Citations

ABSTRACT The paraneurium is a distinct connective tissue sheath that loosely envelopes the epineurium. The space between the para‐ and epineurium (the subparaneurial compartment) holds significant clinical relevance, serving as a target for deposition of regional anesthesia and a potential pathway for the pathological spread of cysts and tumors. However, visualization of the paraneurial layer, as a separate structure from the epineurium, remains challenging with conventional MRI, especially in the absence of pathology. This study evaluated the visualization of the paraneurium using ultra‐high‐field 7‐Tesla (7 T) MRI compared to conventional field strengths. We assessed normal nerve anatomy and a spectrum of pathological cases, including intraneural ganglion cysts, schwannomas, intraneural perineuriomas, and metastatic melanoma. The results demonstrated that 7 T MRI provides superior resolution and delineation of the paraneurium compared to 1.5 and 3 T systems. In normal anatomy, T1‐weighted and proton density (PD) sequences provided optimal contrast between the paraneurium and surrounding adipose tissue. In contrast, for pathological cases, T2‐weighted sequences with fat suppression were superior for depicting the paraneurium and identifying subparaneurial involvement. In conclusion, 7 T MRI offers a significant advantage in identifying the thin paraneurial layer. This capability provides valuable insights into the pathogenesis of tumor spread and holds promise for improving pre‐operative planning and clinical decision‐making in peripheral nerve pathology.

As plastic surgery has grown, leaders within the field have advocated transforming plastic surgery divisions into independent departments. This shift raises critical questions regarding departmental identity, resource allocation, and potential administrative conflicts. The purpose of this study was to explore the potential implications of these institutional hierarchies from the perspective of plastic surgery program chairs. An anonymous online survey was distributed to chairs of accredited plastic surgery programs. Thirty-five of 103 (34.0%) individuals responded to the online questionnaire. Twenty-eight of 33 (84.8%) programs were classified as divisions of surgery, whereas 5 of 33 (15.2%) were independent departments. Nine of 28 (32.1%) respondents believed their division had lost patients to other surgical specialties because their program was not an independent department. Eighteen of 33 (54.5%) respondents believed that plastic surgery is negatively impacted when institutional programs are not independent departments. Respondents who indicated losing patients to other specialties, compared with those who did not, were 22.4 times more likely to be critical of institutional hierarchies ( P =0.004). Respondents from independent departments were 12.6 times more likely to also express a negative view ( P =0.049). Our findings provide a nuanced understanding of the institutional structures within plastic surgery programs. The extent to which institutional hierarchies negatively impact plastic surgery remains unclear. The significant association between patient loss and negative perceptions implies that direct clinical or financial repercussions strongly shape opinions. Moreover, chairs from independent departments may be more critical of institutional hierarchies because they have a deeper appreciation for the advantages that come with departmental status.

Abstract The adoption of electronic health records (EHRs) has transformed health care, improving efficiency and chart accessibility. However, the widespread reliance on unstructured data entry and the lack of standardized documentation frameworks have resulted in significant data fragmentation across health care systems. The prevalence of unstructured data in EHRs limits their potential for clinical decision support, trial matching, real‐world evidence (RWE) generation, and quality measurement. Data fragmentation in health care triggers a cascade of challenges that ultimately compromise patient care. Clinicians face an excessive documentation burden and struggle to locate critical information buried in unstructured notes. Researchers encounter difficulties in extracting reliable clinical data. EHR vendors grapple with standardizing unstructured information for interoperability, and payers are unable to process unstructured clinical data efficiently to support value‐based care models. These challenges are particularly acute in oncology, where complex clinical elements like cancer staging, disease status, and treatment changes require precise, structured documentation. Emerging artificial intelligence (AI) technologies, such as large language models (LLMs) and ambient listening, offer a path to automate structured data generation while reducing the workload on providers. Here, the authors propose LLM‐based workflows that balance automation with clinician verification, streamlining data entry without compromising accuracy. Realizing these benefits requires coordinated efforts among clinicians, researchers, EHR vendors, payers, and policymakers to align regulatory frameworks with AI‐driven innovations. This article outlines a strategy to enhance structured data capture within EHRs, ultimately improving patient care, research, and health care efficiency.

