Citations

Abstract Background Although myeloproliferative neoplasm-unclassifiable (MPN-U) stands as a distinct entity in the International Consensus Classification (ICC) for myeloproliferative neoplasms, its diagnosis and management remain challenging due to the overlapping clinico-pathological characteristics with other MPNs. In the current study, we describe a well-defined cohort of patients with MPN-U and compare their presentation and outcome with those of a separate cohort with essential thrombocythemia (ET). Methods Diagnostic criteria were according to the ICC (Blood 2022; 140:1200). Non-driver mutations were screened by next-generation sequencing (NGS) and were available for subsets of patients with MPN-U or ET. Conventional statistical methods were employed, using JMP Pro 18.0.0 software SAS Institute, Cary, NC, USA. Results Thirty Mayo Clinic patients who met the ICC criteria for MPN-U (median age 64 years; range 33-94; 60% males) and 658 with ET (median age 60 years; range 18-90; 36% males) were included in the current study. Table 1 outlines a comparative list of presenting characteristics (including mutations), clinical outcomes, and treatment details between patients with MPN-U vs. ET. JAK2 mutations were significantly more prevalent in MPN-U cohort (87% vs. 63% in ET; p<0.01) and CALR mutations less common (3% vs. 25%; p<0.01). Other somatic mutations, including ASXL1, SRSF2, and EZH2 were more frequently observed in the MPN-U group (p<0.05 in all). Abnormal karyotype was reported in 14% of patients with MPN-U vs. 8% with ET (p=0.3). Compared to patients with ET, those with MPN-U were more likely to be males (60% vs. 36%; p<0.01) and, as expected from disease presentation and indications for diagnostic work up of suspected cases, present with venous thrombosis (43% vs. 10%; p<0.01) and palpable splenomegaly (40% vs. 8.5%; p<0.01). Similarly, they were less likely to display thrombocytosis or leukocytosis (p<0.01 for both). Portal, superior mesenteric, and hepatic vein thromboses constituted 77% (vs. 27% in ET; p<0.01) of the venous events in MPN-U; in contrast, the incidences of pre-diagnosis arterial events were similar between the two groups (10% vs. 12%; p=0.7). At a median follow-up of 2.8 years in the MPN-U cohort, 6 (20%) deaths and 1 (3%) leukemic transformation were documented while 3 (10%) patients progressed to meet the diagnostic criteria for polycythemia vera. Four (13%) patients with MPN-U experienced venous thrombo-embolic events (VTEs) including 2 with recurrent VTEs, despite active therapy with systemic anticoagulation and aspirin. The ET cohort was followed for a median of 7.3 years with 161 (24%) deaths, 23 (3%) leukemic transformations, and 66 (10%) VTEs. Among the latter, 27 (41%) had recurrent episodes despite treatment with systemic anticoagulation, aspirin, and/or cytoreductive drugs. Notably, patients with MPN-U were less likely to receive aspirin and cytoreductive therapy during the clinical course (77% vs. 91% and 37% vs. 79%; p<0.01 for both, respectively). Long-term survival rates were similar between MPN-U and ET (15-year survival 70% vs. 65%; p=0.1). Consistent with previous associations seen in ET, patients with JAK2-mutated MPN-U were more likely to be males (58% vs. 30%; p<0.01, Table 2), and display a lower platelet and leukocyte count at the time of diagnosis (p<0.01 for both). These patients were also more likely to present with venous thrombosis (50% vs. 12% in JAK2 wild-type cases; p<0.01) and palpable splenomegaly (46% vs. 9%; p<0.01). Of note, the incidences of pre- and post-diagnosis arterial and post-diagnosis venous events were similar between JAK2-mutated MPN-U vs. JAK2-mutated ET patients (12% vs. 14%; p=0.7, 4% vs. 12.5%; p=0.1, and 15% vs. 12%, p=0.6, respectively). Similarly, long-term outcomes including deaths (15% vs. 26%; p=0.2), leukemic transformations (4% vs. 4%; p=0.9), and 15-year survival rates (78% vs. 60%; p=0.9) were also comparable between the two groups. Conclusion The findings from the current study were not unexpected and underline the ongoing challenges that face formal classification systems in formulating accurate diagnostic criteria for patients with MPN-U whose clinical presentation is biased by disease definitions. Regardless, it is comforting to document a favorable long-term outcome that appeared to be similar to that of ET while venous thrombosis remains a treatment challenge in both groups.

