Citations

Background: Major League Baseball (MLB) recently instituted a combine for their prospective players. Purpose/Hypothesis: The purpose of this study was to report initial epidemiological findings for both physical examination and imaging findings for the MLB Combine. It was hypothesized that most players would have no significant abnormal physical examination findings, but most would have abnormal findings on their shoulder and elbow magnetic resonance imaging (MRI) scans. Study Design: Descriptive epidemiology study. Methods: All players who attended the 2023 MLB Combine were included. History was taken and physical examination was performed by an MLB team physician, and MRI of the shoulder, elbow, cervical spine, and lumbar spine was performed on all players and read by 1 of 2 musculoskeletal radiologists. Physical examination notes and imaging reports were reviewed by the authors and recorded. Data were analyzed separately for pitchers and position players. Results: Overall, 219 players (112 pitchers [51%], 107 position players [49%]) were included. The mean age was 20.4 ± 1.5 years. The majority of players (95%) had no current pain or limitations; 54 players (25%) had a history of previous surgery. For pitchers, 18% underwent previous elbow surgery. With regard to MRI, >80% of pitchers had some abnormality within their ulnar collateral ligament (UCL), the majority of which was UCL scarring/remodeling. 70% of pitchers and 60% of position players had evidence of an olecranon osteophyte. Increased signal within the ulnar nerve was common in pitchers (40%) but rare in position players (4%). For the shoulder, 95% of pitchers and 93% of position players had some rotator cuff abnormality (often tendinosis within the supraspinatus and infraspinatus muscles). Also, 75% of pitchers had MRI evidence of a superior labrum anterior to posterior (SLAP) tear, while 24% of position players demonstrated this. Lumbar spine MRI showed that 47% of position players had a disk protrusion and 46% had a previous pars fracture. Conclusion: Most baseball players who participated in the 2023 MLB Combine had MRI evidence of rotator cuff abnormalities, and many showed evidence of an olecranon osteophyte. Eighteen percent of pitchers had a previous history of elbow surgery. The majority of pitchers had some abnormality within their UCL and evidence of a SLAP tear on MRI scans.

Abstract Objectives We aimed to assess the expression of biomarkers of response to antibody-drug conjugates (TROP2 and nectin-4) and immune microenvironment (NKG7, PD-L1, and B7-H3) in penile squamous cell carcinoma (pSCC). Methods Our archive was queried for patients who had a penectomy for pSCC between 2000 and 2022. Primary tumors were immunostained for B7-H3 and NKG7, while metastatic specimens were immunostained for TROP2 and nectin-4. Expression of PD-L1, TROP2, and nectin-4 in primary tumors was previously characterized. H-scores (0-300) were used to quantify expression. Associations between biomarkers, tumor-infiltrating lymphocytes (TILs), and clinicopathologic and outcome parameters were evaluated. Results For both TROP2 and nectin-4, H-scores within the lymph node metastases were higher compared to those within the primary tumors (mean, 264.5 vs 244.8, P = .0003; mean, 170.6 vs 146.7, P = .05, respectively; 33 paired specimens). For B7-H3 (n = 107), 32.7% of the primary tumors had an H-score of more than 0. In 34.8% of the cases, NKG7 expression was observed in 25% to 50% of the TILs. A significant association was noted between TIL density, B7-H3, NKG7, and PD-L1 expression. Conclusions Therapeutic strategies targeting TROP2 and nectin-4 hold promise for patients with advanced pSCC. The potential of PD-L1, B7-H3, and NKG7 for predicting response to immunomodulatory treatment warrants further research.

