Citations

Abstract Background Accurate and efficient dose calculation is essential for online adaptive planning in proton therapy. Deep learning (DL) has shown promising dose prediction results for pencil beam scanning proton therapy (PBSPT) in recent years, but existing DL‐based dose prediction methods still suffer from limited generalizability and an inability to effectively handle outlier clinical cases. This may lead to inaccurate dose delivery to targets or excessive irradiation to organs at risk (OARs), thereby compromising the safety and efficacy of online adaptive proton therapy. Purpose To design a physics‐aware and generalizable AI‐based PBSPT dose prediction method that incorporates underlying physics to enhance generalizability, particularly in handling outlier clinical cases. Methods This study analyzed PBSPT plans of 103 prostate (93 for training and 10 for testing) and 78 lung cancer patients (68 for training and 10 for testing) from our institution, with each case comprising CT images and structure sets. Using the doses generated by our Monte Carlo‐based dose engine as the reference standard, we compared three methods: the region of interest (ROI)‐based method, the beam mask and sliding window method, and the proposed noisy probing dose method, which rapidly generates a low‐statistics dose via uniformly weighted spots on an expanded spot‐placement target volume without optimization. To evaluate the generalizability of these methods to rare treatment planning scenarios, 12 cases with uncommon beam angles or prescription doses were used to assess their performance, which was evaluated using dose‐volume histogram (DVH) indices, 3D Gamma passing rates (3%/2 mm/10%), and Dice coefficients for dose agreement, while prediction times were measured to gauge model efficiency. Results The proposed noisy probing dose method consistently outperformed the ROI‐based and beam mask baselines across all evaluated metrics, with more accurate dose agreement and superior generalizability. For DVH indices, the noisy probing dose method achieved the smallest deviation in clinical target volume (CTV) dose coverage: in prostate cancer, CTV D98 deviation was reduced by 45% (from 0.53 ± 0.22 Gy [RBE] for ROI‐based) and 29% (from 0.41 ± 0.28 Gy [RBE] for beam mask) to 0.29 ± 0.06 Gy [RBE]; in lung cancer, similar improvements were observed, with CTV D98 deviations reduced to 0.34 ± 0.12 Gy [RBE]. The 3D Gamma passing rates improved to 99.65% ± 1.15% for prostate targets and 97.04% ± 1.17% for lung targets. The dice coefficients of the 90% iso‐dose lines were also the highest with the noisy probing dose method (prostate: 0.983 ± 0.005; lung: 0.967 ± 0.01). For the 12 outlier cases, the noisy probing dose method maintained superior generalizability, yielding higher 3D Gamma passing rates (prostate targets: 96.79% ± 0.83%, OARs: 94.29% ± 1.01%; lung targets: 93.38% ± 1.34%, OARs: 93.95% ± 1.32%), demonstrating robust generalizability to rare clinical scenarios. The dose predictions for all testing cases were completed within 0.3 seconds. Conclusions A novel noisy probing dose method was proposed for PBSPT dose prediction in prostate and lung cancer patients. By embedding more proton‐specific physics, this method demonstrated an improvement in the generalizability of dose prediction.

