Conners, Amy Lynn

Abstract Background Preparing tumor board patient summaries is time intensive. Large-language-model based systems may automate summarization but require real-world evaluation prior to clinical use. We performed an exploratory retrospective evaluation of the Microsoft Healthcare Agent Orchestrator (HAO), deployed in a Mayo Clinic–controlled staged environment, to generate tumor board–style patient summaries from retrospective Electronic Health Record (EHR) notes. Methods HAO generated summaries for breast, hepatobiliary, and neuro-oncology tumor board cases using up to the most recent 1,000 clinical notes. Clinician reviewers evaluated outputs via REDCap surveys across perceived factuality, completeness, clarity/conciseness, temporal cohesion, comparative performance, safety, and clinical utility (0–4 Likert scale). Reviewers were permitted to query the HAO chat interface to address missing details. Automated factuality was assessed using TBFact (bidirectional entailment), reporting precision and recall against available reference summaries. Results Among 57 survey responses from 5 different physicians, mean scores exceeded 2.8 across domains, with medians of 3 for most axes. In an exploratory comparison, oncology fellows required less time to review HAO-generated summaries than to manually generate patient summaries (mean difference 13.57 minutes per patient, p<0.001), although this difference may be influenced by prior familiarity with the same cases; 96% of survey responses indicated that HAO would save time. TBFact evaluations showed higher recall than precision across domains, consistent with broad capture of reference content alongside additional content that was not present in gold-standard summaries. Attribution was viewed favorably but showed issues with primary-source specificity and link reliability. Conclusions In a controlled Mayo environment, HAO demonstrated moderate performance and was associated with reduced review time for tumor board preparation. These findings are promising but preliminary and do not establish clinical safety, noninferiority to manual review, or readiness for routine clinical use. Limitations, including verbosity, specialty-specific content gaps, and inconsistent attribution, highlight the need for iterative refinement and further evaluation. Graphical Abstract Created using ChatGPT

Background Breast arterial calcifications (BAC) are commonly observed as an incidental finding on screening mammography and have been linked to cardiovascular disease. Whether BAC is independently associated with stroke risk remains uncertain. Methods We conducted a systematic review and meta-analysis in accordance with PRISMA 2020 guidelines. PubMed, Embase, and Scopus were searched from inception to May 2024 for cohort studies evaluating the association between BAC and stroke. Eligible studies included women undergoing mammography with documented BAC status and subsequent stroke outcomes. Data on patient characteristics and vascular risk factors were extracted. Study quality was appraised using the Joanna Briggs Institute (JBI) Critical Appraisal Checklist. Random-effects meta-analyses with restricted maximum likelihood (REML) estimation were performed to pool risk ratios (RRs) and mean differences (MDs). Heterogeneity was assessed with the I 2 statistic, and publication bias with Egger’s test. Results Ten cohort studies including 52,413 women were analyzed. The presence of BAC was associated with more than a twofold increased risk of stroke (RR 2.09, 95% CI 1.58–2.75; I 2 = 64.4%). This association persisted after adjustment for age, diabetes, hyperlipidemia, and menopausal status. Compared with BAC-negative women, those with BAC were significantly older (MD 7.07 years, 95% CI 5.44–8.70) and more frequently hypertensive (RR 1.48, 95% CI 1.23–1.78), diabetic (RR 1.67, 95% CI 1.38–2.02), hyperlipidemic (RR 1.27, 95% CI 1.07–1.51), and postmenopausal (RR 1.26, 95% CI 1.00–1.59). Interestingly, BAC was less common among smokers (RR 0.62, 95% CI 0.45–0.86). Egger’s test showed no evidence of publication bias ( p = 0.143). Conclusion BAC detected on screening mammography is independently associated with an increased risk of stroke, even after accounting for traditional risk factors. These findings support BAC as a promising, underutilized imaging biomarker of cerebrovascular risk in women and highlight the need for standardized reporting and prospective validation.