Background/Objectives: We performed a systematic review of the current clinical and preclinical literature on the use of particle therapy, DDRi, and/or immunotherapy, specifically for pMMR colorectal cancer. Methods: A systematic review of the literature published between 2014 and 2025 was conducted across major databases. Studies were included if they examined particle radiotherapy (e.g., proton, alpha, carbon) either alone or in combination with DDRi and/or immune checkpoint inhibitors (ICI), or X-ray radiotherapy (XRT) in combination with DDRi/ICI. Results: In total, 133 studies met the inclusion criteria, including 62 clinical studies. Clinically, particle therapies show excellent local control and normal tissue sparing, with manageable toxicity profiles. Trials with any form of radiation and DDRi are few, potentially owing to toxicity concerns. ICI combinations showed promising efficacy with XRT, with no randomized trials comparing them to particle radiation. Conclusions: Particle radiation and/or DDRi have significant preclinical evidence of immunostimulatory effects in pMMR rectal adenocarcinoma and increase response rates to immunotherapy (presented in the companion manuscript). Despite strong preclinical evidence and rationale, clinical trials including all three modalities are scarce. Existing evidence suggests a potential benefit based on extrapolation from photon-based studies and supports prospective evaluation with careful attention to treatment-related toxicity, which remains a concern.

Abstract Objectives: In the RadComp randomized trial, we compared radiation dose to cardiac substructures in patients treated with proton versus photon radiation therapy (RT) ; evaluated changes in cardiac-specific health-related quality of life (HRQOL) outcomes from baseline to 6 months following RT; and determined associations between radiation dose volume histogram (DVH) parameters and changes in cardiac-specific HRQOL. Methods: The study cohort included patients with locally advanced breast cancer enrolled on RadComp who completed treatment per protocol, had available radiation DVH metrics for all cardiac substructures, and completed both the baseline and at least one follow-up cardiac-specific HRQOL over the 6 months post-RT. Cardiac-specific HRQOL measures included PROMIS fatigue, PRO-CTCAE shortness of breath (SOB) (severity and interference with daily activities), and PRO-CTCAE chest pain, chest tightness, and angina (frequency, severity, and interference). Detailed dose volume histograms (DVH) for cardiac substructures included the whole heart, left ventricle (LV), right ventricle (RV), left atrium (LA), right atrium (RA), and left anterior descending artery (LAD) were derived. The Wilcoxon rank-sum test was used to compare dose distributions between photon and proton RT. Associations between DVH metrics and HRQOL outcomes were evaluated using repeated-measures linear regression via generalized estimating equations with an independent working correlation structure. All models were adjusted for RT type, age, cardiovascular risk factors, surgery type, laterality, ethnicity, education, timepoint, and race. Additional stratified analyses was performed according to laterality (right versus left + bilateral). Results: Across 1160 RadComp participants, RT exposure to cardiac substructures differed significantly between proton and photon RT and by laterality. Median mean heart dose was lower with protons, 0.9 Gy (Q1, Q3 0.49, 1.46), compared to photons 2.83 Gy (Q1, Q3 1.49, 4.52) (p< 0.001), and lower for right-sided disease, 0.91 Gy (Q1, Q3 0.44, 1.55), compared to left and bilateral 1.99 Gy (Q1, Q3 1.11, 3.63) (p<0.001). Similarly, for the LAD, LA, LV, and RA, mean proton dose was significantly lower (p<0.001). However, for the whole heart, RA, and RV, maximum doses to small-volume parameters and dose to 0.03cc were higher with proton RT. PRO-CTCAE measures of HRQOL were generally favorable at all timepoints across all participants, and grade ≥3 toxicity was rare (≤5.5%) for all measures. The PROMIS Fatigue T-Score increased