Citations

ABSTRACT Approximately 10% of patients with Ewing sarcoma (EwS) have pathogenic germline variants. Here, we report two cases: first, a novel germline pathogenic variant in partner and localizer of BRCA2 ( PALB2) in a patient with a late EwS relapse. Its impact on homologous recombination is demonstrated, and breast cancer risk is discussed. Second, we report a pathogenic germline variant in RAD50 in a patient with a rare prostate primary EwS and later concurrent relapsed EwS and therapy‐related acute myeloid leukemia (tAML) with two rare KMT2A translocations. Understanding rare pathogenic germline variants and the impact on cancer behavior, response to treatment and treatment‐associated toxicities is important for patients.

Abstract Background Ten years after the original publication, the Centiloid framework is now broadly used to harmonize amyloid‐PET quantification, facilitate data sharing and comparison across cohorts, and even assist visual interpretation in clinical settings. We evaluated the global implementation of Centiloids by comparing their distribution and corresponding positivity thresholds across cohorts. Methods We gathered data from publicly available cohorts and reached out to investigators across the world to collect cross‐sectional Centiloids, demographic and clinical information, and visual read data. Gaussian mixture models (GMM, k=2) were fitted to Centiloid values for each cohort and cutoffs were calculated as mean + 2SD of the lower Gaussian. When visual reads were available, we determined Centiloid cutoffs that maximized correspondence with visual reads (Cohen's kappa). Data was combined across cohorts using random effects meta‐analyses. Results As of January 2025, we included 37 cohorts ( n = 41,678 participants) with heterogeneous pipelines, radiotracers, and clinical and demographic characteristics (Table‐1). The low Gaussian peaks ranged from ‐9 to 10CL; the meta‐analysis identified a common peak at 1CL. The second peaks were more heterogeneous (range=38‐102CL; meta‐analysis outcome=64CL). Across cohorts, the proportion of cognitively unimpaired versus impaired participants impacted the position of both peaks, with better separation in cohorts enriched in impaired individuals (Figure 1C). The meta‐analysis indicated a GMM‐based cutoff of 19CL (95%CI: 16‐21CL, Figure 1B); subgroup analyses showed no evidence of significant effect between single versus multicenter settings (17 versus 20CL, p = 0.30), MRI‐based or PET‐only processing (18 versus 19CL, p = 0.88), and no evidence of difference across radiotracers (Flutemetamol: 16CL; PIB: 17CL, Flutafuranol: 18CL, Florbetaben: 19CL, Florbetapir: 20CL, p = 0.78). In a subset of 29,496 participants with visual reads available, binary visual reads corresponded well to Centiloids (common kappa=0.86, Figure 2A). The visual read‐based cutoff of 24CL (95%CI: 21‐27CL, Figure 2B) maximized correspondence between visual read and quantification and was slightly higher than the GMM‐based cutoff. All meta‐analysis models showed high non‐random heterogeneity (I 2 >80%) across studies, suggesting non‐random differences in peaks and cutoffs. Conclusions Meta‐analysis‐based cutoffs align well with thresholds from the existing literature. High heterogeneity among studies underscores the need to investigate contributing factors, raising concerns about applying common cutoffs.

