Cystic neutrophilic granulomatous mastitis: sensitivity and specificity of 16s rRNA and Sanger sequencing for Corynebacterium spp.

AIMS: Cystic neutrophilic granulomatous mastitis (CNGM) is a subtype of granulomatous mastitis (GM) associated with Corynebacterium spp infection. We aimed to analyse the prevalence of Corynebacteria in CNGM and non-CNGM cases. METHODS: Breast specimens diagnosed as granulomatous inflammation between 2010 and 2020 were reviewed to identify a CNGM cohort and a non-CNGM cohort. Polymerase chain reaction-based identification of Corynebacteria by 16S ribosomal RNA (16S rRNA) primers, followed by confirmatory Sanger sequencing (SS), was performed on all cases. Clinical, radiological and microbiology data were retrieved from the electronic patient records. RESULTS: Twenty-eight CNGM cases and 19 non-CNGM cases were identified. Compared with the non-CNGM cohort, patients in the CNGM cohort were more likely to be multiparous (p=0.01), breast feeding (p=0.01) and presenting with a larger breast mass (p<0.01), spontaneous drainage (p=0.05) and skin irritation (p<0.01). No significant difference in the prevalence of Corynebacteria between the cohorts (7% vs 11%, p=0.68) by microbiological culture was identified. Compared with microbiology culture, the sensitivity and specificity of each Corynebacterial detection method were 50% and 81% for Gram stain, and 25% and 100% for 16S rRNA combined with SS. Regardless of the diagnosis, patients positive for Corynebacteria were more likely to have a persistent disease (p<0.01). CONCLUSION: CNGM presents as a large symptomatic breast mass in multiparous breastfeeding women. The importance of adequate sampling and repeated microbiology culture in conjunction with sequencing on all GM cases with persistent disease is paramount.

Predicting Patterns of Distant Metastasis in Breast Cancer Patients following Local Regional Therapy Using Machine Learning.

Up to 30% of breast cancer (BC) patients will develop distant metastases (DM), for which there is no cure. Here, statistical and machine learning (ML) models were developed to estimate the risk of site-specific DM following local-regional therapy. This retrospective study cohort included 175 patients diagnosed with invasive BC who later developed DM. Clinicopathological information was collected for analysis. Outcome variables were the first site of metastasis (brain, bone or visceral) and the time interval (months) to developing DM. Multivariate statistical analysis and ML-based multivariable gradient boosting machines identified factors associated with these outcomes. Machine learning models predicted the site of DM, demonstrating an area under the curve of 0.74, 0.75, and 0.73 for brain, bone and visceral sites, respectively. Overall, most patients (57%) developed bone metastases, with increased odds associated with estrogen receptor (ER) positivity. Human epidermal growth factor receptor-2 (HER2) positivity and non-anthracycline chemotherapy regimens were associated with a decreased risk of bone DM, while brain metastasis was associated with ER-negativity. Furthermore, non-anthracycline chemotherapy alone was a significant predictor of visceral metastasis. Here, clinicopathologic and treatment variables used in ML prediction models predict the first site of metastasis in BC. Further validation may guide focused patient-specific surveillance practices.

Machine learning analysis of breast ultrasound to classify triple negative and HER2+ breast cancer subtypes.

OBJECTIVES: Early diagnosis of triple-negative (TN) and human epidermal growth factor receptor 2 positive (HER2+) breast cancer is important due to its increased risk of micrometastatic spread necessitating early treatment and for guiding targeted therapies. This study aimed to evaluate the diagnostic performance of machine learning (ML) classification of newly diagnosed breast masses into TN versus non-TN (NTN) and HER2+ versus HER2 negative (HER2-) breast cancer, using radiomic features extracted from grayscale ultrasound (US) b-mode images. MATERIALS AND METHODS: A retrospective chart review identified 88 female patients who underwent diagnostic breast US imaging, had confirmation of invasive malignancy on pathology and receptor status determined on immunohistochemistry available. The patients were classified as TN, NTN, HER2+ or HER2- for ground-truth labelling. For image analysis, breast masses were manually segmented by a breast radiologist. Radiomic features were extracted per image and used for predictive modelling. Supervised ML classifiers included: logistic regression, k-nearest neighbour, and Naïve Bayes. Classification performance measures were calculated on an independent (unseen) test set. The area under the receiver operating characteristic curve (AUC), sensitivity (%), and specificity (%) were reported for each classifier. RESULTS: The logistic regression classifier demonstrated the highest AUC: 0.824 (sensitivity: 81.8%, specificity: 74.2%) for the TN sub-group and 0.778 (sensitivity: 71.4%, specificity: 71.6%) for the HER2 sub-group. CONCLUSION: ML classifiers demonstrate high diagnostic accuracy in classifying TN versus NTN and HER2+ versus HER2- breast cancers using US images. Identification of more aggressive breast cancer subtypes early in the diagnostic process could help achieve better prognoses by prioritizing clinical referral and prompting adequate early treatment.