Abstract Background Molecular breast imaging (MBI) has shown utility as a supplement to mammography in women with dense breasts. Background parenchymal uptake (BPU), which describes the level of Tc-99m sestamibi in fibroglandular tissue on MBI, varies among women with similar density. Higher BPU is associated with greater breast cancer risk. The goal of this work is to investigate the histological correlations of BPU that may explain this association with risk. Methods We compared histologic features of prospectively collected benign breast tissue biopsies from healthy women who had a region of dense tissue on mammography exhibiting one of the extreme categories of BPU on MBI: either photopenic or marked. Tissue composition of specimens was assessed by measuring area of epithelium, stroma, and fat and counting lobules. Lobular involution status was assessed by an expert pathologist as either none, partial, or complete. Ki-67 index and estrogen receptor-alpha expression were also assessed. Results Biopsies were performed in 48 women (mean age 54.9 years [SD 11.1 years]), including 20 with marked BPU and 28 with photopenic BPU. Women with marked BPU were younger (mean age 49.9 vs. 58.4 years, p = 0.004), had higher body mass index ( p = 0.01), and were more likely premenopausal ( p = 0.005). Marked BPU specimens had a higher proportion of epithelium (14.5% vs. 2.2%, p < 0.001), lower stromal content (44.8% vs. 84.2%, p = 0.005), similar fat (18.9% vs. 12.9%, p = 0.41), and greater lobule counts (15.5 vs. 6.3, p < 0.001) compared to photopenic specimens. Complete lobular involution was less frequent in marked tissue than in photopenic tissue (10.5% vs. 76%, p < 0.001). Ki-67 index was higher in marked BPU tissues (5.4% vs. 1.6%, p = 0.006), though the difference was attenuated after adjustment for epithelial area (p = 0.26). Estrogen receptor-alpha expression did not differ between marked and photopenic groups (25.6% vs. 28.4%, p = 0.12). Conclusions Marked BPU on MBI is associated with greater epithelial content, less lobular involution, and higher proliferative activity—features characteristic of tissue at elevated risk for malignant transformation. These findings suggest that BPU provides functional information beyond mammographic density, supporting its role as an imaging biomarker for breast cancer risk in women with dense breasts. Trial Registration: NCT01240278 and NCT01588834

In women with dense breasts, adding molecular breast imaging to digital breast tomosynthesis (DBT) increased both overall and invasive cancer detection rates, while modestly raising the recall rate compared with DBT alone.

Pseudoangiomatous stromal hyperplasia (PASH) of the breast is histologically characterized by anastomosing and slit-like spaces invested by collagenous stroma and lined by flattened, spindle cells. These clear spaces that may mimic microscopic vascular channels do not contain red blood cells. Immunohistochemistry (IHC) studies may also help to confirm a diagnosis of PASH, with the spindled cells marking positively with CD34 and PR while demonstrating no reactivity with more specific endothelial antigens such as CD31 and ERG. In the current case, a 39-year-old female was diagnosed with cellular PASH of the right breast with unique histological patterns showing “tiger-striped” and “zippered” histologies. To our knowledge, this is the first report of these unique variant PASH morphologies.

Abstract To facilitate the delivery of accurate and timely care to patients in complex environments, process improvement methodologies such as Lean can be very effective. Lean is a quality improvement methodology that seeks to add value for patients and employees by continuously improving processes and eliminating waste. At our institution, Lean principles were applied to improve efficiency and minimize waste in the diagnostic breast imaging reading room. This paper describes how we applied Lean principles, including plan-do-study-act testing, level-loading (heijunka), and visual management, to level the workload of the diagnostic radiologists in our practice. Implementation of these principles to improve the diagnostic workflow in breast imaging is described along with examples from our practice, including challenges and future opportunities.

Abstract Purpose: Patient-derived xenografts (PDX) are a research tool for studying cancer biology and drug response phenotypes. While engraftment rates are higher for tumors with more aggressive characteristics, it is uncertain whether engraftment is prognostic for cancer recurrence. Patients and Methods: In a prospective study of patients with breast cancer treated with neoadjuvant chemotherapy (NAC) with taxane ± trastuzumab followed by anthracycline-based chemotherapy, we report the association between breast cancer events and PDX engraftment using tumors derived from treatment naïve (pre-NAC biopsies from 113 patients) and treatment resistant (post-NAC at surgery from 34 patients). Gray test was used to assess whether the cumulative incidence of a breast cancer event differs with respect to either pre-NAC PDX engraftment or post-NAC PDX engraftment. Results: With a median follow-up of 5.7 years, the cumulative incidence of breast cancer relapse did not differ significantly according to pre-NAC PDX engraftment (5-year rate: 13.6% vs. 13.4%; P = 0.89). However, the incidence of a breast event was greater for patients with post-NAC PDX engraftment (5-year rate: 50.0% vs. 19.6%), but this did not achieve significance (P = 0.11). Conclusions: In treatment-naïve breast cancer receiving standard NAC, PDX engraftment was not prognostic for breast cancer recurrence. Further study is needed to establish whether PDX engraftment in the treatment-resistant setting is prognostic for cancer recurrence.