Abstract Introduction: The epidemiology and clinical characteristics of familial myelodysplastic syndromes/acute myeloid leukemia (MDS/AML) remain poorly understood, particularly among Native Hawaiians and Other Pacific Islanders (NHOPI). Methods: To investigate the real-world evidence in this area, we conducted a retrospective study at the Queen's Medical Center (QMC), Hawaii's largest tertiary referral hospital. The data was gathered from QMC's electronic medical record. Familial MDS or familial AML patients in this study were defined as MDS or AML occurring in a patient with a family history of MDS and/or AML. Patients with acute promyelocytic leukemia (APL), as well as those with a family history of APL, were excluded given the differences in disease characteristics and its genetic phenotype from other AML subtypes. Unspecified leukemia was defined when a patient had a family history of leukemia but the specific subtype was unavailable. Some patients underwent comprehensive panel testing, while others had targeted next-generation sequencing (NGS) assays ordered at the discretion of their treating physicians. NGS was performed at Clinical Laboratory Improvement Amendments (CLIA)-certified commercial laboratories. NGS panels typically included the following genes: ASXL1, BCOR, BRAF, CEBPA, CSF3R, DNMT3A, ETV6, EZH2, FLT3, HRAS, IDH1, IDH2, JAK2, KIT, KRAS, MLL, NPM1, NRAS, PDGFRA, PHF6, PML, PTPN11, RUNX1, SETBP1, STAG2, TET2, TP53, and WT1. Results: Among 1,686 MDS/AML patients presented between January 2012 and July 2023, 12 (0.71 %) had familial MDS/AML and 25 (1.48%) had a family history of unspecified leukemia. While no NHOPI were identified among familial MDS/AML cases, 20% of patients with a family history of unspecified leukemia were NHOPI. The median age at diagnosis of familial MDS/AML was 70 years, and 50% of familial MDS patients had MDS with low blasts based on the World Health Organization 2022 classification. Based on International Consensus Classification 2022 criteria, 25% had MDS, not otherwise specified (NOS) with multilineage dysplasia, and 25% had MDS, NOS with single lineage dysplasia. Among the eight patients with familial MDS, the median overall survival (OS) was 12.7 years, with an estimated 5-year OS rate of 75.0% (95% CI: 12.8–96.1), based on Kaplan–Meier analysis. In contrast, the three patients with familial AML had a median survival of 10.2 months and a 3-year OS estimate of 33.3% (95% CI: 0.90–77.4). For patients with a family history of unspecified leukemia, the median survival was not reached during the follow-up period. The estimated 5-year OS rate was 53.3% (95% CI: 12.5–82.7) for the 10 familial MDS cases and 53.3% (95% CI: 20.9–77.3) for the 13 familial AML cases. When combining all familial cases, the median OS for familial MDS (n = 18) was 12.7 years with a 5-year OS rate of 61.9% (95% CI: 26.8–84.0). The median OS for familial AML (n = 16) was 0.90 years, and the estimated 5-year OS rate was 48.2% (95% CI: 20.8–71.3). When all cases were combined, there was no statistically significant difference in overall survival between patients with familial AML (n = 16) and those with familial MDS (n = 18), based on the log-rank test (p = 0.22). Regarding NGS, ASXL1 mutations were the most frequent in the cohort, detected in six of the 32 NGS tests performed (18.8%). Among patients with familial MDS, recurrent somatic mutations were observed in several genes. ASXL1 mutations were identified in two patients; one had a brother with MDS, and the other had a maternal uncle with unspecified leukemia. DNMT3A mutations were also found in two patients with familial MDS; one had a brother with MDS, and the other had a paternal uncle with MDS. SRSF2 mutations were detected in two patients; one had a brother with MDS, and the other had a granddaughter with leukemia. Additionally, SF3B1 mutations were found in two patients; one had a mother with MDS, and the other had a paternal uncle with MDS. Two patients with familial AML were found to have ASXL1 mutation, both with fathers who had unspecified leukemia. Otherwise, no other repeated mutations were found. None of the cases involved NHOPI patients. Conclusion: Our study offers new insights into familial MDS/AML within Hawaii's multiethnic population. Although no NHOPI patients were identified among the familial cases, our findings underscore the genetic heterogeneity and clinical variability of these disorders.

Abstract Introduction Surface proteins in cancer not only reflect tumor–microenvironment interactions but also serve as targets for diagnostic flow cytometry and precision therapies, including monoclonal antibodies and CAR-T cell therapies. A better understanding of plasma cell immunophenotypes could inform flow cytometric prognostic assays and development of targeted therapies. Methods RNA-seq and Seq-FISH data from CD138-enriched bone marrow plasma cells, along with corresponding clinical annotations, were obtained from the Multiple Myeloma Research Foundation-CoMMpass Study. A curated list of surface genes was assembled by integrating three sources: the Human Protein Atlas (plasma membrane and cell junction annotations), the Gene-Set Enrichment Analysis membrane-associated gene set (Ontology:0016020), and the Cell Surface Protein Atlas (validated high-confidence category). A total of 4,579 genes coding for cell surface proteins were identified. Genes below the 25th variance percentile were removed. PCA (≥90% variance) followed by k-means (k=2, silhouette) and confirmed by hierarchical clustering. A Plasma Cell Maturity (PBM) score was computed using singscore on 84 mature plasma cell–associated genes to compare cluster phenotypes. Pathway analysis was done using GSEA software for Hallmark pathways. Clinical biomarkers and survival kinetics were compared between the clusters. Results 761 newly diagnosed patients were included. (Age: 27-93 years, median overall survival/OS: 29 months/mo). Clustering revealed two groups which we termed Atypical/Aggressive Cluster 1 (C1, n=105, 13.8%) and Classical Cluster 2 (C2, n=656, 86.2%), based on the patterns and clinical characteristics. Among non-missing cases, high-risk disease per the International Myeloma Society–International Myeloma Working Group (IMS–IMWG) definition was observed in 27.7% of patients in C2 (n = 552) and 24.7% in C1 (n = 90) (p = 0.4).C1 exhibited enrichment for t(14;16)-MAF (OR = 2.6, 95% CI: 1.1–6.1, p = 0.04) and t(11;14)-CCND1 (OR = 1.7, CI: 1.01–2.8, p = 0.04). In contrast, hyperdiploidy was more common in C2 (OR for C1 = 0.6, CI: 0.4–0.9, p = 0.04). Elevated LDH was observed in C1 (OR = 2.6, CI: 1.6–4, p < 0.001). Median serum Beta-2 microglobulin was lower in C1.