Abstract Background: Endocrine therapy (ET) with aromatase inhibitors (AI) plus ovarian function suppression (OFS) is standard for premenopausal women with ER+/HER2- breast cancer. However, many premenopausal women are intolerant of AI + OFS, leaving tamoxifen (with or without OFS) as their only option. Extensive studies have demonstrated that premenopausal women treated with neoadjuvant ET whose tumors exhibit Ki-67 ≤ 10% following 4 weeks of therapy (defined as endocrine sensitive disease (ESD)) achieve 5-year distant disease-free survival > 96%. Problematically, only 41% of patients reach this ESD threshold with tamoxifen compared to 78% with AI+OFS (Nitz JCO 2022). (Z)-endoxifen (ENDX) is a potent selective estrogen receptor modulator (SERM) that inhibits ERα at plasma concentrations of 3-5 ng/ml and dually targets PKCβ1, a suspected oncogene in breast tumors, at concentrations ≥ 500 ng/ml. We therefore developed the EVANGELINE study (NCT05607004) is an open label, randomized neoadjuvant phase II study that includes a pharmacokinetic (PK) Run-in, aiming to evaluate whether ENDX +/- goserelin is non-inferior to exemestane plus goserelin with respect to ESD rate. Following the completed PK run-in portion of the study, where 40mg/day or 80mg/day ENDX were assessed, the 40 mg/day dose was selected for further evaluation in the second part of the trial. The second part is comprised of a randomized treatment cohort, for patients with a Ki-67% >10% at preregistration randomized to either 40mg (Z)-endoxifen + goserelin or exemestane + goserelin neoadjuvant treatment and a Single Arm Treatment Cohort for patients who have a Ki-67 ≤ 10% at pre-registration assigned to 40mg/day (Z)-endoxifen monotherapy neoadjuvant treatment. Methods: To investigate ENDX further in premenopausal women, a multicenter study in the United States, was designed(EVANGELINE; NCT05607004). This phase 2, open-label, randomized, neoadjuvant trial includes a PK run-in phase designed to assess whether the 4-week ESD rate with (Z)-endoxifen + goserelin is non-inferior to that of exemestane + goserelin in premenopausal women with Stage IIA or IIB ER+/HER2- breast cancer with diagnostic Ki67 > 10%. Additionally, a single arm Phase 2, open-label neoadjuvant Cohort was designed to assess the 4-week ESD rate with ENDX monotherapy in premenopausal women with Stage IIA or IIB ER+/HER2- BC with diagnostic Ki67 ≤10%. Subjects will undergo neoadjuvant treatment for six months before receiving surgical treatment in the involved breast. This study will additionally assess safety, tolerability, response rates, surgical outcomes, and monitor disease progression. Comprehensive biomarker evaluations will be conducted using paired tumor biopsies and blood samples collected at baseline and week four, as well as 6 month surgical specimens. Enrollment in the PK run-in phase (n-22) was completed in Fall 2024 with the second portion of the trial slated to begin early in 2025. Citation Format: Matthew P. Goetz, Vera J. Suman, Claudia Lopez, Hayley Erickson, Lonnissa Nguyen, Sandra S. Hammer, Lida Mina, Pooja Advani, Roberto Leon-Ferre, Karthik Giridhar, Felipe Batalini, Daniel Flora, Jason M. Jones, Nusayba A. Bagegni, Katie N. Hunt, Mara Piltin, Amy Degnim, James N. Ingle, Judy C. Boughey, Sarah A. Buhrow, Joel M. Reid, Matthew Schellenberg, John R. Hawse, Steven C. Quay. A randomized phase 2 non-inferiority trial of (Z)-endoxifen + goserelin and exemestane + goserelin as neoadjuvant treatment for premenopausal women with ER+/HER2- breast cancer (EVANGELINE) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr CT237.

