoleucel (Brexu-cel) is a CD19-targeted chimeric antigen receptor (CAR-)T cell therapy approved for treatment of relapsed/refractory (r/r) B-ALL. Brexu-cel's approval was founded on data from the ZUMA-3 trial, in which 78 patients (pts) were treated at the pivotal dose of 1x10e6 CAR-T cells/kg, resulting in a complete remission (CR) rate of 73%. Three-year follow-up from ZUMA-3 demonstrated median relapse-free and overall survival (RFS/OS) of 11.6 and 25.6 months (mos) for all treated pts. Amongst responders, the median OS was 60.4 mos for those who did not receive a subsequent allogeneic hematopoietic cell transplant (HCT) (n=43), but only 36.3 months for those who did (n=14). This has led some to conclude that HCT may not be beneficial following brexu-cel therapy for r/r B-ALL; however, it must be noted that 29 (37%) of the pts treated on ZUMA-3 received brexu-cel for post-HCT relapse. T cells collected from individuals who relapse in the post-HCT setting are largely of donor origin, making the CAR-T product comparable to a genetically modified donor lymphocyte infusion (DLI). The immunology at play in this scenario may be different from that in which CAR-T cells are of truly autologous origin. Methods: The Real-World Outcomes Collaborative for CAR-T in ALL (ROCCA) consortium database includes pt characteristics and outcomes of brexu-cel therapy given as standard-of-care across 41 US institutions. We identified pts who achieved an MRD negative (neg) CR by either flow cytometry (minimum sensitivity 10e-4) or NGS following brexu-cel for r/r Ph neg B-ALL and categorized them into three groups with regard to use of HCT: Never HCT, Pre-CAR HCT, or Post-CAR HCT. Pts who received second HCT following brexu-cel for post-HCT relapse (n=9) were excluded. Pt/disease characteristics, RFS, and OS were assessed for each group, landmarked at 73 days after brexu-cel infusion (the median time to consolidative allogeneic HCT). Results: In the ROCCA database we identified 165 pts who achieved MRD neg CR following brexu-cel, amongst whom 69 never underwent HCT, 49 were transplanted pre-CAR (ie, brexu-cel served as a genetically modified DLI), and 47 underwent post-CAR consolidative HCT while in CR. The median follow-up is 373 days. The median age was higher in the Never HCT group (49, range 19-83) in comparison with the Pre-CAR HCT (40, 24-72) and Post-CAR HCT (34, 18-70) groups. Thirty-five percent were &gt;60yo in the Never HCT group vs 20% in the Pre-CAR HCT and 9% in the Post-CAR HCT groups. Ph-like disease was present in 38%, 29%, and 43% of the Never HCT, Pre-CAR HCT, and Post-CAR HCT groups, respectively (p=0.35). No significant differences were observed in pre-apheresis bone marrow blasts, presence of CNS disease at apheresis, prior blinatumomab or inotuzumab treatment, or use of bridging therapy across the groups, though there was a higher prevalence of non-CNS extramedullary disease in the Pre- and Post-CAR HCT groups (32, 39%) vs the Never HCT group (13%) (p=0.004). A minority of pts in all groups were deemed to be in MRD neg CR prior to apheresis (6-20%). Median RFS was 260 days in the Never HCT group and not reached in the Pre- and Post-CAR HCT groups (p=0.002), with respective 1-yr RFS of 38%, 64%, and 67%. Median OS was not reached in any group and was similar for all at 1-yr at 68-79% (p=0.13). Characteristics of those in the Never HCT group who were least likely to relapse included age &gt;60 (OR 0.22, CI 0.07-0.73 vs age 18-39, p=0.024) and those with non-Ph-like B-ALL (OR 0.33, CI 0.12-0.94, vs Ph-like, p=0.037) in univariate analyses. Conclusions: Real-world data regarding outcomes following brexu-cel therapy allow evaluation of the role for consolidative allogeneic HCT, which remains one of the most important unresolved questions in the use of CAR-T therapy for r/r B-ALL. This analysis demonstrates that a first consolidative HCT after achieving MRD neg CR with brexu-cel is associated with a significant decrease in relapse. Additionally, use of brexu-cel for post-HCT relapse is also associated with a significantly lower risk of relapse compared with those who never received an allogeneic HCT. With the current duration of followup, OS is not significantly different across the three groups. These data suggest there is a benefit for relapse-free survival in pts receiving an allogeneic HCT at some point in the course of treatment for r/r Ph neg B-ALL.