(p=0.009) Upregulated genes in C1 in decreasing order of log Fold Change(logFC): AHNAK, MT-ND1, EEF2, EZR, HLA-B, TXNIP, HLA-E, TRIB1, MCL1, RHOB, CD74, CFLAR, HLA-A, HERPUD1, RPSA, ERN1, HSP90AB1, RAPGEF2, ADGRE5, and LAPTM5. Downregulated genes in C1 in descending order of |logFC|: B2M, SDC1 (CD138), HSP90B1, SEL1L, CALR, CANX, PECAM1, SLAMF7, STT3A, TNFRSF17 (BCMA), FCRL5, STT3B, PALM2AKAP2, CD53, PARP1, CCR2, RPN1, CD38, EDNRB, and LMAN2. Upregulated pathways in C1: proliferation (MYC, E2F, PI3K/AKT/mTOR, KRAS, Wnt); evasion of apoptosis (p53, TGF-β, Notch, NF-κB, IL6/JAK/STAT3); Epithelial to Mesenchymal Transition (angiogenesis, apical junction); inflammatory (interferon, TNF-α, complement, IL2/STAT5); metabolic reprogramming (oxidative phosphorylation, lipid/cholesterol); and stress (DNA repair, unfolded protein response). The PBM score for C1 was significantly lower, suggesting a less differentiated phenotype. C1 patients had inferior survival (n = 93, median OS 43.9 mo, p = 0.04) compared to those in the C2 cluster (n = 559, median OS 102.5 mo). Patients in the C1 + IMS–IMWG high-risk status had the shortest survival (20.7 mo), compared to C1+low (50 mo), C2/high (68.3 mo), and C2+low (not reached). In a multivariable analysis, C1 status (HR = 1.93; 95% CI: 1.1–3.4, p = 0.02) was an independent predictor, along with older age (1.05), UAMS70 (1.02), EMC-92 (1.4). In C1, early (<18 mo) progression occurred in 76.3% (n=71) (p < 0.001). In patients receiving first line of treatment(LOT1), median PFS was 39.7 mo in C1vs. 47.3 mo in C2 (p=0.8). Among patients who received stem cell transplant in LOT1, the median survival for the C2 was 102.5 mo, compared to 44.5 mo for C1 (n = 52) (p = 0.01). Among non-SCT patients on Immunomodulator + Proteasome Inhibitor therapy, C1 (n=28) had shorter survival of 50.0 vs. 71.7 C2 (n=137; p =0.8). Conclusion Our transcriptomic analysis identified a higher-risk group of patients (C1) exhibiting downregulation of myeloma specific markers (CD38, CD138, BCMA, SLAMF7, FCRL5) and inferior outcomes. It is important to diagnose and find appropriate therapy for these patients as they don't obtain the same benefit from the current standard of therapy as the remaining patients.

Abstract Overview: The emergence of effective immune therapies for multiple myeloma has highlighted the importance of understanding the roles of less abundant components within the tumor microenvironment. Single-cell analyses provide snapshots of the functional relationships of individual cells at precise moments in time, as well as insight into the overall cellular composition. By integrating single-cell datasets from multiple studies, we improve our ability to identify trends within rare cell populations, such as macrophages, and explore their clinical implications. We investigate human data here because our prior research of murine models indicated that macrophages expressing C1q, particularly those co-expressing Fcgr4, tend to correlate positively with increased tumor burden. Methods: We combined 56 human bone marrow samples – primarily collected at baseline, post-induction, or post-ASCT – with three publicly available data sets: GSE210079 (post CAR-T bone marrow microenvironment); GSE223060 (single-cell analysis of the MMRF cohort); and GSE271107 (disease progression from healthy donors and MGUS to SMM and NDMM). These data sets were integrated using Seurat to generate robust cellular clusters from over 300,000 cells across 117 samples. We validated our single-cell findings with bulk Affymetrix, whole bone marrow gene expression data (collected at baseline and post-baseline) from the Total Therapy clinical trials (GSE136324). Results: A distinct cluster of C1QA/B/C-expressing macrophages was consistently identified in all single-cell cohorts. These cells co-express FCER1G, CSF1R, and CD68 (markers also expressed in murine C1Q macrophages) and typically cluster adjacent to FCGR3A-expressing monocytes. Within the disease progression cohort, C1Q macrophages were found in all phases of progression: healthy donors, MGUS, SMM, and NDMM. However, C1Q macrophages in progressive disease (SMM and NDMM) have distinct expression features including lower levels of MRC1 and higher levels of APOBEC3A and ISG15 – genes linked to inflammatory responses. Among 11 post-induction samples submitted to lenalidomide maintenance, five with shorter responses displayed elevated expression of APOBEC3A and ISG15 in C1Q-expressing macrophages compared to six samples with longer responses. In a cohort of 22 samples 100 days post-ASCT, seven poor responders showed both an increased frequency of C1Q macrophages and elevated APOBEC3A expression compared to good responders. The MMRF cohort indicated that higher baseline frequencies of C1Q macrophages often preceded rapid disease progression (e.g. MMRF_1720, MMRF_1641). Anecdotal evidence from the CAR-T therapy cohort suggests that expansion of C1Q macrophages could serve as an early warning for tumor recurrence; sample 16 showed retention and expansion of C1Q macrophages 28 days post-CAR-T, that was subsequently followed by increased tumor cell counts at 3 months post-therapy. Analysis of bulk, whole bone marrow gene expression revealed that elevated levels of C1Q genes correlated with poorer outcomes, particularly in post-baseline samples. Multivariate Cox survival analysis revealed that high C1Q expression independently predicted inferior overall survival in post-baseline samples after adjusting for age, ISS stage, high levels of tumor-associated markers (SDC1, JCHAIN, DERL3), and high-risk UAMS7 subtypes (MF, MS, PR). Moreover, post-baseline GEP70 risk scores (from matched bone marrow aspirates) positively correlated with elevated C1Q expression in whole bone marrow samples. Conclusion: Both single-cell and bulk analyses indicate that the persistence of immunosuppressive C1Q macrophages following therapy is associated with inferior clinical outcomes. Although the precise mechanisms are yet to be fully understood, the close proximity of similar macrophages to tumor cells–as observed in other cancers–likely also occurs in myeloma and reflects the crucial relationship between tumor cells and macrophages within the bone marrow. While C1Q macrophages are typically rare in individual samples, analysis of combined cohorts is building evidence of their crucial role in immune cross-talk and therapeutic resistance mechanisms.