Abstract Background Ten years after the original publication, the Centiloid framework is now broadly used to harmonize amyloid‐PET quantification, facilitate data sharing and comparison across cohorts, and even assist visual interpretation in clinical settings. We evaluated the global implementation of Centiloids by comparing their distribution and corresponding positivity thresholds across cohorts. Methods We gathered data from publicly available cohorts and reached out to investigators across the world to collect cross‐sectional Centiloids, demographic and clinical information, and visual read data. Gaussian mixture models (GMM, k=2) were fitted to Centiloid values for each cohort and cutoffs were calculated as mean + 2SD of the lower Gaussian. When visual reads were available, we determined Centiloid cutoffs that maximized correspondence with visual reads (Cohen’s kappa). Data was combined across cohorts using random effects meta‐analyses. Results As of January 2025, we included 37 cohorts ( n = 41,678 participants) with heterogeneous pipelines, radiotracers, and clinical and demographic characteristics (Table‐1). The low Gaussian peaks ranged from ‐9 to 10CL; the meta‐analysis identified a common peak at 1CL. The second peaks were more heterogeneous (range=38‐102CL; meta‐analysis outcome=64CL). Across cohorts, the proportion of cognitively unimpaired versus impaired participants impacted the position of both peaks, with better separation in cohorts enriched in impaired individuals (Figure‐1C). The meta‐analysis indicated a GMM‐based cutoff of 19CL (95%CI: 16‐21CL, Figure‐1B); subgroup analyses showed no evidence of significant effect between single versus multicenter settings (17 versus 20CL, p = 0.30), MRI‐based or PET‐only processing (18 versus 19CL, p = 0.88), and no evidence of difference across radiotracers (Flutemetamol: 16CL; PIB: 17CL, Flutafuranol: 18CL, Florbetaben: 19CL, Florbetapir: 20CL, p = 0.78). In a subset of 29,496 participants with visual reads available, binary visual reads corresponded well to Centiloids (common kappa=0.86, Figure‐2A). The visual read‐based cutoff of 24CL (95%CI: 21‐27CL, Figure‐2B) maximized correspondence between visual read and quantification and was slightly higher than the GMM‐based cutoff. All meta‐analysis models showed high non‐random heterogeneity (I 2 >80%) across studies, suggesting non‐random differences in peaks and cutoffs. Conclusions Meta‐analysis‐based cutoffs align well with thresholds from the existing literature. High heterogeneity among studies underscores the need to investigate contributing factors, raising concerns about applying common cutoffs.

ABSTRACT Background Isolated chromosome 5/5q losses (–5/5q) without TP53 mutations are associated with favorable outcomes in myeloid neoplasms (MN) with <5% blasts. However, the clinical implication of concurrent −5/5q and TP53 aberrations in MN with ≥5% blasts is poorly understood. Methods Patients with TP53 ‐mutated MN carrying ≥5% blasts assessing the prognostic impact of ‐5/5q on 24‐month overall survival (OS24) were examined. Results Of 587 patients, 515 (88%) exhibited −5/5q overwhelmingly in the context of a complex karyotype (98.3% vs. 61.1% complex karyotype without −5/5q; p < .0001) and multihit TP53 allelic state (88.3% vs. 56.9%; p < .0001). Proportions of patients with blasts ≥20% were comparable between groups with and without −5/5q; p = 0.26. Notably, patients with −5/5q exhibited significantly fewer coalterations; p < .0001. Looking at outcomes, presence of −5/5q was associated with shorter median 24‐month overall survival (7.8 months vs. 11.2 months; p Log‐rank = .012), an effect restricted to subgroups with blasts <20% ( p = .039; N = 163), absent −7/7q ( p = .007; N = 225), or WHO5‐defined single hit allelic state ( p = 0.030; N = 91). Importantly, −5/5q retained independent adverse prognostic significance regardless of TP53 allelic state in a multivariable model. Furthermore, among the subset of 75 (13%) patients undergoing allogeneic stem cell transplantation, −5/5q predicted significantly shorter median 5‐year posttransplant survival (16.2 months vs. median not reached; p Log‐rank = .009). Conclusions These findings emphasize the independent prognostic relevance of chromosome 5/5q losses underscoring the clinical relevance of cytogenetic testing for −5/5q even in this high‐risk cohort.