OBJECTIVE Frailty is a known predictor of adverse outcomes in pituitary surgery, yet its impact regarding craniopharyngioma remains unclear. The authors assessed preoperative frailty, measured by the 11-factor modified frailty index (mFI-11), and its association with perioperative morbidity among adult patients undergoing craniopharyngioma surgery. METHODS Using the Registry for Adenomas of the Pituitary and Related Disorders (RAPID) database, the authors performed a retrospective cohort study of the records of patients ≥ 18 years of age who underwent craniopharyngioma surgery between June 2007 and November 2024. Patients were stratified into the following categories according to their numeric mFI-11 score: fit (scores 0 and 1), managing well (scores 2 and 3), and mildly frail (scores 4–6). Primary outcomes included hospital length of stay (LOS) and discharge disposition. Secondary outcomes included 90-day readmission, surgical outcomes, and complications. Multivariable analyses were performed to identify independent predictors of hospital LOS and non-home discharge, adjusting for age, surgeon experience, American Society of Anesthesiologists class, complications, and surgical approach. RESULTS Among 278 patients (221 fit, 43 managing well, 14 mildly frail), higher-frailty patients were older (mean age 47.8 ± 16.1 years in fit vs 60.1 ± 12.5 years in managing well vs 67.6 ± 8 years in mildly frail, p < 0.001) and more likely to present urgently (24.2% vs 36.4% vs 78.6%, p = 0.001). Gross-total resection was achieved less frequently in frail patients (44.8% vs 24.3% vs 23.1%, p = 0.030). Overall complication rates were similar; however, increasing frailty was associated with higher rates of postoperative mortality (0% vs 2.8% [1/36] vs 7.7% [1/13], p = 0.018), pneumonia (0.5% [1/195] vs 2.8% [1/36] vs 15.4% [2/13], p = 0.006), and reintubation (2.6% [5/195] vs 2.8% [1/36] vs 30.8% [4/13], p = 0.002). Higher frailty was associated with a stepwise rise in median [IQR] LOS (5 [3–8] days vs 7 [4–12] days vs 13 [6.5–21.8] days, p < 0.001) and remained an independent predictor of extended LOS on multivariable analysis (β = 1.729 per point, p = 0.006). Rates of non-home discharge showed a similar pattern (9.2% vs 30.0% vs 64.3%, p < 0.001), with frailty independently predicting non-home discharge (OR 2.352, 95% CI 1.527–3.822). Unplanned 90-day readmission rates were significantly higher in the mildly frail cohorts (22.5% vs 45.7% vs 53.8%, p = 0.002). CONCLUSIONS The mFI-11 scores independently predicted perioperative morbidity following craniopharyngioma resection. Incorporating frailty screening into preoperative screening may improve risk stratification, family and patient counseling, and discharge planning.

Abstract Heterogeneous treatment effects (HTE) are key to precision medicine, but most real-world studies lack the scale and diversity needed to detect them. While multi-site analyses offer a potential solution, data-sharing constraints often prevent access to patient-level information across institutions. We introduce POLARIS, a federated framework for time-to-event target trial emulation. POLARIS converts each site’s weighted Cox risk function into a compact tensor shared once with the coordinating center, enabling lossless reproduction of pooled estimates without sharing patient-level data. We applied POLARIS across five U.S. health systems to study risk of gastrointestinal outcomes after GLP-1 receptor agonist (GLP-1RAs) initiation versus sodium-glucose cotransporter 2 inhibitors (SGLT2is) and dipeptidyl peptidase 4 inhibitors (DPP4is). Results showed that GLP-1RAs were consistently associated with higher risks of nausea and vomiting, particularly among men, individuals with higher baseline HbA1c (≥ 8.5%), and lipid therapy. POLARIS provides a scalable solution for distributed target trial emulation and fine-grained assessment of HTE across diverse health systems.

Background/Objectives: Accurate detection of pulmonary nodules on chest CT is critical for lung cancer screening, yet training robust detectors remains challenging due to the high cost of reliable annotations. In this work, we present a systematic study of curriculum learning for CT-based lung nodule detection on the enhanced LUNA25 benchmark and propose a hybrid curriculum learning framework for data-efficient optimization. Methods: Our approach estimates sample difficulty by fusing clinically interpretable handcrafted factors such as nodule size and count with model-driven signals such as prediction confidence from a teacher model, and constructs a three-stage progressive training curriculum from easy to hard samples. Using YOLOv11s as a strong baseline, the proposed hybrid curriculum is compared against conventional training without curriculum learning and against single-source curricula. Results: On the held-out LUNA25 test set, hybrid curriculum learning increases mAP50 from 0.672 to 0.696, mAP50–95 from 0.369 to 0.385, recall from 0.588 to 0.634, and precision from 0.725 to 0.764. Extensive data-efficiency experiments with proportional reductions (1/2, 1/5, 1/10) and fixed training samples (5000–40,000 slices) further confirm consistent gains across limited-data regimes. Conclusions: These results demonstrate that jointly leveraging intrinsic image complexity and optimization-aware feedback provides effective sample scheduling for robust and data-efficient lung nodule detection.