Abstract Introduction Ciltacabtagene autoleucel (cilta-cel) has demonstrated unprecedented response rates and progression-free survival (PFS) in relapsed/refractory multiple myeloma (RRMM), but a proportion of patients experience early relapse. Identifying these patients is essential to guide future therapeutic strategies, such as enhanced bridging therapy or consolidation, while avoiding overtreatment in those likely to achieve durable disease control. Methods We conducted a multi-center study of patients treated with cilta-cel at 15 U.S. academic centers. Outcomes were evaluated based on high-risk (HR) features, including functional HR disease (FHR; relapse within 18 months of first-line therapy initiation), true extramedullary disease (EMD), traditional HR cytogenetics as defined in CARTITUDE-1 [deletion 17p, t(4;14), t(14;16)], and a 2024 IMS-IMWG-adapted HR definition (deletion 17p; t(4;14) or t(14;16) with gain 1q or deletion 1p; and gain 1q with deletion 1p). Multivariable Cox regression was used to examine the association of HR features (FHR, adapted IMS-IMWG HR, and EMD, individually and collectively) with PFS while adjusting for baseline ferritin, prior BCMA, and ECOG performance status (PS ≥2). Among patients with ≥18 months of follow-up, we evaluated predictors of early relapse (≤18 months post infusion) using logistic regression. Results Our cohort included 598 patients treated with cilta-cel between 5/2022 and 12/2024 who had available data on all HR features. The median age was 65 years (range 33-83), and the median prior lines of therapy (pLOT) was 5 (range: 1-18, 84% with >3 pLOT); 8% had received prior BCMA-directed therapy, and 27% were penta-refractory. Most patients (88%) received bridging therapy and 30% achieved ≥ partial response. In our cohort, 33% (n=195) had FHR, 43% (n=258) had traditional HR cytogenetics, 42% (n=253) had adapted IMS-IMWG HR disease, and 11% (n=64) had EMD. Median follow up was 10.6 months (range 0.4-34.1). Overall response rate (ORR) and complete response (CR) rates were lower in patients with vs without HR features: FHR (ORR: 89% vs 94%, CR: 61% vs 75%), traditional HR (ORR: 90% vs 94%, CR: 65% vs 74%), adapted IMS-IMWG HR (ORR: 91% vs 94%, CR: 64% vs 75%) and EMD (ORR: 77% vs 94%, CR: 61% vs 71%). Corresponding 12-month PFS was also lower in patients with vs without HR features: FHR (55% vs 79%), traditional HR (63% vs 78%), adapted IMS-IMWG HR (61% vs 80%) and EMD (59% vs 73%). ORR, CR rate, and 12-month PFS in patients without any HR features compared to those with ≥1 HR feature were 96% vs 90%, 81% vs 64%, and 83% vs 65%, respectively. In multivariable models, each HR feature (FHR, adapted IMS-IMWG HR, and EMD) was independently associated with inferior PFS after adjusting for baseline ferritin, prior BCMA, and ECOG PS ≥ 2. However, because early relapse cannot always be identified a priori based on traditional risk factors, we sought to identify predictors of relapse within 18 months of cilta-cel in a subset of patients with ≥18 months of follow-up (N=247; infused by 12/2023). Notably, while FHR (OR=3.04, 95% CI=1.59-5.89) and adapted IMS-IMWG HR (OR=2.07, 95% CI=1.10-3.89) were independently associated with odds of relapse within 18 months, EMD was not (OR=0.90, 95% CI=0.32, 2.40) when accounting for prior BCMA, ECOG PS ≥ 2, and baseline ferritin. We also used a time-varying Cox model to examine each HR feature associated with PFS at two time periods, <18 and ≥18 months, while adjusting for baseline ferritin, ECOG PS, and prior BCMA. We confirmed that FHR and adapted IMS-IMWG HR were both associated with relapse <18 months (HR=2.05, 95% CI=1.48-2.83 and HR=1.56, 95% CI=1.11-2.19) but not in the ≥18-month time period (HR=0.70, 95% CI=0.19-2.59 and HR=0.76, 95% CI=0.24-2.40). The magnitude of the association for EMD with PFS was similar across both time periods (<18 months HR=1.67, 95% CI=1.09-2.57 and ≥18 months HR=2.17, 95% CI=0.45-10.40). Conclusion While cilta-cel leads to high response rates and durable PFS in RRMM, a subset of patients with FHR and adapted IMS-IMWG HR remains at increased risk for early relapse. In this large multi-center cohort, we describe PFS outcomes across high-risk subgroups and identify key predictors of relapse within 18 months of cilta-cel. These findings inform risk-adapted therapeutic strategies and interventional trials, including bridging, consolidation, or maintenance approaches, to improve outcomes in the highest-risk RRMM patients.