Abstract Pancreatic cancer, projected to become the second-deadliest malignancy in Western societies by 2030, requires urgent innovations in early detection and risk stratification. Intraductal papillary mucinous neoplasms (IPMNs) represent critical precursor lesions, but current clinical guidelines demonstrate suboptimal accuracy in malignancy prediction, leading to either unnecessary surgeries or missed opportunities for early intervention. In this paper, we present Cyst-X, an artificial intelligence (AI) framework that accurately predicts IPMN malignant transformation using multicenter magnetic resonance imaging (MRI) data. Unlike most previous approaches that rely on computed tomography (CT), our method capitalizes on the superior soft tissue contrast of MRI, allowing for more precise identification of subtle imaging biomarkers. We developed and validated deep learning models on 723 T1-weighted and 738 T2-weighted MRI scans from 764 patients among seven international institutions, demonstrating significantly superior performance (AUC=0.82) compared to current clinical (Kyoto) guidelines (AUC=0.75) and expert radiologists. AI-based imaging features correlate strongly with clinically recognized malignancy markers, providing potential biologically relevant insights. This approach holds promise to significantly refine clinical decision-making, reduce unnecessary surgeries, and better identify high-risk IPMN patients for timely intervention. Our approach integrates a novel pancreas segmentation network with robust classification models that identify subtle imaging biomarkers associated with malignancy risk. Importantly, we demonstrate that these models retain high performance in a privacy-preserving federated learning setting, where institutions collaboratively train AI models without exchanging patient data to address key regulatory and ethical barriers. We publicly release the Cyst-X dataset--the first large-scale, multi-center pancreatic cyst MRI collection--to accelerate research in this field. This study addresses a critical clinical need while establishing technical foundations for privacy-preserving AI in radiology that could transform pancreatic cancer management through earlier intervention and reduced unnecessary procedures. The dataset can be accessed at https://osf.io/74vfs/, and the source code for our deep learning segmentation and classification models is available at https://github.com/NUBagciLab/Cyst-X.

Abstract Pyroptosis is a form of programmed cell death characterized by the cleavage of the gasdermin (GSDM) family proteins that form pores in the plasma membrane, cell rupture, and the release of proinflammatory cytokines. In this study, we performed immunohistochemistry for cleaved gasdermin D (GSDMD), gasdermin E (GSDME) N-terminus, and gasdermin C (GSDMC) in 2 different cohorts of diffuse large B-cell lymphoma (DLBCL) and analyzed the impact of GSDM expression on prognosis and immunity. The results showed frequent cleaved GSDMD (N-terminal) expression. Only cytoplasmic GSDMD N-terminal expression correlated with significantly better patient survival in the 2 cohorts. In contrast, GSDME was mainly expressed in the vascular endothelium, and correlated with significantly adverse prognostic effect. Correlating with the multiplex fluorescent immunohistochemistry results, we found that cytoplasmic GSDMD N-terminal expression was associated with increased CD38+ (activated) M1 macrophages in both cohorts, cognate interactions between live DLBCL cells and activated M1 macrophages (and T cells), and lower PD-1/PD-L1 expression in the analyzed cases. In contrast, T-cell pyroptosis, lymphoma cell-resistance to cell death, and phagocytosis by M2 macrophages were observed in tissues with predominantly nuclear GSDMD N-terminal expression. Bulk gene expression profiling and deconvolution analysis revealed associations of cytoplasmic GSDMD N-terminal expression with downregulation of “Don’t eat me”–signaling genes, upregulation of many RNA genes, decreased frequency of “inflammatory” lymphoma microenvironment subtype, increased frequencies of prognostically favorable cell states and ecotypes, and decreased frequency of T-cell exhaustion state. In summary, this study showed distinct cellular and subcellular patterns of 3 GSDM proteins and their associated immune response phenotypes and prognostic effects, with implications for novel therapeutic strategies for B-cell lymphoma.