Abstract While the use of NCCN guidelines is essential in lymphoma care, adherence among oncologists remains highly variable. Real-world concordance ranges from 65–80% overall, dropping to as low as 11–46% for certain lymphoma subtypes. Although 88% of clinicians report using NCCN guidelines for Hodgkin lymphoma, adherence to PET-adapted approaches is inconsistent. This gap highlights a critical need for tools that support guideline-concordant care.Large language models (LLMs) such as GPT–4 promise decision support, yet hallucinations and divergence from evidence-based care remain obstacles. We assessed whether retrieval-augmented generation (RAG) with National Comprehensive Cancer Network (NCCN) guidelines improves GPT–4 performance in lymphoma treatment recommendation tasks. To address this, we developed a RAG-enhanced GPT-4 agent (RAG-GPT) by indexing the 2025 NCCN Lymphoma Guidelines into a vector database. Fifty clinical vignettes spanning diverse lymphoma subtypes were compiled from published case reports; this sample provides 80 % power (α = 0.05) to detect a mean score difference of ≥ 0.18 on the modified Generative AI Performance Score (mG-PS). Guideline concordance was assessed based on NCCN guidelines and clinical judgment reflecting real-world oncological decision-making. Scores were normalized between -1.0 and 1.0 for consistent evaluation. Readability and rationale clarity were rated on a 5-point Likert scale. Baseline GPT-4 and RAG-GPT generated treatment plans via standardized prompts. An oncologist scored all outputs with mG-PS, which rates NCCN concordance (0.0–1.0) and penalizes hallucinations (-0.25 to –1.0). RAG-GPT achieved a higher mean mG-PS (0.32) than baseline GPT–4 (0.10) (t(98)=-2.11, p=0.038; mean score difference ΔmG-PS=0.22). Importantly, RAG-GPT eliminated severe hallucinations, reinforcing confidence in the safety of its treatment reccomendations. Fully concordant “gold-standard” plans were produced in 46 % of vignettes with RAG-GPT versus 22 % with baseline. Readability scores favored RAG-GPT modestly without statistical significance. Lymphoma's bimodal age distribution, heterogeneity, and lack of standardized treatment algorithms, coupled with regimen toxicity and the need to account for patient frailty and comorbidities, create complex clinical scenarios that constrain LLM applicability for nuanced decision-making. Non-Hodgkin lymphoma, particularly relapsed/refractory, transformed disease, or rare subtypes, lacks a defined standard of care, leading reviewers to rely on clinical judgment or published case series beyond NCCN guidelines. This underscores a key limitation of LLM-supported tools: their performance is highly dependent on the presence of clear algorithmic consensus within a given disease. Linking GPT-4 to NCCN guidelines via RAG significantly improved treatment accuracy and reduced hallucination severity in lymphoma, though performance remained lower than prior RAG studies in cancers like breast and CNS suggesting a greater impact on hallucination reduction than on retrieval of a consistent gold standard regimen. These findings support continued development and prospective validation of guideline-anchored LLMs for clinical use.

