Citations

Abstract Background Standardizing beam modeling in the treatment planning system is challenging due to machine‐specific configuration variations. Because of the proton therapy system (PTS) complexity and continuous development, there is no vendor's representative (“golden”) beam data like in LINACs at this stage. Purpose The primary goal of this study is to create a PTS‐specific repository of source model parameters and beam commissioning data for scanning proton therapy treatment planning systems and establish a set of reference values with associated bands of variations for this data. Methods We conducted a survey of 31 PTSs at 27 proton therapy centers to gather beam commissioning source model parameters used in their treatment planning system, ensuring a diverse selection of PTSs from major vendors: Hitachi, IBA, Varian, and Mevion. For each proton therapy center, we collected (i) beam meterset calibration in protons per MU, (ii) in‐air spot size at isocenter in terms of standard deviation in x‐ (crossline, σ x ) and y‐ (inline, σ y ) directions, (iii) in‐air angular spread at isocenter in x‐direction (θ x ) and y‐direction (θ y ) and (iv) distal falloff. For each PTS, we plotted each parameter as a function of nominal energy and performed polynomial regression analysis with optimal model selection based on Akaike Information Criterion. We calculated intraclass correlation coefficients (ICC) to quantify inter‐PTS variability and generated uncertainty bands representing standard deviations across PTSs for each parameter. Additionally, we computed percentage deviations of each individual PTS from the overall multi‐vendor mean to identify systematic vendor‐specific patterns and outliers across the 31 total PTSs analyzed. Results Multi‐vendor beam parameter analysis across 31 PTSs revealed systematic vendor‐specific patterns in inter‐PTS consistency and energy dependencies. Beam meterset calibration demonstrated excellent inter‐PTS consistency across most vendors (ICC: Varian ProBeam ≈ 0, IBA Proteus ONE = 0.006, Mevion S250i with HYPERSCAN = 0.015), with IBA Proteus PLUS systems showing moderate variation (ICC = 0.156 for the dedicated nozzle, ICC = 0.248, for the universal nozzle). Spot size and angular spread variability were highly vendor‐dependent, ranging from excellent consistency in IBA Proteus ONE to substantial variation in IBA Proteus PLUS with the universal nozzle (angular spread ICC = 0.566). Distal falloff maintained excellent consistency across all vendors (ICC < 0.023), indicating reliable energy spread characteristics. Uncertainty band analysis revealed characteristic energy‐dependent patterns, with most systems showing larger uncertainties at lower energies for spot size and angular spread, while IBA Proteus PLUS with the universal nozzle exhibited the largest uncertainty bands overall. Percentage deviation analysis demonstrated vendor‐specific clustering patterns, with some systems showing systematic positive or negative deviations from the multi‐vendor mean. Conclusions The reference data provided in this work will help to cross‐check the beam model during treatment planning system commissioning. Furthermore, this information is useful for Monte Carlo applications and research, where beam characteristics are required to set up phase space parameters, thereby avoiding the need to model the entire PTS. Finally, the results of this study will have indispensable educational value by providing valuable insights into the characteristics and capabilities of different PTSs.

Abstract Background: For large language models (LLMs) to reach their potential as information technology tools that make medication use safer, clinically relevant benchmarks capable of automated grading and designed specifically to measure the performance of LLMs for medication tasks are required. The purpose of this study was to design a suite of benchmarking tests reflective of Comprehensive Medication Management (CMM; the standard of care for medication optimization) and quantify the baseline performance of the latest LLMs. Methods: We established six benchmarks representing critical stages of the CMM process: drug formulation matching, drug order (sig) generation, drug route matching, drug-drug interaction identification, renal dose identification, and drug-indication matching. For each benchmark, we curated a clinician-annotated dataset comprising 250 standardized input-output pairs including both inpatient and outpatient medications. We evaluated the clinical knowledge retrieval capabilities of three LLMs: GPT-4o-mini, MedGemma-27B, and LLaMA3-70B. We employed a zero-shot prompting strategy, excluding in-context examples, to assess the models’ internal clinical knowledge rather than their few-shot learning potential. To check reliability, each model was run three times using a temperature of 0.7 (a mid-range value of an LLM setting controlling text generation randomness). Performance was assessed using task-specific evaluation metrics including precision (positive predictive value), recall (sensitivity), F1-score, accuracy, and correctness consistency across trials. Results: Across six benchmarks, LLaMA3-70B demonstrated the highest performance in four tasks: drug-formulation matching (F1, 54.0% [95 CI: 50.1-58]), drug-order generation (accuracy, 88.0%), drug-route identification (F1, 74.3% [95 CI: 71-78]), and drug-indication identification (accuracy, 97.6% [95 CI: 95.6-99.2]). In the drug–drug interaction task, GPT-4o-mini achieved the highest overall accuracy (70.4% [95 CI: 64.8-75.7]). For renal dose–adjustment identification, GPT-4o-mini demonstrated the highest F1 score (83.3% [95 CI: 77.6-88]). Correctness-consistency scores ranged from 8.0% to 97.6% across benchmarks, with no model exhibiting uniformly superior consistency. Conclusions: Model performance varied substantially across medication-related tasks. LLaMA3-70B demonstrated promising baseline performance in tasks involving formulation, ordering, route, and indication. GPT-4o-mini showed potential advantages in drug–drug interaction detection and renal dose adjustment. These findings underscore the need for task-specific evaluation when deploying models for medication-focused clinical decision support.