Abstract Background: Early detection is the most potent strategy to improve survival in PDA, but its success hinges on an unresolved question: does radiologically invisible disease represent biologically early or indolent disease, or established malignancy beneath the threshold of detection? We leveraged prediagnostic CTs—incidental scans obtained 3-36 months prior to diagnosis without a focal mass—to define the relationship between imaging phenotype and tumor biology. Methods: From a multi-institutional pool of 5,046 biopsy-confirmed PDAs, we assembled a cohort of 346 prediagnostic CTs, which were independently evaluated by three radiologists using a standardized protocol. Based on majority vote, CTs were dichotomized into imaging-occult (IO) (normal pancreas) and imaging-apparent (IA) (abnormal pancreas with indirect signs such as focal atrophy, duct dilatation, etc.) groups. Tumor size and metastases (M) at diagnosis, and overall survival (OS) were compared using multivariable linear (size), logistic (M1), and Cox proportional hazards (OS) models, adjusting for lead time to diagnosis, age, sex, diabetes, CA 19-9, and surgical resection status, to isolate the independent biological relevance of the imaging phenotype. Results: IO phenotype was seen in 150 patients (mean age 66.7±11.2; 37.3% females) while 196 had IA phenotype (mean age 69.9±10.6; 49.5% females). Median lead time was 396 days (range: 90-1088) with the IA group being enriched in 3-12-month lead time (57.7% vs 33.3%; p&lt;0.001). Both groups had comparable diabetes rates (IA 33.2% vs IO 32.7%; p=1.0), median CA 19-9 (IA 193 vs IO 322 U/mL; p=0.41), and surgical resection (IA 38.3% vs IO 38.0%; p = 1.0). While unadjusted mean tumor size at diagnosis was larger in IA group (3.8 cm vs 3.2 cm), there was no difference after adjusting for clinical factors (-7.1% difference; 95% CI: -17.2, 4.2; p=0.21). Metastases were also comparable (M rate: IA 22.4% vs IO 27.3%; adjusted OR 0.97; 95% CI: 0.50, 1.88; p=0.93). Adjusted Cox models demonstrated equivalent OS (median: 18.7 vs 17.6 months; HR 1.06; p=0.66), which persisted even when stratified by surgical status (HR 1.07; 95% CI: 0.83, 1.38; p=0.61). Even in the critical 3-12 months lead time (n=163), OS remained equivalent (HR 0.89; p=0.62), and groups did not differ in either tumor size (-6.4%; 95% CI: -22.4, 12.9; p=0.49) or metastases (OR 0.48; 95% CI 0.14, 1.59; p=0.23). Conclusions: A normal-appearing pancreas on prediagnostic imaging is common yet signifies neither early nor indolent disease—it harbors PDA biologically equivalent to the imaging-apparent pancreatic phenotype. Because the disease at this stage lies beyond human perceptual capacity, AI augmentation represents the only potential path to realizing the survival benefit of early detection. Citation Format: Khurram Khaliq Bhinder, Sovanlal Mukherjee, Armin Zarrintan, Angela Ammirabile, Ahmed Khan Jadoon, Ziding Feng, Suresh T. Chari, Ajit H. Goenka. Dissociation between prediagnostic imaging phenotype and tumor biology defines the early detection challenge in pancreatic ductal adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB205.

Diagnosing rare genetic diseases remains a major challenge despite widespread clinical testing. Long-read RNA sequencing (RNA-seq) offers a powerful approach to capturing the effects of genetic variants on the transcriptome, yet challenges with sequencing coverage, cost, tissue selection, and scalability have limited its clinical adoption. To address this, we developed STRIPE (Sequencing Targeted RNAs Identifies Pathogenic Events), a targeted long-read RNA-seq–based strategy for rare disease diagnosis and variant interpretation. STRIPE enables deep sequencing of full-length transcripts for any customized disease-specific gene panel such that a wide range of clinically informative readouts, including transcript aberrations and sequence variants, can be detected at haplotype-level resolution. Applying STRIPE to 88 individuals spanning two major rare disease groups, we accurately reidentified known pathogenic variants and revealed their transcript consequences, including many unexpected ones. For 8 of 15 splice site region variants, we observed more complex RNA processing defects beyond single exon skipping or cryptic splice site activation. Notably, we find that donor splice site variants frequently activate cryptic intronic polyadenylation sites, leading to premature transcript termination. Leveraging unique strengths of long-read RNA-seq, STRIPE also resolved variants of uncertain significance and uncovered disease-causing variants in five previously undiagnosed individuals. Overall, STRIPE is a powerful, adaptable, and scalable strategy with broad potential to improve clinical variant interpretation and advance genetic diagnosis of rare diseases.