INTRODUCTION: The risk of progression to esophageal adenocarcinoma in Barrett's esophagus (BE) increases with advancing degrees of dysplasia. There is a critical need to improve the diagnosis of BE dysplasia, given substantial interobserver variability and overcalls of dysplasia during manual community pathologist reads. We aimed to externally validate a previously cross-validated BE dysplasia diagnosis deep learning model (BEDDLM) that predicts dysplasia grade on whole slide images (WSIs). METHODS: We digitized nondysplastic BE (NDBE), low-grade (LGD), and high-grade dysplasia (HGD) histology slides from 3 external academic centers. A consensus read by 2 expert study pathologists was used as the criterion standard. Slide stain characteristics were normalized using cycle-generative adversarial networks. WSIs were assessed by BEDDLM using an ensemble approach, combining a “You Only Look Once” model followed by a ResNet101 classifier model. RESULTS: We included 489 WSIs. Consensus histopathology revealed 232 NDBE, 117 LGD, and 140 HGD WSIs. The mean age (SD) was 66.9 (11.4) years; 413 (84.7%) were men. Using the BEDDLM ensemble model, sensitivity and specificity for NDBE were 73.3% (95% confidence interval [CI]: 67.09%–78.85%) and 93.4% (95% CI: 89.62%–96.10%); for LGD, 84.6% (95% CI: 76.78%–90.62%) and 80.6% (95% CI: 76.26%–84.54%); and for HGD, 80.7% (95% CI: 73.19%–86.89%) and 94.8% (95% CI: 91.97%–96.91%), respectively. The F1 score was 0.81, 0 0.69, and 0.83 for NDBE, LGD, and HGD, respectively. DISCUSSION: Our externally validated deep learning model demonstrates substantial accuracy for the diagnosis of BE dysplasia grade on WSIs.

Abstract Background We evaluated the Tecan Fluent Mix and Pierce Workstation as an automated platform to aliquot directly from whole blood collection tubes. The Fluent Mix and Pierce Workstation uses the Tube Rotator™, an on-deck rotisserie, to keep whole blood in suspension, and 8 Piercing Tips™ to cross the pierceable caps of collection tubes to aliquot samples. Methods We performed accuracy and precision testing by aliquoting residual whole blood specimens into an on-deck gravimetric scale. To test carryover, we aliquoted alternating blanks and high calibrators for several analytes and performed mass spectrometry on the blanks. We performed 168 runs over a 21-day period to test the robustness of the platform. We aliquoted and analyzed 2 batches of samples on both our clinical platform and the Mix and Pierce Workstation to check concordance. Results We observed that the Fluent Mix and Pierce Workstation can aliquot 200 µL with 0.5% inaccuracy and 0.9%CV and can aliquot 300 µL with 0.1% inaccuracy and 0.7%CV. We were unable to detect any analyte carryover using mass spectrometry. We observed no hardware failures during the instrument stress test. Clinical comparison runs indicated that the performance of aliquots made on the Fluent Mix and Pierce Workstation was equivalent to that of aliquots made by our current clinical method. Conclusions Overall, our evaluation showed that the Tecan Fluent Mix and Pierce Workstation is a viable platform for aliquoting whole blood for downstream work flows and can eliminate labor-intensive sample decapping and recapping in high-throughput settings.

Background We automated subarachnoid hemorrhage volume (SAHV) calculation with artificial intelligence (SAHVAI) and created 3‐dimensional volumetric images (SAHVAI‐3D) using noncontrast head computed tomography data for patients with aneurysmal SAH. We also defined 4‐dimensional SAHVAI (SAHVAI‐4D), representing SAHV over time. We aimed to compare automated SAHVAI values and computational times to manual SAHV measurement methods, explore the potential of imaging biomarkers to identify at‐risk brain regions for delayed cerebral ischemia and explore potential insights in future neurotherapeutic interventions for patient recovery after SAH. Methods A training set of 10 consecutive patients with aneurysmal SAH was used to manually compute SAHV, SAHVAI‐3D, and SAHVAI‐4D, including 92 noncontrast computed tomography scans (182 slices each). The SAHVAI deep learning algorithm generated automated SAHV values in cubic centimeters. A 3‐dimensional SAH brain map was created for each patient for the SAHVAI and manual evaluations. Blood volumetric outputs were analyzed and compared to neurologic outcomes at discharge, including delayed cerebral ischemia events, symptomatic vasospasm, and areas with the thickest SAH blood concentration. Results SAHVAI quantified SAHV in a mean of 6.7 seconds per scan, significantly faster than the manual method, which took >60 minutes per scan (Fisher exact test, P <0.001). SAHVAI demonstrated an accuracy of 99.8%, Dice score of 0.701, false‐positive rate of 0.0005, and negative predictive value of 0.999. The mean absolute error between SAHVAI and manual methods was 5.67 mL. SAHVAI‐3D brain map and total SAHV at admission were inversely associated with Glasgow Coma Scale (R 2 = 0.23, P = 0.017) and directly associated with length of hospital stay (R 2 = 0.175, P = 0.004), especially in regions with dense blood concentration. Conclusion SAHVAI‐3D and SAHVAI‐4D brain mapping techniques represent innovative imaging biomarkers for SAH. These advancements enable rapid evaluation and targeted interventions, potentially improving patient care in SAH management.