Abstract Introduction: Cilta-cel is approved in the US and the EU for the treatment of patients (pts) with lenalidomide-refractory multiple myeloma (MM) after ≥1 prior line of therapy (pLOT), including a proteasome inhibitor and an immunomodulatory agent based on the CARTITUDE-1 (CART-1, NCT03548207) and CARTITUDE-4 (CART-4, NCT04181827) studies. Cilta-cel led to deep and durable responses and significant survival benefit in pts with relapsed/refractory MM (RRMM) after ≥3 pLOT in CART-1. In CART-4, cilta-cel led to a significant overall survival benefit with a higher proportion of pts achieving deep, sustained minimal residual disease negativity vs standard of care in pts with lenalidomide-refractory MM after 1–3 pLOT. We investigated the mechanism of action of cilta-cel by correlating biomarkers from peripheral blood and the bone marrow TME with progression-free survival (PFS) and number of pLOT in pts with RRMM from CART-1 and CART-4. Methods: Biomarker analyses were performed using peripheral blood and bone marrow aspirates (BMAs) collected in both CART-1 and CART-4 studies. Immunophenotyping by flow cytometry was performed using peripheral blood at baseline/time of apheresis in samples from both studies. Immune fitness at baseline was assessed in pts by pLOT and in association with PFS. Gene set enrichment scores were derived from RNA sequencing data of the TME using BMAs. Mixed-effects models were used to identify signatures/pathways modulated over time (day [D] 28 and at 6 months [mo] post infusion vs screening) and in association with PFS and pLOT. Results: In total, 176 CART-4 and 97 CART-1 pts received cilta-cel as study treatment. Immunophenotyping data were available from 248 peripheral blood samples (1 pLOT, n=56; 2 pLOT, n=61; ≥3 pLOT, n=131). At apheresis, CD4+ naïve T cells (%) were higher in pts with 1 or 2 pLOT, compared with those with ≥3 pLOT, while no difference was found in pts with ≥3 pLOT with further breakdown (i.e., 3–4 vs 5–6 vs ≥7 pLOT). Further, higher baseline levels of CD4+ naïve T cells were associated with longer PFS. These data suggest that the negative impact on peripheral immune fitness with the addition of LOT may plateau when pts have received ≥3 pLOT. TME gene expression data were available from 148 BMA samples (screening, n=50; D28, n=50; 6 mo, n=48) from CART-4. Analyses of gene expression signatures suggest depletion of B cells and antibodies at D28 post cilta-cel infusion, with partial recovery at 6 mo. This was corroborated in parallel by flow cytometry and immunoglobulin quantification. Further, elevated expression of genes associated with myeloid cells, including tumor-associated macrophages (TAM, likely M1), and of genes associated with cytotoxic T cells were also observed at D28. Elevated expression of genes associated with B-cell receptors, T-cell differentiation and activation, and cytokine signaling pathways were found at both D28 and 6 mo. Pts with longer PFS (>18 mo) had higher levels of M1 TAM at D28, while those with shorter PFS had higher levels of regulatory T cells and more suppressed interferon pathway genes at 6 mo, suggesting that relapse may be associated with a more suppressive immune response. In pts with fewer pLOT, more profound B-cell depletion on D28, better recovery at 6 mo, and elevated B-cell receptor signaling were inferred from gene expression data. Pts with 1 pLOT also demonstrated higher elevation of M1 TAM expression on D28 from screening as compared with pts with 3 pLOT, similar to what was observed in pts with longer PFS. Conclusions: Correlative biomarker data suggest that longer PFS is associated with better immune fitness at baseline and stronger immune responses post cilta-cel infusion, as observed in peripheral blood and within the TME of pts with RRMM in CART-1 and CART-4. The peripheral immune fitness was more pronounced in pts with 1 and 2 pLOT vs 3 pLOT and beyond, where deterioration plateaued. This suggests the impact of T-cell immune fitness on PFS may be limited beyond the third LOT. Importantly, other covariates, including factors within the immune TME and tumor burden, may play a significant role in balancing this compromised intrinsic T-cell immune fitness at apheresis, thereby contributing to PFS durability after 3 pLOT. Overall, these results support increased benefits in treating pts with MM with cilta-cel in earlier LOTs.

