Citations

Abstract Background: Polycythemia vera (PV) is characterized by excessive erythrocytosis and elevated hematocrit (Hct), which is associated with increased risk of thrombotic events. Controlling Hct levels (<45%) is a key therapeutic goal to reduce thrombotic events. Sapablursen (ISIS 702843) is a liver-directed, ligand-conjugated antisense oligonucleotide that suppresses transmembrane serine protease 6 (TMPRSS6) to enhance hepcidin production thereby restricting iron availability and reduced erythrocytosis. We are reporting the topline results of the IMPRSSION study (NCT05143957), an ongoing global, open-label phase 2 trial evaluating the safety and efficacy of sapablursen for Hct control and reducing phlebotomy rate in patients with PV. Methods: Eligible patients met 2016 WHO diagnostic criteria for PV, were required to have >3 phlebotomies within six months prior to screening, including >1 phlebotomy in the last 12 weeks. There were no exclusions for HCT prior to randomization. Patients received monthly subcutaneous doses of sapablursen in Cohort A (high dose) or Cohort B (low dose), during a 37-week treatment period. The primary efficacy endpoint was the reduction in the frequency of phlebotomy during the last 20 weeks of the treatment period (Week 17-37) compared to baseline (defined as the mean weekly phlebotomy rate during the six months prior to screening through the first dose of sapablursen). Phlebotomy was performed when Hct levels were ≥45% or at the investigator's discretion, confirmation of Hct was not required for phlebotomy. Serial assessments of serum hepcidin, TSAT and ferritin were conducted. The 10-item Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score (MPN-SAF-TSS) assessed the change in symptoms from baseline to Week 37. Results: Forty-nine (49) patients received at least one dose of sapablursen (full analysis set) in either Cohort A (n=32) or Cohort B (n=17). The median age was 61 years, 40 (82%) were male, and 29 (59%) classified as high risk. The primary endpoint was met at both dose levels. The weekly phlebotomy rate (mean±SD) decreased from baseline of 0.15±0.07 (~7.8/yr) to 0.05±0.09 (~2.6/yr) during the primary endpoint window (Week 17-37) in Cohort A (n=32, p<0.0001; 95%CI, -0.13, -0.07), and from 0.17±0.07 (~8.9/yr) to 0.07±0.07 (~3.6/yr) in Cohort B (n=17, p=0.0001; 95%CI, -0.14, -0.06). The mean MPN-SAF-TSS was reduced at Week 37 by 32% in Cohort A (change from baseline LSM -6.2, p=0.005) and 11% in Cohort B (change from baseline LSM -2.7, p=0.344). The proportion of patients with Hct control (<45%) without phlebotomy from week 17 to 37 in Cohort A and B was 53% (17/32) and 35% (6/17), respectively. Serum hepcidin was increased from baseline values of 0.25±0.6 to 5.05±4.7 nM at Week 13 in Cohort A (p<0.0001) and from 0.68±1.7 to 2.06±3.6 nM in Cohort B (p=0.0142). A sustained low TSAT was observed between Weeks 17-37 (6.0±2.1% and 5.5±2.0% in Cohort A and 8.9±5.0% and 7.0±3.4% in Cohort B). Serum ferritin increased over time from 18±22 ng/mL and 28±58 ng/mL at baseline in Cohort A and B, respectively, to 44±47 ng/mL and 41±85 ng/mL at Week 37, respectively. Sapablursen was generally safe and well-tolerated. The incidence of injection site reactions was 12.2% (6/49). Injection site reactions did not recur following multiple injections, were not progressive and resolved spontaneously. There was one death on study due to transformation to AML which was deemed as not related to the study drug. Conclusions: Treatment of PV patients with sapablursen increased serum hepcidin, controlled Hct, reduced phlebotomy need and improved quality of life in a dose-dependent manner. These findings highlight the potential of sapablursen as a therapeutic approach in PV and support further development.