from baseline (median 49 [Q1, Q3 46, 57]) to end of RT (55 [Q1, Q3 49, 63]), and returned to baseline levels at 6 months (p<0.001). While there were no significant differences in PRO-CTCAE SOB severity (p=0.24) and interference (p=0.11) over time, PRO-CTCAE chest pain worsened compared to pre-RT with respect to frequency (p<0.001), severity (p<0.001), and interference (p=0.008). However, the absolute changes were small. No DVH metrics were significantly associated with any cardiac-specific HRQOL outcomes. Stratification by laterality yielded similar results. Conclusions: Proton RT generally conferred significantly lower doses to the heart and cardiac substructures compared to photon RT. HRQOL outcomes were favorable across all participants, with fatigue returning to baseline, no change in SOB parameters, and a slight worsening of chest pain outcomes over 6 months following RT. Cardiac-specific HRQOLs appear not to be significantly affected by RT dose to cardiac substructures within the first 6 months post-RT. Other organs at risk that could also impact cardiac HRQOL endpoints, longer follow-up, and major cardiac events will be evaluated in future analyses. Citation Format: J. L. Wright, K. Ko, W. Kenworthy, W. Bosch, L. Braunstein, G. Freedman, E. Gillespie, R. Jimenez, S. Pugh, H. Lu, M. Pankuch, N. Remmes, W. Straube, W. Zou, E. Berlin, L. Fang, S. MacDonald, N. Taunk, J. Kim, A. Thukral, A. Taghian, M. Mishra, R. Mutter, V. Croog, D. Boggs, J. Urbanic, N. Ohri, A. Chawla, T. Meier, L. Rosen, M. Fegundes, O. Cahlon, K. Hoffman, A. Ferriozzi, J. Bekelman, A. Lin, B. Ky. Associations Between Cardiac Radiation Dose Volume Metrics and Cardiac Specific Health Related Quality of Life for Patients with Breast Cancer Randomized to Proton versus Photon Radiation on the RadComp Trial [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr RF6-04.

Abstract Benign breast disease (BBD) confers elevated risk for breast cancer, yet current tools lack precision in identifying lesions likely to progress. Immune mechanisms may influence progression, but the spatial and molecular features of immune responses in BBD versus invasive cancer remain poorly characterized. We applied Digital Spatial Proteomics (DSP) to evaluate immune-related protein expression across 1,540 regions of interest (ROIs) from breast tumors (n=58, Carolina Breast Cancer Study [CBCS]), a UNC BBD cohort (n=20), and a Mayo BBD cohort (n=177). Using Consensus Clustering in CBCS and UNC BBD, we identified two stable immune phenotypes: “High Immune,” enriched in tumors and high-risk benign lesions, and “Quiet Immune,” predominant in low-risk BBD. ROIs were classified by immune marker expression, and patients were labeled as “High” if ≥50% of their ROIs were High Immune. In CBCS, High Immune profiles were associated with ER-negative status (OR=2.92 [0.99-8.65] participant-level; OR=3.28 [1.34-8.04] ROI-level). In BBD, High Immune profiles trended toward higher odds of high-risk lesions (OR=1.00 [0.15-6.70] participant-level; OR=1.68 [0.57-4.97] ROI-level), though this pattern was not replicated in the Mayo cohort. BBD samples exhibited substantial spatial heterogeneity in immune expression (40% within-patient ROI agreement vs. 74% in tumors), potentially limiting biomarker reliability. Nonetheless, four immune proteins (CD44, CD80, CD14, CD66b) were significantly associated with future cancer development in the Mayo cohort, though these findings require validation due to within-person variability. Our findings highlight the challenges posed by spatial heterogeneity in assessing immune biomarkers in BBD and suggest that immune activation may emerge later in disease progression. These results raise important questions about the timing and consistency of immune responses in the natural history of breast cancer. Citation Format: A. Zhang, A. Cozzo, L. Olsson, E. Kirk, X. Gao, S. Nyante, B. Calhoun, R. Vierkant, M. Sherman, M. Troester. Immune Heterogeneity in Breast Cancer Natural History: Benign Breast Disease to Breast Cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS2-05-23.