Ki67 in Breast Cancer Assay: An Ad Hoc Testing Recommendation from the Canadian Association of Pathologists Task Force.

Ki67, a marker of cellular proliferation, is commonly assessed in surgical pathology laboratories. In breast cancer, Ki67 is an established prognostic factor with higher levels associated with worse long-term survival. However, Ki67 IHC is considered of limited clinical use in breast cancer management largely due to issues related to standardization and reproducibility of scoring across laboratories. Recently, both the American Food and Drug Administration (FDA) and Health Canada have approved the use of abemaciclib (CDK4/6 inhibitor) for patients with HR+/HER2: high-risk early breast cancers in the adjuvant setting. Health Canada and the FDA have included a Ki67 proliferation index of ≥20% in the drug monograph. The approval was based on the results from monarchE, a phase III clinical trial in early-stage chemotherapy-naïve, HR+, HER2 negative patients at high risk of early recurrence. The study has shown significant improvement in invasive disease-free survival (IDFS) with abemaciclib when combined with adjuvant endocrine therapy at two years. Therefore, there is an urgent need by the breast pathology and medical oncology community in Canada to establish national guideline recommendations for Ki67 testing as a predictive marker in the context of abemaciclib therapy consideration. The following recommendations are based on previous IKWG publications, available guidance from the monarchE trial and expert opinions. The current recommendations are by no means final or comprehensive, and their goal is to focus on its role in the selection of patients for abemaciclib therapy. The aim of this document is to guide Canadian pathologists on how to test and report Ki67 in invasive breast cancer. Testing should be performed upon a medical oncologist's request only. Testing must be performed on treatment-naïve tumor tissue. Testing on the core biopsy is preferred; however, a well-fixed resection specimen is an acceptable alternative. Adhering to ASCO/CAP fixation guidelines for breast biomarkers is advised. Readout training is strongly recommended. Visual counting methods, other than eyeballing, should be used, with global rather than hot spot assessment preferred. Counting 100 cells in at least four areas of the tumor is recommended. The Ki67 scoring app developed to assist pathologists with scoring Ki67 proposed by the IKWG, available for free download, may be used. Automated image analysis is very promising, and laboratories with such technology are encouraged to use it as an adjunct to visual counting. A score of <5 or >30 is more robust. The task force recommends that the results are best expressed as a continuous variable. The appropriate antibody clone and staining protocols to be used may take time to address. For the time being, the task force recommends having tonsils/+pancreas on-slide control and enrollment in at least one national/international EQA program. Analytical validation remains a pending goal. Until the data become available, using local ki67 protocols is acceptable. The task force recommends participation in upcoming calibration and technical validation initiatives.

Burnout Interventions for Resident Physicians: A Scoping Review of Their Content, Format, and Effectiveness.

CONTEXT.­: Physicians face a high rate of burnout, especially during the residency training period when trainees often experience a rapid increase in professional responsibilities and expectations. Effective burnout prevention programs for resident physicians are needed to address this significant issue. OBJECTIVE.­: To examine the content, format, and effectiveness of resident burnout interventions published in the last 10 years. DESIGN.­: The literature search was conducted on the MEDLINE database with the following keywords: internship, residency, health promotion, wellness, occupational stress, burnout, program evaluation, and program. Only studies published in English between 2010 and 2020 were included. Exclusion criteria were studies on interventions related to the COVID-19 pandemic, studies on duty hour restrictions, and studies without assessment of resident well-being postintervention. RESULTS.­: Thirty studies were included, with 2 randomized controlled trials, 3 case-control studies, 20 pretest and posttest studies, and 5 case reports. Of the 23 studies that used a validated well-being assessment tool, 10 reported improvements postintervention. These effective burnout interventions were longitudinal and included wellness training (7 of 10), physical activities (4 of 10), healthy dietary habits (2 of 10), social activities (1 of 10), formal mentorship programs (1 of 10), and health checkups (1 of 10). Combinations of burnout interventions, low numbers of program participants with high dropout rates, lack of a control group, and lack of standardized well-being assessment are the limitations identified. CONCLUSIONS.­: Longitudinal wellness training and other interventions appear effective in reducing resident burnout. However, the validity and generalizability of the results are limited by the study designs.

Resident Depression and Burnout During the COVID-19 Pandemic: A Survey of Canadian Laboratory Medicine Trainees.