Src homology 2-domain containing protein tyrosine phosphatase 2 (SHP2), encoded by the Ptpn11 gene (Tyrosine-protein phosphatase non-receptor type 11), is a key downstream effector of PD-1/PD-L1 signaling and is likely important, in addition to immune modulation, in tumor development and vascular homeostasis. SHP2 conveys PD-1 mediated inhibitory signaling in T cells, and is emerging as a therapeutic target. Importantly, there is an association between immune checkpoint inhibitors (ICIs), immune-related adverse events (irAEs), and cardiovascular complications, underscoring the need to understand SHP2’s role in these processes. This review aims to summarize current knowledge on SHP2/PTPN11 biology, its role in immune regulation, cancer progression, and vascular homeostasis, and to discuss emerging therapeutic strategies targeting this pathway. The concept of using SHP2 inhibitors with immune checkpoint inhibitors (ICIs) is being investigated to address ICI resistance and to improve anti-tumor efficacy substantially. SHP2 is also being studied in non-cancer cell contexts, and signaling responses can differ by large magnitudes depending on the biological context and stimuli. Under normal circumstances, SHP2 promotes vascular homeostasis in endothelial cells (ECs) and myeloid cells and inhibits inflammation, and the reduction in SHP2 activity by oxidative stress, such as in atherosclerosis or diabetes, upregulates inflammation. In contrast, in response to radiation, the fibrotic response and subsequent lung injury were increased by endothelial SHP2 induction via Notch-Jag1 signaling. Vascular smooth muscle cells SHP2 act as a pro-atherogenic effector by enhancing ERK/MAPK signaling, and the upregulation of mitochondria localized SHP2 can also induce cellular senescence-associated inflammation by upregulating mitochondrial reactive oxygen species. Taken together, the two opposite signaling effects of SHP2 suggest that both the immune and vascular system responses appear to be more modulated by the redox, cell, and compartment-specific signaling of SHP2. More studies are needed for mitigating cardiovascular toxicity to patients, particularly with ICI-based treatment regimens.

Chronic traumatic encephalopathy (CTE) is a neurodegenerative disease associated with exposure to repetitive head impacts (RHI). In CTE, hyperphosphorylated tau (p-tau) aggregates are found in neurons at the depth of cortical sulci close to the gray matter/white matter (GM/WM) boundary. To date, CTE can only be diagnosed postmortem by neuropathological examination. Traumatic encephalopathy syndrome (TES) is the clinical syndrome purported to be associated with CTE pathology. The aim of this study is to investigate microstructural properties at the GM/WM boundary in individuals with a history of exposure to RHI and clinical features of CTE (i.e., TES). Diffusion magnetic resonance imaging (dMRI), TES diagnoses, and cerebrospinal fluid (CSF) biomarkers were acquired from 165 male former American football players (age: 57.29 ± 8.23 years) from the DIAGNOSE CTE Research Project, a multicenter, observational cohort study. Fractional anisotropy (FA) was measured at the GM/WM boundary of the whole brain. In addition, a widely used method (tract-based spatial statistics [TBSS]) was applied to measure FA of central WM. We used analyses of covariance to test associations between FA and TES. Furthermore, we used linear regressions to test associations between FA and nine CSF biomarkers (i.e., p-tau-181, -217, -231, total tau, amyloid β [Aβ] 1–40 , Aβ 1–42 , glial fibrillary acidic protein [GFAP], neurofilament light [NfL], and soluble triggering receptor expressed on myeloid cells-2 [sTREM2]). We report an association between higher FA at the GM/WM boundary and higher levels of certainty for CTE pathology ( F (1, 147) = 5.781, 95% confidence interval (CI) = 0.0003–0.003, p = 0.035) as well as neurobehavioral dysregulation ( F (1, 148) = 7.559, 95% CI = 0.001–0.009, p = 0.020), and functional dependence/dementia ( F (1, 148) = 5.046, 95% CI = 0.0004–0.006, p = 0.039). In addition, we report an association between higher FA at the GM/WM boundary and higher CSF p-tau-181 (β = 0.272, 95% CI = 0.078–0.466, p = 0.029) and p-tau-217 (β = 0.295, 95% CI = 0.102–0.488, p = 0.027). FA of the central WM was not associated with TES diagnoses. Taken together, these findings suggest that dMRI at the GM/WM boundary could be used to investigate microstructural alterations suggestive of tau pathology-associated neurodegeneration in individuals with TES, the clinical presentation of CTE. Future studies are needed to validate this approach and to identify clinically useful cutoff values for dMRI metrics.