Abstract Background: Hispanic/Latinx (HL) and African American (AA) breast cancer (BC) patients (pts) have increased prevalence of high-risk features such as incidence of BC at age <50 and estrogen receptor negativity. We previously showed that AA and HL had a lower fraction of neutrophils and M2 macrophages (Mɸ) despite higher percentage of PD-L1 positive cases. This study aims to further characterize differences in the tumor immune microenvironment between racial and ethnic groups, with a focus on how these differences vary by site of biopsy. Methods: We analyzed 19,459 BC samples tested by NGS (592, NextSeq; WES, NovaSeq) and WTS (NovaSeq; Caris Life Sciences). Race/ethnicity data were self-reported by pts. Immune cell proportions were estimated using WTS deconvolution (Quantiseq). Gene expression was analyzed for T-cell inflammed score (TIS), MAPK Activation Score (MPAS) and interferon-gamma (IFNy) score. Real-world overall survival (OS) data were derived from insurance claims, calculated from tissue collection or treatment initiation to last contact, and analyzed using Kaplan-Meier. Statistical significance was assessed by chi-square and Mann-Whitney U test with p-values adjusted for multiple comparisons (q<0.05). Results: There were 9674 Non-Hispanic Whites (NHW) (primary BC, defined as biopsy (bx) from breast) pBC: n=3,537, 36.6%; (metastatic BC, defined as biopsy outside of the breast) mBC: n=6,137 63.4%, 1800 AA (pBC: n=750, 42%, mBC: n=1,050, 58%), and 1795 HL (pBC: n=842, 47%; mBC n=953, 53%) samples. There were 2364 TNBC (66% NHW, 19% NHB, 15% HL), 4318 HR+/HER2- (76% NHW, 11.6% NHB, 12.4% HL), and 893 HER2+ (68% NHW, 15% NHB, 17% HL). In pBC overall, AA had lower median cell fraction (%) of B cells (4.8% vs 5.3%), M2 Mɸ (3.4% vs 4.4%), but higher median cell fraction of dendritic cells (DC) (2.8% vs 2.6%) compared to NHW while HL had lower median cell fraction of M2 Mɸ (3.9% vs 4.4%) and higher median cell fraction of CD8+ T cell (0.4% vs 0.2%), all q<0.05. In pBC, HL had higher IFNy score (-0.3 vs -0.34), while AA had lower MPAS (-1.2 vs -0.96), all q<0.05. In lymph node bx, AA had lower median % of M2 Mɸ (3.4% vs 4.4%, q<0.05). No significant differences were seen in the liver bx by race but notably there were lower median cell fractions of CD4+, CD8+, and regulatory T cells (0.0 across NHW, AA, HL) across all race/ethnicities vs. other biopsy sites. By subtype, TNBC primary bx yielded much of the significant changes with AA having lower median cell fraction of M2 Mɸ (2.49% vs 3.04%, q= 0.01) compared to NHW. In primary bx, AA (35.6 m vs 48.5 m, HR 1.4, 95% CI 1.2-1.5, p<0.001) and HL (42.4 m vs 48.5 m, HR 1.2, 95% CI 1.1-1.4, p<0.001) had worse median OS versus NHW. In lymph node bx, AA also had worse OS (29.2 m vs 37.2 m, HR 1.2, 95% CI 1-1.5, p=0.03) compared to NHW. In liver bx, there was worse mOS overall but no difference by race (NHW 18.65m, AA 18.79m, HL 20.6m, p =0.73). In HR+/HER2- liver bx AA had worse OS (14.2 m vs 20.9 m, HR 1.3, 95% CI 1-1.7, p=0.03) versus NHW. In HER2+ primary bx, HL had worse OS (43.2m vs 66.9m, HR 1.7, 95% CI 1.1-2.6, p=0.009). There was a trend towards worse mOS in TNBC AA patients receiving pembrolizumab (18.42m vs HL 23.69m vs NHW 35.30m; p=0.16) and HER2+ AA patients receiving trastuzumab deruxtecan (AA: 16.06m vs HL: 26.81m vs NHW: not reached; p = 0.05). Conclusions: We found race and biopsy site specific differences in the breast cancer immune microenvironment. AA pts had lower M2 Mɸ in primary and lymph node bx while NHW pts showed higher CD8+ T cell in primary tumors. Liver biopsies exhibited low T cell infiltration across all groups. AA pts had worse mOS in primary and lymph node bx compared to NHW but not in liver bx of which we saw worse mOS in all races. These results highlight the need to consider both race and bx site in treatment decisions. Citation Format: R. Hsu, S. Deshmukh, N. Gyabaah-Kessie, E. Rallos, B. Al-Zubeidy, E. Gonzalez, A. G. Baugh, S. Carrel, M. Li, D. Zavala, A. Martynova, P. Jayachandran, D. Stewart, D. Spicer, S. Wu, J. Xiu, G. W. Sledge Jr., R. L. Mahtani, A. C. Sandoval Leon, S. Gandhi, S. Chumsri, M. Lustberg, M. C. Stern, E. T. Roussos Torres. Immune microenvironment and survival differences among Hispanic and African American breast cancer cases by biopsy site [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-12-06.