CONTEXT.­: Resident physicians face a higher rate of burnout and depression than the general population. Few studies have examined burnout and depression in Canadian laboratory medicine residents, and none during the COVID-19 pandemic. OBJECTIVE.­: To identify the prevalence of burnout and depression, contributing factors, and the impact of COVID-19 in this population. DESIGN.­: An electronic survey was distributed to Canadian laboratory medicine residents. Burnout was assessed using the Oldenburg Burnout Inventory. Depression was assessed using the Patient Health Questionnaire 9. RESULTS.­: Seventy-nine responses were collected. The prevalence of burnout was 63% (50 of 79). The prevalence of depression was 47% (37 of 79). Modifiable factors significantly associated with burnout included career dissatisfaction, below average academic performance, lack of time off for illness, stress related to finances, lack of a peer or staff physician mentor, and a high level of fatigue. Modifiable factors significantly associated with depression further included a lack of access to wellness resources, lack of time off for leisure, and fewer hours of sleep. Fifty-five percent (41 of 74) of participants reported direct impacts to their personal circumstances by the COVID-19 pandemic. CONCLUSIONS.­: Burnout and depression are significant issues affecting Canadian laboratory medicine residents. As the COVID-19 pandemic continues, we recommend the institution of flexible work arrangements, protected time off for illness and leisure, ongoing evaluation of career satisfaction, formal and informal wellness programming with trainee input, formal mentorship programming, and a financial literacy curriculum as measures to improve trainee wellness.

Comparative Evaluation of Tumor-Infiltrating Lymphocytes in Companion Animals: Immuno-Oncology as a Relevant Translational Model for Cancer Therapy.

Despite the important role of preclinical experiments to characterize tumor biology and molecular pathways, there are ongoing challenges to model the tumor microenvironment, specifically the dynamic interactions between tumor cells and immune infiltrates. Comprehensive models of host-tumor immune interactions will enhance the development of emerging treatment strategies, such as immunotherapies. Although in vitro and murine models are important for the early modelling of cancer and treatment-response mechanisms, comparative research studies involving veterinary oncology may bridge the translational pathway to human studies. The natural progression of several malignancies in animals exhibits similar pathogenesis to human cancers, and previous studies have shown a relevant and evaluable immune system. Veterinary oncologists working alongside oncologists and cancer researchers have the potential to advance discovery. Understanding the host-tumor-immune interactions can accelerate drug and biomarker discovery in a clinically relevant setting. This review presents discoveries in comparative immuno-oncology and implications to cancer therapy.

Quantitative digital histopathology and machine learning to predict pathological complete response to chemotherapy in breast cancer patients using pre-treatment tumor biopsies.

Complete pathological response (pCR) to neoadjuvant chemotherapy (NAC) is a prognostic factor for breast cancer (BC) patients and is correlated with improved survival. However, pCR rates are variable to standard NAC, depending on BC subtype. This study investigates quantitative digital histopathology coupled with machine learning (ML) to predict NAC response a priori. Clinicopathologic data and digitized slides of BC core needle biopsies were collected from 149 patients treated with NAC. The nuclei within the tumor regions were segmented on the histology images of biopsy samples using a weighted U-Net model. Five pathomic feature subsets were extracted from segmented digitized samples, including the morphological, intensity-based, texture, graph-based and wavelet features. Seven ML experiments were conducted with different feature sets to develop a prediction model of therapy response using a gradient boosting machine with decision trees. The models were trained and optimized using a five-fold cross validation on the training data and evaluated using an unseen independent test set. The prediction model developed with the best clinical features (tumor size, tumor grade, age, and ER, PR, HER2 status) demonstrated an area under the ROC curve (AUC) of 0.73. Various pathomic feature subsets resulted in models with AUCs in the range of 0.67 and 0.87, with the best results associated with the graph-based and wavelet features. The selected features among all subsets of the pathomic and clinicopathologic features included four wavelet and three graph-based features and no clinical features. The predictive model developed with these features outperformed the other models, with an AUC of 0.90, a sensitivity of 85% and a specificity of 82% on the independent test set. The results demonstrated the potential of quantitative digital histopathology features integrated with ML methods in predicting BC response to NAC. This study is a step forward towards precision oncology for BC patients to potentially guide future therapies.

A Canadian national guideline on the neoadjuvant treatment of invasive breast cancer, including patient assessment, systemic therapy, and local management principles.