Abstract Background The purpose of this study was to develop a series of benchmarks specifically for large language model (LLM) performance in the identification of drug-drug interactions (DDIs). Methods We evaluated six LLMs using a clinician-designed dataset of 750 DDI scenarios. Tasks included: (1) a pointwise two-drug classification task (2) a pairwise three-drug discrimination task and (3) a listwise 4–6 drug selection task. Zero-shot prompting was applied and performance was assessed using precision, recall, F1 score, accuracy, and self-consistency. Results Model performance varied by task structure and interaction severity. GPT-5-Chat demonstrated the highest recall and F1 score in the pointwise task (80.7% and 80.2%), whereas DrugGPT achieved superior accuracy in the pairwise (93.7%) and listwise task (92.6%). All models had improved performance with more severe compared to less severe DDIs (accuracy 72.0-99.2% vs. 58.3-86.7%). All models tended to have higher self-consistency in the pairwise and listwise experiments compared to the pointwise experiment (68.8-93.6% vs. 31.0-74.2%). No model exhibited uniformly high reliability across all tasks. Conclusions Current LLMs show inconsistent capabilities in identifying DDIs. Performance degrades as reasoning space expands with limited stability. These findings emphasize the need for multi-format evaluation frameworks and reliability-aware assessment when considering LLMs for medication-safety applications.

Abstract Background: Intravenous (IV) isatuximab plus pomalidomide and dexamethasone (Isa-Pd) is approved for treatment in patients with relapsed/refractory multiple myeloma (RRMM) based on the ICARIA-MM study. Subcutaneous (SC) administration of Isa offers potential advantages such as shorter administration duration and greater patient comfort compared with IV routes. Isa SC via an on-body injector (OBI), an investigational wearable bolus injector, demonstrated safety and efficacy in a Phase 1b study in patients with RRMM. The IRAKLIA trial (NCT05405166) was the first Phase 3 trial evaluating the non-inferiority of Isa-Pd SC via an OBI vs Isa-Pd IV in patients with RRMM. Here, we report the findings of the IRAKLIA trial in the East Asian subgroup with RRMM. Methods: This global, multicenter, open-label study enrolledpatients aged ≥18 years with RRMM who had received ≥1 prior line of therapy (LOT) including lenalidomide and a proteasome inhibitor (PI). Patients were randomized 1:1 to Isa SC (1400 mg) or Isa IV (10 mg/kg) weekly in Cycle (C)1, then every 2 weeks in addition to P (4 mg orally daily, Day [D]1-21) and d (40 mg orally [20 mg if age ≥75 years] weekly). Treatment was administered in 4-week cycles and continued until disease progression, unacceptable toxicity, or patient withdrawal. Co-primary endpoints included overall response rate (ORR) and observed Isa concentration before dosing (Ctrough) at steady state (predose at C6D1). Key secondary endpoints included rate of very good partial response or better (≥VGPR), Ctroughlevel Cycle 2 Day 1 (C2D1), infusion reaction incidence, and patient satisfaction rate based on the patient experience and satisfaction questionnaire (PESQ) at C5D15. Results: The East Asian subgroup consisted of 105 patients from China (Mainland, n=82 and Taiwan, n=1) and Japan(n=22) (SC, n=50; IV, n=55). Baseline characteristics of the East Asian subgroup were generally balanced between the two arms and generally comparable to the overall global population (median age 62 years; median 2 prior LOT; 29.5% high-risk cytogenetics; 82.9% refractory to lenalidomide; 7.6% prior exposure to daratumumab). After a median of 13.0 months follow-up, ORR in the SC and IV arm was 74.0% and 70.9%, respectively (relative risk [95% CI] = 1.044 [0.815 – 1.335]). The mean (SD) of Ctroughat C6D1 was 502 (232) µg/mL for SC arm and 282 (123) µg/mL for IV arm. The geometric mean ratio (90% CI) comparing SC to IV for Ctrough at steady state was 1.756 (1.346 – 2.290). The ≥VGPR rates were 58.0% for SC and 50.9% for IV (relative risk [95% CI] = 1.139 [0.798 – 1.631]). The mean(SD) of C2D1 Ctroughlevel was 420 (161) µg/mL for the SC arm and 262 (111) µg/mL for IV arm. The geometric mean ratio (90% CI) comparing SC to IV for Ctrough at C2D1 was 1.673 (1.321 to 2.120). The infusion reaction incidence was 0% and 21.8% for SC and IV, respectively. Grade ≥3treatment-emergent adverse events were observed in 82.0% and 87.3% of patients in the SC and IV arms, respectively. Treatment discontinuation rates were 6% for the SC arm and 12.7% for the IV arm. Injection site reactions (ISRs) occurred in 4% (2/50) of patients in the SC arm and in 0.21% (2/964) of SC injections (all Grade 1), with 99.3% of SC injections completed successfully. Isa SC patients expressed higher satisfaction with their injection method, with 64.0% of patients reporting “satisfied” or “very satisfied” vs 52.7% of Isa IV patients (odds ratio [95% CI] = 1.594 [0.728 – 3.489]). Conclusions: The efficacy, safety and pharmacokinetic results between Isa SC vs IV + Pd in the East Asian population from the IRAKLIA study were consistent with the results of the overall global population, in which non-inferiority of ORR, Ctrough at C6D1, ≥VGPR and Ctrough at C2D1 was demonstrated and statistically significantly fewer infusion reactions and higher patient satisfaction were also noted for SC vs IV. These results support the potential use of Isa SC delivered via OBI in East Asian patients and demonstrate the potential of this delivery method to improve patient experience and practice efficiency.