Abstract Introduction: The CARTITUDE-4 study (NCT04181827), which enrolled patients with lenalidomide-refractory multiple myeloma (MM) after 1–3 prior lines of therapy (pLOT), demonstrated a significant benefit of ciltacabtagene autoleucel (cilta-cel) over established triplet regimens. However, the benefit of cilta-cel for patients with standard-risk cytogenetics remains less defined. Here, we report outcomes in patients with standard-risk cytogenetics from the intent-to-treat and as-treated populations in CARTITUDE-4. Methods: InCARTITUDE-4, patients randomized to the cilta-cel arm underwent apheresis and bridging treatment with either pomalidomide, bortezomib, and dexamethasone (PVd) or daratumumab, pomalidomide, and dexamethasone (DPd), followed by lymphodepletion therapy with cyclophosphamide and fludarabine, and then a single cilta-cel infusion. Progression-free survival (PFS) was assessed using a validated computerized algorithm. The intent-to-treat population included 208 patients; 32 patients progressed or died on bridging therapy, resulting in an as-treated population of 176 patients. Patients with high-risk cytogenetics, defined as del(17p), t(14;16), t(4;14), gain/amp(1q) (n=105), or with unknown cytogenetics (n=12), were excluded from the as-treated analysis. The 12-month minimal residual disease (MRD)-negative complete response (CR) was defined per the International Myeloma Working Group criteria as the proportion of participants with CR or better prior to and at 12 months (±3 months), achieving MRD-negative status at 12 months (+3 months) after cilta-cel infusion, as determined by next-generation sequencing (10-5), prior to progressive disease or subsequent anti-myeloma therapy. For the as-treated population, PFS rates were measured from the time of cilta-cel infusion. Results: In CARTITUDE-4, in the intent-to-treat population, at a median follow-up of 33.6 months, patients with standard-risk cytogenetics had a 30-month PFS rate (95% CI) of 71.0% (58.8–80.2) in the cilta-cel arm (N=69) vs 43.2% (31.3–54.5) in the standard-of-care (SOC) arm (N=70). Patients with standard-risk cytogenetics in the as-treated population (n=59) had a 30-month PFS rate of 80.5% (95% CI, 67.2–88.8). In CARTITUDE-1 (NCT03548207), which evaluated cilta-cel in patients with heavily pretreated relapsed/refractory MM (RRMM; ≥3 pLOT), the 30-month PFS rate among patients with standard-risk cytogenetics (negative for del(17p), t(14;16), or t(4;14); n=68) was 59.9% (95% CI, 47.2–70.5). In the CARTITUDE-4 as-treated population with standard-risk cytogenetics, 8 PFS events occurred within 1 year and 4 PFS events beyond 1 year of cilta-cel infusion. Twenty-six patients achieved MRD-negative CR at 12 months after cilta-cel; 100.0% of these patients were progression free at 30 months. Fourteen patients were not evaluable for MRD due to: calibration failure (n=12), no sample availability for testing (n=1), or indeterminate results post baseline (n=1). Conclusions: The PFS rate at 2.5 years for patients with standard-risk RRMM was higher in CARTITUDE-4 compared with CARTITUDE-1, supporting the use of cilta-cel as early as second line in the treatment course. In CARTITUDE-4 (as-treated population), 80.0% of patients with standard-risk cytogenetics were progression free and off treatment at 2.5 years. In patients with standard-risk disease who achieved MRD-negative CR at 1 year, this rate increased to 100.0%. The low rate of progression events in cilta-cel-treated patients with standard-risk cytogenetics shows the profound benefit of a single cilta-cel infusion in this population.

Abstract Background: Recent advances in computer vision have given rise to image foundation models (FMs), large-scale neural networks pre-trained on millions of unlabeled histopathology patches, which can learn generalizable representations of tissue morphology. In many cancers, these FMs have predicted genetic mutations and prognoses directly from histology, suggesting that morphology encodes “hidden” molecular information. In lymphoma, this field is nascent but growing.General-purpose histopathology FMs have limited representation of lymphoid morphology. We developed and evaluated LymphoVision, a lymphoma-specialized image foundation model pretrained with self-supervised learning, to extract clinically meaningful features from routine hematoxylin and eosin (H&E) histology. Methods: LymphoVision was pre-trained using knowledge distillation and consistency (DINOv2) on over 37 million multi-resolution patches (5x, 10x, 20x, 40x) sampled using a density-aware strategy from 31,211 H&E whole slide images (WSIs) from 9,155 archival lymphoma and reactive lymphoproliferative cases. To ensure diverse, representative samples, our patch sampling method uses an efficient graph-based clustering algorithm and prioritizes morphologically informative regions by clustering patch embeddings and performing density sampling within each cluster. The pretrained encoder was evaluated using a clustering-constrained attention multiple instance learning (MIL) framework on three diagnostic tasks: (1) cell-of-origin (COO) classification in diffuse large B-cell lymphoma (DLBCL), distinguishing germinal center B-cell (GCB) from non-GCB subtypes based on the Hans immunohistochemistry algorithm; (2) follicular lymphoma (FL) grading, differentiating low-grade (grades 1–2) from high-grade (grade 3); and (3) multiclass lymphoma subtype classification across six categories: DLBCL, FL, marginal zone lymphoma (MZL), mantle cell lymphoma (MCL), classical Hodgkin lymphoma (CHL), and benign/reactive lymphoid conditions. WSI-level labels were derived from pathology reports, and none of the evaluation cases were included in pretraining. Models used a ViT-Giant LymphoVision backbone to generate 20x patch embeddings, followed by MIL using 50% training data and 2-fold cross-validation (25% validation, 25% test). Performance was assessed using area under the receiver operating characteristic curve (AUC), accuracy, recall (sensitivity) and F1 score. Results: For DLBCL COO classification (total n=358 cases), LymphoVision achieved a mean test AUC of 0.93 (range 0.92–0.94), accuracy of 84.5% and F1 of 0.85 across two independent test folds (n=91 per fold, 50% GCB). These results outperformed general-purpose FMs on the same task and cases, including Virchow-v2 (AUC 0.85) and UNI-v2 (AUC 0.88), despite those models being pretrained on 10–100 times more images. For FL grading (total n=660 cases), the test set in each fold included 166 cases (18% grade 3), yielding a mean test AUC of 0.88 (range 0.87–0.90), accuracy of 89.2% and F1 of 0.80. In multiclass lymphoma subtype classification (total n=1,176 cases), the test set included 296 cases per fold. LymphoVision achieved a mean test AUC of 0.98 (range 0.97-0.99), mean test accuracy of 87.1% and mean F1 of 0.84. Class-wise recall was 92.0% for DLBCL (n=112), 88.2% for FL (n=72), 77.2% for MZL (n=29), 70.8% for MCL (n=35), 87.5% for CHL (n=24), and 95.8% for benign/reactive lymphoid conditions (n=24). Conclusion: LymphoVision is a lymphoma-specialized FM trained using self-supervision that achieves high diagnostic accuracy across multiple clinically meaningful tasks using weak supervision and standard H&E images, suggesting the encoder has learned fine-grained lymphoma morphology. Based on historical benchmarks from the 1997 International Lymphoma Study Group classification project, expert hematopathologists achieved diagnostic agreement rates of 55% to 84% using H&E morphology alone for the non-Hodgkin lymphoma (NHL) entities included in our multiclass task. LymphoVision approached or exceeded these benchmarks. These findings underscore the value of our approach using disease-specialized pretraining and intelligent patch sampling for data-efficient histopathology learning. LymphoVision has the potential to augment diagnostics, enable scalable AI-powered classification in real-world workflows, and serve as a foundation for prognostic and theragnostic applications. External validation and multimodal integration are ongoing.

Abstract Introduction: Only 30% of patients (pts) presenting with abnormal bleeding symptoms receive a clear diagnosis, with many classified as having bleeding disorder of unknown cause, or BDUC (Baker et al. 2021). Currently, the recommended diagnostic workup for BDUC by the International Society on Thrombosis and Haemostasis (ISTH) includes plasmatic assays for coagulation factors, von Willebrand factor (vWF) antigen/activity, and platelet function testing (PFT), specifically platelet function analyzer-100 (PFA-100) and light transmission aggregometry (LTA). However, PFTs are not widely available and cannot be mailed to reference laboratories. We studied the diagnostic yield of an expanded testing panel which includes quantitative platelet glycoprotein flow cytometry (GpFL) and platelet electron microscopy (PTEM), which we have optimized for specimen mailing. Methods: We conducted a minimal risk, retrospective study by reviewing the electronic health records of pts (18 years and older) who were seen in the Coagulation Clinic of a tertiary academic medical center for evaluation of a bleeding disorder between October 2017 and June 2025. Testing included plasma-based assays including PT, APTT, vWF assay, FVIII, FIX, FXIII screen, and thrombin time (and additional reflex assays as needed). At the discretion of the treating clinician, PFTs (PFA-100 and LTA), GpFL, PTEM, and fibrinolysis assays (plasminogen activator inhibitor-1 antigen [PAI-1 Ag] and alpha-2 plasmin inhibitor) were included. We abstracted clinical data on pt demographics, ISTH bleeding scores, and the results of coagulation testing. Pts with an established diagnosis of a bleeding disorder, underlying malignancy, and those who did not have plasma-based coagulation testing were excluded from this study. Results: 109 pts met study criteria, and 82.6% were females (n = 90) with a median age of 45 years (range = 18-88 years). The median ISTH-BAT score was 7 (range: 2-21). About half of the pts, 49.5% (n = 54), had normal coagulation profiles and PFTs. All 54 had PTEM performed, of which 14.8% (n = 8) were abnormal (alpha and/or dense granule deficiencies). 85.2% (n = 46) had GpFL testing performed, with one showing reduced Gp2b/CD42 and Gp3a/CD61 expression, four showing reduced Gp1a/CD49b expression, and three showing reduced GpVI expression. Half of the pts, 50.5% (n = 55) had an abnormal coagulation profile from plasma and/or PFTs, of which 20% (n = 11) had a clinically significant coagulation factor deficiency (FV, FVII, FX). Roughly 9.1% (n = 5) had type 1 vWD of which 1 also had PAI-1 deficiency. One pt had isolated PAI-1 deficiency. Abnormal coagulation profiles were noted in 53 pts who had PFA-100 and LTA done, and both tests were normal in 22.6% (n = 12). Within the abnormal coagulation profile group, 10.9% (n = 6) had ultrastructural abnormalities on PTEM (dense and alpha granule abnormalities). Majority of the pts (n = 46) had GpFL done, and one had abnormal reductions in Gp1b-aCD42b, three in Gp1aCD49b, and seven in GPVI. Comparison of the ISTH BAT score between the normal and abnormal coagulation profiles/PFTs groups by non-parametric (Mann-Whitney U test) and parametric (Welch's t-test) showed no significant difference (p = 0.56 and p = 0.87) respectively. Conclusion: In this cohort of pts with abnormal bleeding, standard recommended testing resulted in a diagnosis in 66.4% pts. The remainder would have been classified as BDUC. The addition of GpFL and PTEM resulted in establishing a diagnosis that explains the bleeding phenotype in 14.8% of pts with an otherwise normal coagulation profile and PFT. Although abnormal PFTs indicate a platelet function defect, GpFL and PTEM may further define the abnormalities particularly differentiating syndromic from non-syndromic inherited platelet disorders. Additional studies are needed to confirm our findings and explore the possibility of addition of GpFL and PTEM for patients with BDUC.