Abstract Background: Artificial intelligence (AI) algorithms based on deep learning approaches show promise in improving breast cancer (BC) detection on mammography and may also improve prediction of future BC risk compared with clinical risk prediction models. Historical clinical risk prediction models underperform among women with BBD; however, AI models for BC detection on mammography have not been tested among women at elevated BC risk due to benign breast disease (BBD). We evaluated if an AI cancer detection algorithm can aid BC risk prediction among women with BBD. Methods: We examined women without a prior BC who had an initial BBD biopsy at Mayo Clinic, Rochester, between 2002 and 2013. Incident BC occurring after BBD was identified using the Mayo Tumor Registry and supplemented by follow-up questionnaires. Only women with a full field digital mammogram at least 2 years prior to BC diagnosis were eligible. A breast pathologist evaluated BBD according to increasing severity, as non-proliferative (NP), proliferative disease without atypia (PDWA) or atypical hyperplasia (AH). AI malignancy scores derived from the Transpara detection algorithm (1-10) and volumetric percent density (VPD) from Volpara (per one standard deviation, SD) were assessed from mammograms close to BBD diagnosis. Cox proportional hazards regression models were applied to estimate hazard ratios (HRs) with 95% Confidence Intervals (CIs), adjusted for age and BMI. C-statistics were estimated to assess the contribution of each factor to BC risk prediction. Likelihood ratio tests (LRTs) and bootstrapping methods were used to compare model performance with the addition of AI malignancy score. Results: The BBD cohort with mammography consisted of 3,125 women followed for a median 12.9 years, with 250 incident BC. Of these, 221 BC occurred at least two years after the mammogram and were used in analyses. As expected, increased BC risk was associated with BBD severity [for AH HR=3.18 (95%CI: 2.04, 4.95) and for PDWA HR=1.59 (95%CI: 1.17, 2.14) compared to NP] and with higher VPD [HR=1.24 per SD (95% CI: 1.05, 1.45)] (Table ). The AI-malignancy score alone was also associated with BC risk [HR=1.16 per 1 unit score (95%CI: 1.09,1.23)] and showed similar discriminatory accuracy [C-statistic=0.626 (95%CI: 0.586, 0.665)] as the model with BBD severity and VPD combined [C-statistic=0.627, (95%CI: 0.585, 0.669)] (Table ). In models with BBD severity and VPD, the AI-malignancy score was an independent risk factor for BC [HR=1.14 (95%CI: 1.07, 1.21); PLRT<0.001] but only achieved a marginally significant improvement in discriminatory accuracy [C-statistic=0.651 (0.609, 0.693), ΔC-statistic=0.024 (95% CI: 0.000, 0.047)]. Conclusion: In this preliminary study, an AI BC detection algorithm achieved similar performance in predicting future BC risk compared to established risk factors. Citation Format: C. M. Vachon, S. Winham, M. Jensen, D. Hursh, L. Pacheco-Spann, A. Norman, J. Fischer, S. Schrup, L. Seymour, D. Gehling, S. Nyante, M. Troester, N. Karssemeijer, R. Vierkant, D. Radisky, C. Scott, S. I. Maimone, A. Degnim, M. Sherman. An AI algorithm for breast cancer detection improves future BC risk prediction among women with benign breast disease [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PD4-02.