Abstract Computer vision now enables deep learning models to link histopathologic features to molecular features in a quantitative manner not previously achievable by human pathologists. However, human interpretability is often limited. We trained a deep neural network to predict gene expression from digital whole slide images (WSIs) using 848 central nervous system (CNS) tumors with paired WSI and RNA-sequencing data. We used inferred RNA expression levels as a surrogate for protein expression of the CNS tumor lineage markers GFAP, OLIG2, and SSTR2, as well as the proliferation marker MKI67 (Ki67). We established thresholds for categorizing the inferred expression levels as positive or negative based on levels observed in cross-validation testing in select tumor types to approximate routine clinical interpretation of immunohistochemical (IHC) staining. We tested the sensitivity and specificity of our ‘virtual’ IHC on an independent, multi-institutional cohort of over 2,000 CNS tumors with objective diagnostic labels derived from DNA methylation-based tumor classification. As a dichotomous variable, inferred GFAP expression showed a sensitivity of 74% and a specificity of 99% as a glial marker in a cohort of gliomas and meningiomas. In the same cohort, OLIG2 showed a sensitivity of 75% and a specificity of 97% as a glial marker, while SSTR2 showed a sensitivity of 83% and a specificity of 97% as a marker of meningioma. In a cohort of ependymomas and gliomas, virtual IHC for OLIG2 showed a sensitivity of 75% and a specificity of 82% in distinguishing ependymomas from gliomas. Finally, in a cohort of gliomas, inferred expression of MKI67 demonstrated a trend consistent with what would be expected by actual IHC, with increasing MKI67 expression with increasing tumor grade. Our model provides the basis for a human interpretable and clinically applicable deep neural network to aid human pathologists in the diagnosis and grading of CNS tumors.

Abstract Molecular testing has become critical components in the diagnostic classification of the central nervous system (CNS) tumors. However, these methods require substantial resources, limiting accessibility for many patients. Recent advances in artificial intelligence (AI) and computer vision empower deep learning models to infer molecular features from histopathology images, which may often be more readily available than molecular testing, to classify CNS tumors. We trained pan-CNS tumor models to predict DNA methylation and gene expression from whole-slide images (WSIs). These molecular predictions were then used in a hierarchical machine learning framework, termed Neuropath-AI, to predict nine broad tumor families and 52 tumor types using a large diverse cohort of 5,715 histopathology samples, of which 2,988 had paired DNA methylation profiling and 848 had paired RNA-sequencing. We evaluated Neuropath-AI on another large independent multi-institutional cohort of 5,516 CNS tumors, for which Neuropath-AI predicted tumor class with an associated confidence score. Neuropath-AI designated 46% of test samples as predictable with high-confidence, which it then accuracy classified in 97% of samples. It made moderate-confidence (and above) predictions for 87% of test samples, for which it achieved a diagnostic accuracy of 80% for the top-1 prediction and 86% accuracy for a top-2 predictions. A comparison with human neuropathologists showed that Neuropath-AI is comparable at classifying tumors. A second test showed that human neuropathologists were more accurate when the Neuropath-AI classifier results were provided to them, testifying to the potential translational benefit of integrating AI classification tools into the pathologist’s workflow. Our model provides the basis for a clinically applicable deep learning assistant to improve human efficiency and accuracy of CNS tumor diagnosis. The model will be made publicly available and can be readily implemented to assist human pathologists in clinical workflows, in further expanded prospective studies.

Abstract Repetitive head impacts are common in contact and collision sports and are linked to structural brain changes and an elevated risk of neurodegenerative diseases such as Chronic Traumatic Encephalopathy. Identifying early in vivo structural markers remains challenging. Although diagnosis currently requires post-mortem confirmation, clinical symptoms, including cognitive impairment and behavioural changes, are reflected in the diagnosis of Traumatic Encephalopathy Syndrome. These symptoms align with dysfunction in key brain regions—amygdala, hippocampus and thalamus—which support memory, emotion and behaviour and commonly show tau pathology in Chronic Traumatic Encephalopathy. This study uses shape analysis to examine structural differences in these regions between former American football players and unexposed asymptomatic controls and evaluates the influence of age, head impact exposure and clinical diagnosis on brain structure. We analyzed brain morphology in former American football players (n = 163) and unexposed, asymptomatic controls (n = 53). Structural segmentation was performed with FreeSurfer 7.1, and the shape analysis pipeline was used to generate subregional reconstructions. Vertex-level morphometry, based on the logarithm of the Jacobian determinant and radial distance, quantified local surface area dilation and thickness. Group differences were examined with covariate-adjusted linear regression models contrasting football players and controls, as well as participants with and without a Traumatic Encephalopathy Syndrome diagnosis. Partial correlations examined the influence of age, age of first football exposure and cumulative head impact index metrics, including frequency, linear acceleration and rotational force. Models were adjusted accordingly for age, body mass index, education, race, imaging site, apolipoprotein ϵ4 status and total intracranial volume. Former football players exhibited bilateral surface area contractions in the hippocampus and amygdala, along with reduced amygdala thickness, compared to controls. Older age was associated with widespread surface contractions and thinning across all regions, except for preserved thickness in the left hippocampus. An earlier age of first exposure to football correlated with surface contractions in the thalamus and left hippocampus. Greater cumulative linear acceleration was linked to bilateral hippocampal surface contractions and reduced thickness in the left thalamus, while greater rotational force exposure was associated with hippocampal thinning. No significant structural differences were found between players with and without a diagnosis of Traumatic Encephalopathy Syndrome. These findings extend volume-based research by revealing localized alterations in surface area dilation and thickness and emphasize the roles of age and repetitive head impact exposure in long-term brain changes.