Abstract Introduction: CD19-directed CAR T-cell (CAR-T) therapy has substantially improved clinical outcomes for adults with relapsed or refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL). However, notable toxicities are associated with CAR-T, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Variable practice patterns have emerged regarding the use of CRS/ICANS prophylaxis (ppx) via corticosteroids and/or other pharmacologic agents, but the utility of these approaches on reducing CRS or ICANS following CAR-T in B-ALL remains unknown. Thus, we utilized the Real-World Outcomes Collaborative for CAR T in ALL (ROCCA) database to evaluate associations between ppx and the development and severity of CRS or ICANS following brexucabtagene autoleucel (brexu-cel) for B-ALL. Methods: Adult patients (pts) who received standard-of-care brexu-cel for relapsed/refractory B-ALL across 40+ US centers were included. Primary endpoints were the incidence of CRS and ICANS within 28 days of CAR-T infusion. Secondary endpoints included timing of CRS and ICANS onset, progression-free survival (PFS) and overall survival (OS). Univariable and multivariable logistic regression evaluated associations between ppx and incidence of CRS/ICANS; KM and log-rank methods were used for survival analyses. Results: A total of 378 pts were included: 53 pts (14.0%) received CRS/ICANS ppx and 325 (86.0%) received no ppx. Ppx agents included steroids only (26.4%), steroid and anakinra (52.8%), anakinra only (15.1%), and steroid and tocilizumab (5.7%). Pts in the ppx group had a higher proportion of Hispanic pts (64% vs 34%), Ph-like (36% vs 19%) and a lower proportion had 0-5% bone marrow blasts at apheresis (38% vs 53%). The ppx and no ppx groups were similar in age, prior therapies, CAR-T indication, and marrow blasts &gt;50%. Cumulative incidence of any grade (gr) CRS was 94% with ppx and 86% with no ppx (p=0.025), of any gr ICANS was 53% with ppx and 51% with no ppx (0= 0.85). Cumulative incidence of gr 3-4 CRS was 9% in the ppx group and 10% in the no ppx group (p=1.00), of gr 3-4 ICANS was 26% in the ppx group and 29% in the no ppx group (p=0.71). Median time to onset for any gr CRS or any gr ICANS was identical in the ppx and no ppx cohorts (CRS: 5 days, range ppx 1-9 days, no ppx 0-14 days; ICANS: 7 days, ppx 1-20 days, no ppx 0-16 days). Univariate logistic regression analyses showed no significant association between ppx and the incidence of gr 2-4 CRS (OR 1.34, 95% CI: 0.75-2.40, p=0.326) or gr 3-4 ICANS (OR 0.88, 95% CI: 0.46-1.70, p=0.708). Multivariate analysis adjusting for ethnicity, leukemia genetic sub-type, and pre-apheresis disease burden similarly found no significant associations between ppx and gr 2-4 CRS (adjusted OR 1.06, 95% CI: 0.55-2.01, p=0.870) or gr 3-4 ICANS (OR 1.23, 95% CI: 0.59-2.55, p=0.584). When comparing steroid-only ppx (N=14) to steroid combinations or other drug ppx (N=39), there was no significant differences in gr 2-4 CRS (57% vs 44%, p=0.345) or gr 3-4 ICANS (43% vs 21%, p=0.115). Comparing cumulative incidence of dexamethasone (dex) exposure, pts in the ppx group had numerically greater dex exposure than pts in the no ppx group, with median (range) dosing of 182 mg (10-1386mg) and 100mg (1-4000mg), respectively. Finally, we found no significant differences in PFS or OS based on ppx. Median PFS was 9.6 months in the ppx group vs 13.3 months in the no ppx group (p=0.646) and median OS was 17.2 months in the ppx group vs 34.4 months in the no ppx group (p=0.170). Conclusions: In this large cohort of adult patients with R/R ALL receiving brexu-cel CAR-T therapy, prophylaxis intended to prevent CRS/ICANS was ineffective at reducing CRS/ICANS incidence, time to onset, or overall steroid exposure. Furthermore, PFS and OS were comparable between patients who received prophylaxis and those who did not. In the absence of prospective trials in B-ALL, these data suggest that current strategies of CRS/ICANS prophylaxis has limited benefit.