Key Points Inositol polyphosphate-5-phosphatase E (INPP5E) and type Ig phosphatidylinositol-4-phosphate 5-kinase (PIPKIγ) coordinated the homeostasis of phosphatidylinositol-4-phosphate and phosphatidylinositol-4,5-bisphosphate in primary cilia. Modulating INPP5E or PIPKIγ activity changed the level of polycystin-1 and polycystin-2 in primary cilia. INPP5E inhibition increased the hypomorphic PKD1 variants (PC1-R3277C) in cilia and reduced cystogenesis in the kidney in vitro . Background Autosomal dominant polycystic kidney disease is mainly caused by mutations in PKD1 and PKD2 , which encode polycystin-1 (PC1) and polycystin-2 (PC2), respectively. PC1 and PC2 assemble a cation channel complex enriched in primary cilium, a sensory organelle associated with various developmental diseases, including polycystic kidney disease (PKD). Accumulating evidence supports the necessity of functional polycystin (PC) complex in cilia to prevent cystogenesis in the kidney, indicating that improving their ciliary levels may ameliorate defects underlying PKD pathogenesis. Yet, molecular mechanisms underlying the ciliary targeting and homeostasis of the PC complex are not fully understood. Methods Indirect immunofluorescence microscopy was used to monitor ciliary levels of PC1 and PC2 in renal epithelial cells. Electrophysiology analysis in oocytes was used to determine the channel activity of the PC complex. Cystogenesis in the kidney was measured using in vitro 3D-Matrigel cell models and ex vivo mouse embryonic kidney models. Results Suppressing inositol polyphosphate-5-phosphatase E (INPP5E) or activating type Ig phosphatidylinositol-4-phosphate 5-kinase raised ciliary levels of the PC complex in both normal renal epithelial cells and cells carrying autosomal dominant polycystic kidney disease mutations that interrupt the trafficking of PCs into cilia, including GANAB inactivation and the trafficking PKD1 mutation p.Arg3277Cys (RC). PC1 RC formed a complex with PC2 and exhibited normal channel activity in vitro . An INPP5E that increases PC1 and PC2 in cilia, suppressed in vitro forskolin-induced cystogenesis of inner medullary collecting duct epithelial cell line 3 cells in 3D Matrigel and ex vivo cyst formation in embryonic Pkd1 RC/RC mouse kidneys. Conclusions Our results demonstrated that increasing the ciliary level of PCs, by manipulating a ciliary phosphoinositide signaling axis, enhanced the functionality of PCs and suppressed cystogenesis of renal epithelial cells in vitro .