Abstract Background: Partial tandem duplications of the histone methyltransferase KMT2Agene(KMT2A-PTD) are the most common KMT2A-aberration in adult acute myeloid leukemia (AML). KMT2A-PTD AML represents a high-risk disease subset with a poor prognosis. However, KMT2A-PTD leukemia is under studied, likely due to a lack of experimental models, and therefore the underlying mechanisms of leukemogenesis remain elusive. Since KMT2A-PTD leukemias possess high-level expression of HOXAgenes, similar to KMT2A-rearranged (KMT2A-r) and NPM1c AML, patients with this type of AML are currently enrolled into several phase I/II trials assessing menin inhibitors (MIs). However, mechanistic or clinical evidence for treatment response is lacking. To date, no targeted therapeutic approaches for KMT2A-PTD leukemia are available. Results: To better understand KMT2A-PTD mediated leukemogenesis, we endogenously tagged the KMT2A-PTD or KMT2A-wildtype (WT) allele in EOL-1 cells with a dTAG that can be degraded with PROTACs. Targeted protein degradation showed that the KMT2A-PTD, but not the KMT2A-WT protein, is critical to maintain leukemia cell growth. We identified KMT2A-PTD specific genome-wide chromatin binding sites and direct target genes as assessed by changes in nascent RNA production that started to occur within 30 minutes of KMT2A-PTD degradation. KMT2A-PTD target genes identified in this study included HOXA9, HMX2, HMX3, CDKN2C, MYC and CDK6 among others. Having defined the KMT2A-PTD target gene program, we assessed the effect of MI on KMT2A-PTD function. In KMT2A-PTD cell lines, PDXs and cells from a published murine KMT2A-PTD/FLT3-ITD knockin model, we observed intrinsic resistance to revumenib characterized by retained chromatin binding of KMT2A-PTD, but not KMT2A-WT after MI treatment. We next defined critical domains for KMT2A-PTD function via a KMT2A domain scan. This revealed the CXXC domain and AT hooks as critical dependencies in EOL-1 cells. Overexpression of KMT2A or mutants with duplicated domains in 293T cells showed that a duplication of the CXXC domain or the AT hooks is sufficient to mediate retained KMT2A chromatin binding under MI. Therefore, the resistance to MIs is likely due to the duplicated protein of KMT2A subverting menin function at critical target loci. Given that KMT2A-PTD remains chromatin-bound upon menin inhibition, we focused on ENL as a potential regulator of leukemia gene expression programs. In KMT2A-r and NPM1c leukemia, the ENL protein has been shown to be a critical dependency as it associates with acetyl lysine via its YEATS domain. ENL/AF9 YEATS inhibitors reverse ENL-dependent gene expression including HOXA9 and MYC. Using the small molecules TDI-11055 and TM-7, we show that YEATS inhibition leads to loss of members of the super-elongation complex and DOT1L from KMT2A-PTD driven loci and downregulation of target gene expression. Treatment of KMT2A-PTD PDXs with YEATS inhibitors induced differentiation and significantly reduced leukemia burden in vivo. Interestingly, we show that in contrast to previous findings in KMT2A-r leukemia, loss of ENL alone is insufficient to abrogate leukemic proliferation of EOL-1 cells. Instead, by combining ENL degradation with CRISPR-Cas9 mediated knockdown of AF9, we find that both YEATS proteins, ENL and its paralogue AF9, support aberrant gene expression in KMT2A-PTD leukemia. Finally, we show that the novel combination of MI and YEATS inhibition (revumenib plus TDI-11055) synergistically affects proliferation and induces differentiation in cell line and PDX models of KMT2A-PTD leukemia, which in turn prolongs survival of mice with KMT2A-PTD leukemia. Conclusions: KMT2A-PTD leukemia is dependent on the mutated, but not the WT KMT2A allele. It is characterized by a relative, intrinsic resistance to MI monotherapy, which is mediated by a duplication of the CXXC domain and AT hooks of KMT2A. Targeting both, the ENL and AF9 YEATS domain, is a possible treatment strategy and when combined with revumenib, YEATS inhibition has synergistic effects and overcomes intrinsic MI resistance. However, based on our studies, MI monotherapy may not be effective for most patients with KMT2A-PTD leukemia. Given that numerous compounds targeting the ENL (and AF9) YEATS domain are under study and MIs have recently obtained FDA approval, the combination of MI and YEATS inhibition could be translated into clinical trials for patients with KMT2A-PTD leukemia in the relatively near future.