Abstract Introduction: Brexu-cel is a highly effective treatment for R/R B-ALL, with complete responses achieved in 71% of patients (pts) treated in the pivotal ZUMA-3 trial. Patients with B-ALL with high-risk (HR) genetics are known to have inferior outcomes with chemotherapy, but little is known about the influence of HR ALL genetics on responses to chimeric antigen receptor (CAR)-T in adults. In this study, we assessed the impact of HR ALL genetics on outcomes in a large cohort of adult patients treated with brexu-cel for R/R Philadelphia chromosome (Ph)-negative B-ALL. Methods: Patient characteristics and outcomes were obtained from the Real-World Outcomes Collaborative of CAR-T in Adult ALL (ROCCA) database which includes data from 40 U.S. centers. NCCN guidelines (NCCN Guidelines in Oncology, ALL v2.2025) were used to define HR B-ALL genetics; lack of HR genetics was defined as standard-risk (SR). The primary outcome was event-free survival (EFS), defined as lack of response, disease progression, or death from any cause. Secondary outcomes were overall survival (OS), relapse free survival (RFS) and response rates including complete response (CR), CR incomplete count recovery (CRi), measurable residual disease negative CR (MRD-) by flow or NGS and progressive disease (PD). Multivariable models were adjusted for age, pre-infusion blast percentage, prior allo-HCT, and post CAR-T allo-HCT (modeled as a time-varying covariate). Results: The cohort included 286 pts: 169 with HR and 117 with SR B-ALL. The HR cohort included TP53m (n=24, 14%), KMT2Ar (n=16, 9%), CRLF2r (n=65, 39%) and other HR genetics such as JAK2 and other Ph-like, IKZF1, complex cytogenetics (n=64, 38%). Relative to SR pts, HR pts were more commonly Hispanic (50% vs 28%). HR and SR pts had similar disease burden pre-apheresis (&gt;5% blasts: 58% vs 53%), prior blinatumomab (60% vs 58%), prior inotuzumab (42% vs 46%), and prior allo HCT (27% vs 37%). The presence of CNS disease (11% HR vs 12% SR) and extramedullary disease (EMD, 22% HR vs 29% SR) at apheresis was also similar. CR/CRi was achieved in 90.5% of evaluable HR pts (n=147) and 90.2% of evaluable SR pts (n=102); MRD- rates were also similar between HR and SR pts at 78.9% and 78.2% respectively. In unadjusted analyses, EFS did not significantly differ between HR pts (median 9.6 mos; 1-yr: 46%, 95% CI 38%-54%) and SR pts (median 11.6 mos; 1-yr: 48%, 95% CI 38%- 58%). Median RFS (HR: 12.2 mos, SR: 13.3 mos) and median OS (HR: 22.4 mos, SR: 21.0 mos) were also similar. Multivariable models demonstrated no significant associations between HR B-ALL and EFS (HR 1.09, 95% CI 0.73-1.63), RFS (HR 1.13, 95% CI 0.71-1.80), or OS (HR 0.96, 95% CI 0.62- 1.48) relative to SR ALL. 28% of HR patients and 21% of SR patients underwent consolidative allo-HCT. Among the HR subgroup, CR/CRi rate was lowest in KMT2Ar (71%) relative to TP53m (95%), CRLF2r (95%), and other HR genetics (89%). The median EFS was lowest in KMT2Ar (4.1 mos), followed by TP53m (6.9 mos), CRLF2r (11.6 mos), and other HR (11.1 mos). Median OS was also lowest in KMT2Ar (6.8 mos) vs 22.4 mos in TP53m and not reached in CRLF2r or other HR. PD was highest in KMT2Ar pts (29%), as compared to 11% HR other, 10% SR, and 5% TP53 and CRLF2r. The rate of relapse among those that achieved CR across the groups was similar and ranged between 28% -40%. Rates of CD19 negative relapse were similar between all subgroups (13%-15%), however 21% of KMT2A pts underwent myeloid transformation as compared 0% for TP53 and CRLF2 pts, and 2% among other HR and SR pts. 25% of KMT2Ar patients underwent consolidative allo-HCT as compared to 32% CRLF2 pts, 29% TP53 pts, 25% other HR pts and 21% SR pts. Multivariable models evaluating the different genomic subgroups demonstrated that relative to SR B-ALL, only KMT2Ar was associated with inferior OS (HR 3.17, 95% CI 1.53-6.58, p=0020). No associations were seen between HR subgroups and EFS. Conclusion: To our knowledge, this analysis represents the largest examination of adult B-ALL recipients of CAR-T assessed by genomic risk group. We found that response rates and survival outcomes following brexu-cel were similar among HR and SR B-ALL pts, including pts with TP53m and CRLF2r. However, pts with KMT2Ar had lower response rates and dismal EFS and OS, with higher rates of PD and myeloid transformation at relapse, suggesting that alternative strategies are necessary to improve outcomes for patients with R/R KMT2Ar B-ALL.