Abstract Background DNA methylation profiling can be used to robustly predict postsurgical outcomes and response to radiotherapy (RT) for meningioma patients. To allow for seamless integration of these complementary models into clinical practice, a practical framework is needed. Methods We leveraged a cohort of nearly 2000 surgically-treated meningiomas with DNA methylation profiling and clinical outcomes data. Existing methylation-based prediction models were dichotomized to yield four risk groups: low and high recurrent risk, each with RT sensitive and resistant subgroups. Risk groups were correlated with progression-free survival in the context of existing biomarkers including extent of resection and WHO grade. Results We first demonstrated that all risk groups benefit from gross total resection. All “high-risk, RT sensitive” tumors (n = 306, 15.7%) also benefited from adjuvant RT: after GTR, median PFS increased from 4.68 (4.13-9.48) years to not reached (P = .003); after subtotal resection (STR), from 2.12 (1.59-3.02) to 4.09 (3.41-not reached) years (P = .004). “Low-risk, RT sensitive cases” (n = 1207, 61.8%) also benefited from RT after STR (median PFS 7.39 (6.66-12.8) vs. 16.53 (10.35-not reached) years, P = .03), suggesting that RT be considered in these patients. Neither “low-risk RT resistant” (n = 84, 4.3%) nor “high-risk RT resistant” (n = 356, 18.2%) cases benefited from RT, and the latter group was associated with universally poor outcomes. Conclusions We identify methylation-defined risk groups of meningioma for which additional benefit is gained from adjuvant RT, leading to a clinical decision-making framework for straightforward integration of molecular models into clinical practice.

Abstract A new mouse model for the classical subtype of human glioblastoma (GBM) has been bred using Cre-mediated EGFRvIII overexpression and homozygous p19-ARF deletion, the mouse homolog of human p14ARF/CDKN2A, in GFAP expressing cells. Transgenic mice develop intraparenchymal and/or leptomeningeal brain lesions with some spinal cord invasion as early as 1 month old and 95% of mice die by 6 months due to hydrocephalus and/or paralysis. Mice with high grade tumors have worse survival and similar features to human classical GBM such as necrosis, high levels of mitosis and infiltration of tumor cells into normal brain. Immunohistochemical analysis confirms EGFRvIII overexpression and p19-ARF loss in Cre-expressing cells, along with patchy positive GFAP and high proliferation by Ki67. Bulk RNA sequencing reveals GEC3 tumors cluster closely with human classical GBM and have significant upregulation of downstream markers in the JAK/STAT and P13K/AKT pathways. Adherent and neurosphere primary culture of dissociated tumors indicate that tumor cells maintain EGFRvIII expression in culture and generate xenograft tumors by 3 weeks after intracranial injections into NODSCID mice. Xenograft tumors are reminiscent of the primary tumor, with similar histopathological features and immunohistochemical staining. Our spontaneous glioma model is a powerful tool for future studies focused on the role of the immune microenvironment on glioma recurrence and resistance.

Abstract Recurrent or refractory IDH-mutant (mt) gliomas represent a population with significant unmet clinical need. The Notch ligand delta-like 3 (DLL3) is a tumor-associated antigen that is highly and homogeneously expressed in IDH-mt gliomas but absent in normal brain tissue, making it an attractive therapeutic target. Tarlatamab is a half-life extended bispecific T-cell engager that targets DLL3 and CD3 to redirect cytotoxic T cells toward DLL3-expressing tumor cells. Given the specific expression of DLL3 in IDH-mt gliomas and promising activity of tarlatamab in other DLL3-positive tumors, this study investigates the safety, efficacy, and intracranial activity of tarlatamab in this patient population. TARGID is an open-label, phase II trial enrolling adult patients with histologically confirmed IDH1 or IDH2-mt astrocytoma or oligodendroglioma following up to two prior lines of systemic therapies. The study comprises two cohorts. Cohort 1 (n=10) includes patients with resectable disease who will receive up to three cycles of tarlatamab prior to surgery; postoperative continuation is permitted. The primary endpoint for this cohort is the change in CD8+ T-cell infiltration in tumor tissue and blood. Cohort 2 (n=34) includes patients with unresectable disease and uses a Simon’s optimal two-stage design (H0: P ≤ 0.1 versus H1: P ≥ 0.25). In the 1st stage, 13 patients will be accrued. If 2 or more responses are observed, 21 additional patients will be enrolled. Tarlatamab is administered intravenously at 10 mg every two weeks, following a 1 mg priming dose on Cycle 1 Day 1. The primary endpoint is objective response rate by RANO 2.0. Exploratory endpoints include DLL3 expression, circulating biomarkers, and immune correlatives. The trial is currently in activation, with first patient enrollment anticipated in Q3 2025. The study is supported by the provision of study drug from Amgen as an investigator-initiated trial.