Abstract BTK degraders promote proteasomal degradation of BTK and can overcome resistance to covalent and noncovalent BTK inhibitors in patients with relapsed CLL. Currently the molecular basis for BTK degrader resistance is unknown. We therefore set out to understand BTK degrader resistance based on observations from the phase 1 trial of zelebrudomide (NX-2127) (NCT04830137), the first BTK degrader to enter clinical studies. Targeted 500-gene sequencing was performed on blood and/or tissue from four CLL patients treated with zelebrudomide. In one patient previously treated with chemoimmunotherapy, ibrutinib, venetoclax, and pirtobrutinib, BTK A428D was detected in the baseline blood sample with variant allele frequency (VAF) of &lt;2%, along with BTK V416L (13%) and mutations in TP53 (60%) and SF3B1 (37%). Following daily treatment with 100mg zelebrudomide and subsequent dose escalation to 200mg, at 6 months the VAF of all four mutations increased, with BTK A428D VAF increasing to 6% in the blood and 29% in a lymph node. BTK A428D was detected at baseline but not at progression in a second patient. To evaluate the impact of BTK A428D on drug response, we transduced BTK-dependent TMD8 cells with BTK WT, C481S, L528W, and A428D, and found that BTK A428D uniquely conferred resistance to all FDA-approved BTK inhibitors (ibrutinib, acalabrutinib, zanubrutinib, pirtobrutinib) as well as three BTK degraders under clinical evaluation (zelebrudomide, bexobrutideg, BGB-16673). To model the subclonal nature of BTK A428D we conducted a competition assay in which TMD8 cells expressing GFP + BTK A428D (5%) were mixed with TMD8 cells expressing mCherry + BTK WT (95%); over 9 days the GFP + BTK A428D expanded from 5% to 88-97% with all four BTKi and all three BTK degraders. In vitro kinase assays revealed that the BTK A428D mutant protein lacks kinase activity yet facilitates calcium flux and ERK/AKT signaling upon B-cell receptor activation, suggesting a scaffolding effect similar to other kinase deficient BTK mutations. To determine whether BTK A428D affects BTK degradation, we developed a reporter that fuses the BTK kinase domain to GFP. BTK A428D showed minimal degradation following 20 hours of treatment with a titration of all three BTK degraders. Similarly, we knocked BTK A428D into the endogenous BTK locus of HBL1 cells and confirmed a lack of degradation at doses up to 1 µM of all 3 degraders. Consistent with these findings, surface plasmon resonance revealed that the FDA-approved BTK inhibitors as well as zelebrudomide, bexobrutideg, BGB-16673, and AbbVie-1 (Compound 1 from WO 2023183811) are unable to bind to BTK A428D. Additionally, an in-cell NanoBRET assay demonstrated that zelebrudomide was unable to mediate the interaction between cereblon and BTK A428D. Strikingly, NanoBRET also demonstrated that a panel of promiscuous kinase-binding compounds was unable to bind to BTK A428D. The lack of binding of both selective and promiscuous kinase binders to BTK A428D led us to speculate that the mutation disrupted the ATP binding pocket to an unusual degree. We therefore solved the crystal structure of apo BTK A428D and found that, in addition to its predicted steric clash with ligand binding in the hinge region, the mutation led the protein to crystallize with its activation loop occupying the catalytic cleft, rendering it inaccessible. This is the first apo BTK structure observed in this autoinhibited conformation, and explains the observed binding profile. We hypothesized that the addition of a second targeted agent such as venetoclax might mitigate expansion of the A428D subclone. To test this, we introduced the BTK A428D mutation into the endogenous BTK gene in 3% of TMD8 cells and then exposed them to 0.1µM of zelebrudomide, bexobrutideg, or BGB-16673 +/- venetoclax 0.1µM. Combining the BTK degraders with venetoclax slowed the emergence of the BTK A428D resistant population and, in the case of zelebrudomide, eradicated both the wild-type and A428D clones. In summary, we identified and validated BTK A428D as a clinically relevant acquired pan-resistance mutation that confers resistance by blocking access to the ATP binding pocket. This mechanism differs from other kinase dead BTK mutations in which the kinase's enzymatic pocket remained accessible.These data also provide rationale for combining BTK degraders with BCL2 inhibitors, particularly in patients with BTK A428D mutations, or high-risk mutations such as in TP53.