Abstract In both type 1 and type 2 diabetes (T1D and T2D), insulin-producing β cells undergo progressive dysfunction due to inflammation, leading to impaired glucose responsiveness, dedifferentiation, and cell loss. While bile acid (BA) dysregulation under diabetic conditions is known to influence metabolic and inflammatory pathways, its mechanistic role in β cell regulation remains incompletely defined 1–3 . Here we show that bile acid sensor Farnesoid X receptor (FXR) and Bromodomain and Extra-Terminal motif (BET) signaling cooperatively regulates β cell inflammatory response and β cell identity. We identified the physiological protein-protein interaction between FXR and the bromodomain-containing protein 4 (BRD4) as a regulatory axis that protects against β cell dysfunction. We show that FXR activation by Fexaramine (Fex) together with BRD4 inhibition by JQ1 synergistically suppressed IL-1β-induced inflammation while also improving β cell identity and insulin secretion in both db/db model and high-fat diet (HFD) plus multi low-dose streptozotocin (MLD-STZ) model of diabetes. Importantly, this cooperative effect is abolished in β cell–specific FXR knockout (βFXRKO) mice, establishing that FXR is required for the functional synergy between these pathways in vivo . Mechanistically, structure-guided modeling and mutational analyses identified a direct interaction between FXR and the BD2 domain of BRD4, depending on specific lysine acetylation sites. Additionally, inhibition of the BD2 domain of BET combined with FXR activation markedly improved β cell survival in human T1D and T2D models established from human pluripotent stem cell (hPSC)-derived islet-like organoids (HILOs). Collectively, these findings establish a BA–bromodomain axis as a transcriptional interface linking metabolic signaling and chromatin regulation, and highlight FXR–BET targeting as a promising strategy to counter progressive β cell failure in diabetes.

Abstract Intercellular communication between multiple myeloma (MM) cells and osteoclast precursor cells (pre-OCs) contributes extensively to the occurrence and development of myeloma-related bone destruction. However, key interacting modes and the exchanged substances involved in this communication remain unclear. In this study, we discover that tunneling nanotubes (TNTs) directly connect MM and pre-OCs. Using the stable isotope labeling with amino acids in cell culture assay and a positive-negative double selection strategy, we identify filamin A (FLNA) as a major protein transported from MM to pre-OCs. FLNA acts as a molecular clutch linking extracellular matrix–bound MAC1 to the cytoskeleton, activating Rho and MAPK signaling pathways and promoting F-actin polymerization, which subsequently enhances osteoclast differentiation by modulating cellular stiffness, traction force, and deformability. In addition, FLNA directly binds vinculin and promotes its recruitment to podosomes, thereby enhancing the functions of podosomes and the bone resorption capacity of osteoclasts. The conditional depletion of Flna in mice suppresses podosome activity and reduces stiffness, traction force, and deformability in pre-OCs, leading to significantly impaired osteoclast differentiation and increased bone mass. In the Vk∗MYC mouse model of myeloma, the administration of the TNT inhibitor latrunculin B disrupts FLNA transport to osteoclasts and alleviates osteolytic bone disease. These findings highlight the critical role of MM-transferred FLNA in osteoclastogenesis and suggest that targeting TNTs may represent a therapeutic strategy to limit pathological bone resorption associated with MM.