BACKGROUND: The association between breast cancer diagnosis, treatment, and the risk of incident ischemic stroke remains unclear. We aimed to investigate ischemic stroke risk among breast cancer survivors and evaluate the association by age, follow-up duration, and cancer treatments. METHODS: Using the Korean National Health Insurance Service database, we studied 113,232 women newly diagnosed with breast cancer (aged ≥18 years) without prior stroke history who underwent breast cancer surgery between January 2010 and December 2016. Each was matched 1:3 by birth year to a cancer-free female population (n=322,818). Subdistribution hazard ratios (sHRs) and 95% confidence intervals (CIs) were estimated, accounting for death as a competing risk and adjusting for sociodemographic factors and cardiovascular and noncardiovascular comorbidities. RESULTS: Over a mean follow-up of 7.2 years, ischemic stroke occurred in 1,155 (1.0%) breast cancer surgery survivors. Overall, breast cancer survivors had a slightly lower risk of stroke than cancer-free women (sHR 0.94; 95% CI 0.88–1.00). However, stroke risk was elevated in the short term following diagnosis (sHR 1.59, 95% CI 1.34–1.89 at 1 year; sHR 1.17, 95% CI 1.05–1.30 at 3 years) across all age groups, with stronger associations observed at 3 and 6 months post-diagnosis. A reduced risk was observed after 1 year in a landmark analysis that included only individuals event-free at the 1-year follow-up (sHR 0.87, 95% CI 0.81–0.93). Among breast cancer survivors, treatment with anthracycline (sHR 1.25) and the combination of tamoxifen and aromatase inhibitors (sHR 1.49) were associated with increased stroke risk, whereas radiation therapy was associated with decreased risk (sHR 0.84). These associations weakened and became nonsignificant after 1 year. Stroke risk was also higher in breast cancer survivors with low income, hypertension, diabetes, or current smoking. CONCLUSION: The association between breast cancer and ischemic stroke risk is time-dependent, with increased short-term risk post-diagnosis and treatment, followed by a decline over time. These findings highlight the importance of proactive stroke risk management, including baseline cardiovascular assessments and ongoing monitoring for thromboembolic events.

Abstract Aims The Society for Cardiovascular Angiography and Interventions (SCAI) Classification provides risk stratification of patients with acute myocardial infarction complicated by cardiogenic shock (AMI-CS). This sub-study of the ECLS-SHOCK trial investigates the prognostic impact of SCAI stages in AMI-CS and the influence of SCAI stages on the effect of extracorporeal life support (ECLS) therapy in AMI-CS patients. Methods Patients with AMI-CS enrolled in the multicentre, randomized ECLS-SHOCK trial were included. The outcomes, treatment effect and safety of ECLS were stratified according to SCAI stage at admission using a post-hoc classification. Results From a total of 417 patients enrolled in the ECLS-SHOCK trial between June 2019 and November 2022, 51.6% (n = 215), 13.4% (n = 56) and 35.0% (n = 146) presented in SCAI Stages C, D and E, respectively. SCAI stages were associated with the risk of 30 day all-cause mortality (C vs. D vs. E: 32.6% vs. 67.9% vs. 64.4%, P < 0.001), with rates of renal replacement therapy at 30 days (C vs. D vs. E: 7.0% vs. 19.6% vs. 13.7%, P = 0.03) and with poor neurological outcomes (C vs. D vs. E: 17.2% vs. 44.4% vs. 36.5%, P < 0.001). No interaction was observed between SCAI stage and the treatment effect of ELCS on 30 day all-cause mortality (ELCS vs. control SCAI C: 32.7% vs. 32.4%; SCAI D: 68.4% vs. 66.7%; SCAI E: 59.7% vs. 68.4%, P for interaction = 0.65). Conclusions In AMI-CS patients included in the ECLS-SHOCK trial, SCAI stages at admission were predictive for mortality and for the incidence of safety events. The efficacy of ECLS treatment was not affected by SCAI stage.