Abstract Background: Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma (NHL), accounting for approximately 30% of all NHL cases. DLBCL diagnosis relies on histologic examination, immunophenotyping, molecular and fluorescence in situ hybridization (FISH) analysis to differentiate it from similar NHL subtypes with overlapping morphological features, including more high-grade lymphomas like double-hit lymphoma. Because FISH is performed in a panel of single- or dual-target test format and analyzed through microscopy and manual image analysis, assessment of multiple target biomarkers can be cumbersome and expensive. As additional biomarkers for diagnosis, subtyping, prognosis, and therapeutic decision-making become available, target panels will become larger, posing additional challenges to clinical laboratories. Patients and Methods In this study, Hi-C sequencing of FFPE samples was performed to identify all genomic rearrangements by a single sequencing analysis for an unselected DLBCL case series of 159 patients, and the results were compared to FISH for the detection of clinically important variants. Results Hi-C sequencing showed superior performance compared to FISH: standard FISH for DLBCL includes testing for MYC, BCL2, and BCL6. Between FISH and Hi-C, these biomarkers were detected 105 times, of which, FISH detected 80/105 (76%) and Hi-C detected 102/105 (97%). The cohort included 4 double-hit lymphomas by FISH but Hi-C sequencing detected five. Besides DLBCL diagnostic biomarkers (MYC, BCL2, BCL6), many additional biomarkers were detected by Hi-C sequencing including variants not tested by FISH because the clinical suspicion was ambiguous and the relevant FISH probes were not used, and those not tested because the biomarkers were not thought to be common and, hence, left out of the FISH panel. In total, Hi-C sequencing detected additional biomarkers in 42 (26%) cases, including (1) variants targeted by FISH (i.e. MYC, BCL2, BCL6) but negative, (2) classification biomarkers for other lymphomas (CCND1, MALT, IRF4), and (3) those not tested because the biomarkers are not common and not included in the FISH panel. These additional biomarkers were supportive of suspected diagnosis of DLBCL versus double-hit lymphoma (24 cases), prompted reconsideration of a different diagnosis that may have led to different treatments (6 cases, e.g. mantle cell lymphoma), or biomarkers such as PDL1, ALK, and ATM which have implications for targeted therapies and/or inclusion in clinical trials (5 cases). Conclusion The Hi-C FFPE sequencing technology has the potential to improve the workflow of pathology labs for detection of a wide range of structural variants informing precision diagnosis and treatment for lymphomas.

Abstract Background: BCMA- and GPRC5D-targeting immunotherapies including CART cell therapy and Bispecific antibodies are currently approved for the treatment of multiple myeloma. However, clinical outcomes after dual antigenic exposure remain undefined. Herein we report the first report of BCMA-GPRC5D dual targeted multiple myeloma (DEMM). Methods: We conducted a retrospective multi-center study of RRMM patients treated with both BCMA- and GPRC5D-targeted treatments (TT) (CAR T, bispecific antibodies, ADCs) between November 1, 2022, and July 1, 2025. Outcomes included PFS1 (post-dual exposure), and overall survival (OS). Kaplan-Meier estimates and event-free probabilities were calculated. Results: One hundred and fifty patients who underwent TCE treatments were included and 49 patients met the criteria for inclusion. Median age was 65 years, 22.4% had creatinine clearance &lt;=30 and ECOG 2 or greater was present in 22.2%; 57% had high-risk cytogenetics (IMWG 2016), and 60.5% were penta-refractory. Patients received a median of 6 prior lines of therapy (LOT) before 1st TCE and 9 prior LOT before DEMM. 65.3% received 1 TCE, 30.6% received 2 TCE and 4.1% received 3 TCE before DEMM. Similarly, 67.3% received 1 BCMA, 24.5% received 2 BCMA TT and 4.1% received 3 BCMA TT before DEMM. Prior TT included- BCMA CART (57.1%), BCMA Bispecific (57.1%), BCMA ADC (18.4%), other TCE (10.2%). The last dual exposure was GPRC5D Bispecific in 91.8%, BCMA Bispecific in 6.1% and BCMA CART in 2%. The median followup post relapse was 3.4 months (range 0.0-16 mos). The median PFS during the last dual exposure TT was 3.98 months (95% CI: 2.96–6.70). 32.7% did not receive any subsequent LOT and died due to progressive disease. 67.3% of patients had progressive MM and received next LOT. Median PFS1 was 1.77 months (95% CI: 1.28–6.21) with 3-, 6-, and 12-month event-free probabilities of 44%%, 19% and 0% respectively. Median OS was 5.7 months (95% CI: 2.53-12.16) for DEMM from the time of refractoriness to dual antigenic targets. Conclusions: Patients dual-exposed to BCMA- and GPRC5D-directed therapies have a poor prognosis, 32.7% had not received further treatment and in those who received treatment the PFS1 was less than 2 months and an OS of 5.7 months from progression. This reflects an area of high unmet need for the patients that are dual-exposed to BCMA- and GPRC5D-directed therapies. Further data on the cohort will be updated at the meeting. Acknowledgements include, the International Myeloma Foundation, IMWG.