Abstract Purpose: In RET-rearranged non–small cell lung cancer (NSCLC), treatment options after first-generation selective RET inhibitors (SRI) are limited, and the value of SRI retreatment remains unclear. This study evaluates outcomes of SRI retreatment. Experimental Design: This multicenter retrospective study included patients with advanced RET-rearranged NSCLC who received ≥2 SRI-based therapy lines. Of 411 SRI-treated patients, 41 (10%) underwent SRI retreatment. Outcomes included objective response rate (ORR), progression-free survival (PFS), 6-month PFS, time to treatment failure, adverse events (AE). Results: Among 41 patients, 14 (34%) discontinued the initial SRI because of toxicity and 27 (66%) because of progression. After discontinuation for toxicity, all switched to an alternate SRI, achieving an ORR of 67%, median PFS of 9.9 months, and 6-month PFS of 80.3%. AEs reoccurred in nine patients (64%), with grade ≥3 AEs in 3 (21%) who switched at full dose. Among patients who discontinued because of progression, SRI monotherapy (n = 13) achieved an ORR of 23%, median PFS of 7 months, and 6-month PFS of 61.5%. Benefit was observed in patients with brain-only or oligoprogressive disease or dose reduction on first SRI. Combination therapy (n = 14; targeted agents, n = 11; chemotherapy, n = 3) achieved an ORR of 39%, median PFS of 4 months, and 6-month PFS of 27.3%. Conclusions: Same-class SRI switch after toxicity is feasible and clinically active but warrants cautious, dose-adjusted switching to mitigate toxicities. The efficacy of SRI rechallenge after progression seems limited overall. However, selected patients, such as those with brain-only or oligoprogressive disease, may derive benefit.

Abstract High-grade serous ovarian cancer (HGSOC), the most prevalent and lethal ovarian cancer subtype, is typically diagnosed at an advanced stage due to the absence of effective early detection strategies. Mounting evidence indicates that HGSOC originates from the fallopian tube (FT), particularly from secretory epithelial cells in the fimbria. However, incomplete understanding of the cellular and molecular origins of HGSOC continues to hinder early interception efforts. To address this, we established a clinically and genetically annotated living FT organoid biobank at Mayo Clinic, representing over 200 patient donors. Using optimized enrichment protocols, we isolated FT epithelial stem/progenitor cells from tissue and Tao brushings, generating organoids independent of anatomical site or laterality that recapitulate native FT epithelial architecture and lineage differentiation. We applied integrated multi-omic analyses, including single-cell RNA sequencing, single-nucleus RNA/ATAC sequencing, and bulk RNA-seq, across fresh cells and organoids. Integration with public datasets produced the largest single-cell atlas of normal and high-risk FTs to date. This analysis defined high-specificity gene signatures for secretory and multiciliated lineages that outperform canonical lineage markers (e.g. PAX8, FOXJ1), identified lineage-defining transcription factors and regulatory networks, and showed concordant protein expression in the human FT epithelium. Notably, subcluster-level analysis revealed a rare hybrid epithelial-mesenchymal population transcriptionally and proteomically aligned with the mesenchymal-like subtype of HGSOC. Spatial immunoprofiling demonstrated that these cells express multiple defining markers and reside above the basement membrane, intercalated between epithelial cells and lack luminal contact, reminiscent of the histologically described peg cell population. Comparative analyses with fetal mesonephric tissues uncovered developmental parallels, and this population was stably maintained in 3D organoid cultures, providing a robust ex-vivo model to study early transformation. Together, our findings identify a rare, multipotent epithelial population with mesenchymal features as a candidate origin for mesenchymal-like HGSOC and establish the FT organoid biobank as a powerful platform for mechanistic and translational studies in ovarian carcinogenesis. Citation Format: Megan L. Ritting, Wenmei Yang, Syed Mohammed Musheer Aalam, Hui Zhao, Liang Feng, Jianning Song, Mihai G. Dumbrava, Wazim M. Ismail, Kenneth Schaufelberger, S. John Weroha, Scott H. Kaufmann, Alexandre Gaspar-Maia, Mark E. Sherman, Jamie N. Bakkum-Gamez, Nagarajan Kannan. A rare multipotent peg-like epithelial cell is a candidate cell-of-origin for high-grade serous ovarian cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7636.