Background and Aims: Among those with primary sclerosing cholangitis (PSC), perihilar cholangiocarcinoma (pCCA) is often diagnosed at a late stage and is a leading source of mortality. Detection of pCCA in PSC when curative action can be taken is challenging. Our aim was to create a deep learning model that analyzed MRI to detect early-stage pCCA and compare its diagnostic performance with expert radiologists. Approach and Results: We conducted a multicenter, international, retrospective cohort study involving adults with large duct PSC who underwent contrast-enhanced MRI. Senior abdominal radiologists reviewed the images. All patients with pCCA had early-stage cancer and were registered for liver transplantation. We trained a 3D DenseNet-121 model, a form of deep learning, using MRI images and assessed its performance in a separate test cohort. The study included 398 patients (training cohort n=150; test cohort n=248). pCCA was present in 230 individuals (training cohort n=64; test cohort n=166). In the test cohort, the respective performances of the model compared to the radiologists were: sensitivity 87.9% versus 50.0%, p <0.001; specificity 84.1% versus 100.0%, p <0.001; area under receiving operating curve 86.0% versus 75.0%, p <0.001. Even when a mass was absent, the model had a higher sensitivity for pCCA than radiologists (91.6% vs. 50.6%, p <0.001) and maintained good specificity (84.1%). Conclusions: The 3D DenseNet-121 MRI model effectively detects early-stage pCCA in PSC patients. Compared to expert radiologists, the model missed fewer cases of cancer.

Abstract Background Automated data extraction from echocardiography reports could facilitate large-scale registry creation and clinical surveillance of valvular heart diseases (VHD). We evaluated the performance of open-source large language models (LLMs) guided by prompt instructions and chain of thought (CoT) for this task. Methods From consecutive transthoracic echocardiographies performed in our center, we utilized 200 random reports from 2019 for prompt optimization and 1000 from 2023 for evaluation. Five instruction-tuned LLMs (Qwen2.0-72B, Llama3.0-70B, Mixtral8-46.7B, Llama3.0-8B, and Phi3.0-3.8B) were guided by prompt instructions with and without CoT to classify prosthetic valve presence and VHD severity. Performance was evaluated using classification metrics against expert-labeled ground truth. Mean squared error (MSE) was also calculated for predicted severity’s deviation from actual severity. Results With CoT prompting, Llama3.0-70B and Qwen2.0 achieved the highest performance (accuracy: 99.1% and 98.9% for VHD severity; 100% and 99.9% for prosthetic valve; MSE: 0.02 and 0.05, respectively). Smaller models showed lower accuracy for VHD severity (54.1%-85.9%) but maintained high accuracy for prosthetic valve detection (&gt;96%). Chain of thought reasoning yielded higher accuracy for larger models while increasing processing time from 2-25 to 67-154 seconds per report. Based on CoT reasonings, the wrong predictions were mainly due to model outputs being influenced by irrelevant information in the text or failure to follow the prompt instructions. Conclusions Our study demonstrates the near-perfect performance of open-source LLMs for automated echocardiography report interpretation with the purpose of registry formation and disease surveillance. While larger models achieved exceptional accuracy through prompt optimization, practical implementation requires balancing performance with computational efficiency.