Importance Accumulation of amyloid plaque drives pathogenesis of Alzheimer disease (AD). Reduction of amyloid via amyloid-targeting therapies may result in clinical benefit. Objective To assess the correlation of posttreatment amyloid levels with clinical outcomes and biomarkers in AD. Design, Setting, and Participants This was a post hoc exploratory analysis from the randomized, placebo-controlled phase 3 trial, TRAILBLAZER-ALZ 2, conducted June 2020 through April 2023 at 277 medical research centers/hospitals in 8 countries. A total of 8240 participants aged 60 to 85 years with early symptomatic AD with amyloid and tau pathology based on positron emission tomography (PET) imaging were assessed for eligibility. Of these, 6504 participants were excluded predominantly due to inadequate amyloid or tau pathology. The current analysis included 1582 participants (766 in the donanemab group and 816 in the placebo group) with baseline and at least 1 posttreatment assessment. Data analysis took place from July 2024 to March 2025. Interventions Participants were randomized 1:1 to receive donanemab (700 mg for the first 3 doses and 1400 mg thereafter) or placebo intravenously every 4 weeks for up to 72 weeks, with outcomes assessed through 76 weeks. Main Outcomes and Measures Participants were categorized into 1 of 10 groups (deciles) based on their lowest amyloid value observed posttreatment. Clinical progression was assessed via changes in integrated Alzheimer’s Disease Rating Scale (iADRS) and Clinical Dementia Rating–Sum of Boxes (CDR-SB) scores. Plasma biomarkers measured included phosphorylated tau 217 (p-tau217), p-tau181, glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL). Correlations between the median amyloid level in each decile were assessed with 76-week least-squares mean changes in each outcome and biomarker. Results Analyses included 1582 participants, including 766 treated with donanemab and 816 with placebo. The mean (SD) age was 72.9 (6.2) years, and 900 participants (56.9%) were female. Participants who received donanemab had lower posttreatment amyloid values than those in the placebo group. Across the trial population, lower posttreatment amyloid levels were correlated with slower clinical progression as measured by iADRS score ( R 2 , 0.73 [95% CI, 0.37-0.97]) and CDR-SB score ( R 2 , 0.87 [95% CI, 0.70-0.97]) and with decreases in p-tau217 ( R 2 , 0.86 [95% CI, 0.73-0.97]), p-tau181 ( R 2 , 0.88 [95% CI, 0.77-0.97]), and GFAP ( R 2 , 0.87 [95% CI, 0.76-0.97]). There was no correlation between posttreatment amyloid value and NfL ( R 2 , 0.03 [95% CI, 0.00-0.54]). Conclusions and Relevance The findings in this secondary analysis of a randomized clinical trial demonstrating a correlation between posttreatment amyloid plaque level and clinical benefit support amyloid plaque removal as the mechanism of action for donanemab treatment and the level of amyloid plaque as a potential surrogate biomarker in amyloid-targeting therapies. Trial Registration ClinicalTrials.gov Identifier: NCT04437511

Abstract Background Dystonia and Parkinson's disease (PD) exhibit clinical and genetic overlap, but the relevance of dystonia gene variants in PD remains unclear. Objective The aim was to assess the frequency of dystonia‐linked pathogenic variants in PD. Methods We screened sequencing data from 15,684 individuals (8272 PD, 3200 atypical parkinsonism, and 4212 unaffected) from the Global Parkinson's Genetics Program (GP2) and Accelerating Medicines Partnership‐Parkinson's Disease (AMP‐PD) for variants in genes linked to isolated dystonia, dystonia‐parkinsonism, and myoclonus‐dystonia. Results Pathogenic variants were identified only in PD patients. Forty‐five PD individuals (0.54%) carried 26 distinct (likely) pathogenic variants in nine dystonia‐linked genes, most frequently in GCH1 , followed by VPS16 . Conclusion Though rare, pathogenic variants in dystonia‐linked genes are present in clinically and pathologically diagnosed PD. Our results reinforce GCH1 as a PD‐relevant gene with clinical implications, whereas variants identified in other genes are rare and of uncertain relation to the PD phenotype. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.