Abstract BACKGROUND Post-operative prognostication in glioma patients remains a clinical challenge, particularly given the heterogeneity in tumor biology and patient outcomes. We evaluated whether image-derived features from post-operative imaging improve survival modeling beyond established clinicopathologic factors. METHODS We analyzed a cohort of 71 post-surgical glioma patients treated at Princess Margaret Cancer Centre with known clinicopathologic features including age, histological classification, IDH mutation status, MGMT methylation status, extent of resection, and performance status. A baseline Cox proportional hazards model was fit to clinical variables. Radiomic features were extracted using PyRadiomics from post-operative CT and MRI across three regions of interest: whole brain, tumor bed, and peritumoral ring. Features underwent stepwise unsupervised reduction (variance filtering, exclusion of volume-correlated features, and removal of remaining collinear features), followed by dimensionality reduction via principal component analysis. Combined clinical and imaging models were evaluated in the overall cohort and stratified by IDH mutation status. RESULTS In the overall cohort, the clinical model was highly prognostic (C-index 0.74, p ~ 3.2×10⁻⁶). Adding imaging features improved performance (C-index 0.79) but was not significant (p ~ 0.31). Subset analysis revealed differential effects by IDH status. In IDH-mutant gliomas (n=24), neither clinical nor combined models were significant. In IDH wild-type glioblastoma (n=47), the clinical model achieved C-index 0.75 (p=0.04). The addition of imaging features significantly improved model performance (C-index 0.83, p=0.03), with selected imaging components contributing independent prognostic value after adjustment for clinical variables. Post-hoc analysis of radiomic feature importance (random forest ranking) revealed that top contributing features were drawn from both CT and MRI modalities, with most sourced from the whole brain region. CONCLUSIONS In post-surgical patients with IDH wild-type glioblastoma, image-derived features from post-operative imaging can significantly enhance prognostic modeling beyond known clinicopathologic factors. These preliminary findings support the potential role of radiomic biomarkers in refining risk stratification for this patient subgroup.

Abstract INTRODUCTION While TERT promoter mutation (TPM) has been established as a marker of clinically aggressive meningiomas, this alteration is rare and found in less than 5% of all cases. However, a larger subset of meningiomas may exhibit aberrant TERT expression in the absence of TPMs. This study investigated the effect of TERT gene expression on clinical outcome in meningioma patients. METHODS Clinical and molecular data were retrospectively collected on 1241 meningiomas, split into a Toronto discovery cohort and a multi-institutional validation cohort. Sanger sequencing and bulk RNA sequencing were used to determine TPM status and TERT gene expression. The effect of TERT expression on progression-free survival (PFS) was assessed using Kaplan-Meier and Cox regression analysis. RESULTS While meningiomas with TPM showed expectedly higher TERT gene expression compared to wildtype (TP-WT) cases (p&lt;0.0001), TERT expression was still detected in 28.7% (157/547) of TP-WT meningiomas. Meningiomas with TERT expression showed significantly worse PFS compared to meningiomas without any TERT expression. In fact, WHO grade 1 meningiomas with TERT expression had PFS outcomes resembling WHO grade 2 meningiomas lacking TERT expression (p=0.59). In turn, WHO grade 2 meningiomas with TERT expression had clinical outcomes similar to WHO grade 3 meningiomas without TERT expression (p=0.42). Furthermore, the proportion of meningiomas expressing TERT as well as overall TERT expression levels increased with increasing WHO grade. Multivariable analysis showed that TERT expression was significantly associated with worse PFS even when controlling for other known predictors of clinical outcome including TPM, CDKN2A/B loss, 1p/22q status and WHO grade (HR 1.85 [95% CI 1.33-2.57], p=0.00024). CONCLUSION TERT expression is a novel independent biomarker of outcome for meningiomas identifiable in up to one-third of cases that may be utilized to reclassify tumours to a higher WHO grade.