PURPOSE: The neoadjuvant treatment of breast cancer (NABC) is a rapidly changing area that benefits from guidelines integrating evidence with expert consensus to help direct practice. This can optimize patient outcomes by ensuring the appropriate use of evolving neoadjuvant principles. METHODS: An expert panel formulated evidence-based practice recommendations spanning the entire neoadjuvant breast cancer treatment journey. These were sent for practice-based consensus across Canada using the modified Delphi methodology, through a secure online survey. Final recommendations were graded using the GRADE criteria for guidelines. The evidence was reviewed over the course of guideline development to ensure recommendations remained aligned with current relevant data. RESULTS: Response rate to the online survey was almost 30%; representation was achieved from various medical specialties from both community and academic centres in various Canadian provinces. Two rounds of consensus were required to achieve 80% or higher consensus on 59 final statements. Five additional statements were added to reflect updated evidence but not sent for consensus. CONCLUSIONS: Key highlights of this comprehensive Canadian guideline on NABC include the use of neoadjuvant therapy for early stage triple negative and HER2 positive breast cancer, with subsequent adjuvant treatments for patients with residual disease. The use of molecular signatures, other targeted adjuvant therapies, and optimal response-based local regional management remain actively evolving areas. Many statements had evolving or limited data but still achieved high consensus, demonstrating the utility of such a guideline in helping to unify practice while further evidence evolves in this important area of breast cancer management.

Hot Seat Diagnosis.

CONTEXT.­: Competency-based medical education relies on frequent formative in-service assessments to ascertain trainee progression. Currently at our institution, trainees receive a summative end-of-rotation In-Training Evaluation Report based on feedback collected from staff pathologists. There is no method of simulating report sign-out. OBJECTIVE.­: To develop a formative in-service assessment tool that is able to simulate report sign-out and provide case-by-case feedback to trainees. Further, to compare time- versus competency-based assessment models. DESIGN.­: Twenty-one pathology trainees were assessed for 20 months. Hot Seat Diagnosis by trainees and trainee assessment by pathologists were recorded in the laboratory information system. In the first iteration, trainees were assessed by using a time-based assessment scale on their ability to diagnose, report, use ancillary tests, comment on clinical implications, and provide intraoperative consultation and/or gross cases. The second iteration used a competency-based assessment scale. Trainees and pathologists completed surveys on the effectiveness of the In-Training Evaluation Report versus the Hot Seat Diagnosis tool. RESULTS.­: Scores from both iterations correlated significantly with other assessment tools including the Resident In-Service Examination (r = 0.93, P = .04 and r = 0.87, P = .03). The competency-based model was better able to demonstrate improvement over time and stratify junior versus senior trainees than the time-based model. Trainees and pathologists rated Hot Seat Diagnosis as significantly more objective, detailed, and timely than the In-Training Evaluation Report, and effective at simulating report sign-out. CONCLUSIONS.­: Hot Seat Diagnosis is an effective tool for the formative in-service assessment of pathology trainees and simulation of report sign-out, with the competency-based model outperforming the time-based model.

Gynecologic pathology services in low- and middle-income countries.

OBJECTIVE: The International Gynecologic Cancer Society (IGCS) offers multidisciplinary conferences to underserved communities. Mentor pathologists have become an integral part of these tumor boards, as pathology services in low-to-middle-income countries are often inadequate and disjointed. The IGCS Pathology Working Group conducted a survey to assess barriers to quality pathology services in low-to-middle-income countries and identified potential solutions. METHODS: A 69-question cross-sectional survey assessing different aspects of pathology services was sent to 15 IGCS Extension for Community Healthcare Outcomes (ECHO) training sites in Africa, Asia, Central America, and the Caribbean. Local gynecologic oncologists distributed the survey to their pathology departments for review. The responses were tabulated in Microsoft Excel. RESULTS: Responses were received from nine training sites: five sites in Africa, two in Asia, one in Central America, and one in the Caribbean. There were no pathologists with subspecialty training in gynecologic pathology. Most (7/9, 78%) surveyed sites indicated that they have limited access to online education and knowledge transfer resources. Of the eight sites that responded to the questions, 50% had an electronic medical system and 75% had a cancer registry. Synoptic reporting was used in 75% of the sites and paper-based reporting was predominant (75%). Most (6/7, 86%) laboratories performed limited immunohistochemical stains on site. None of the sites had access to molecular testing. CONCLUSIONS: Initial goals for collaboration with local pathologists to improve diagnostic pathology in low- and middle-income countries could be defining minimal gross, microscopic, and reporting pathology requirements, as well as wisely designed educational programs intended to mentor local leaders in pathology. Larger studies are warranted to confirm these observations.

Assessment of Digital Pathology Imaging Biomarkers Associated with Breast Cancer Histologic Grade.

BACKGROUND: Evaluating histologic grade for breast cancer diagnosis is standard and associated with prognostic outcomes. Current challenges include the time required for manual microscopic evaluation and interobserver variability. This study proposes a computer-aided diagnostic (CAD) pipeline for grading tumors using artificial intelligence. METHODS: There were 138 patients included in this retrospective study. Breast core biopsy slides were prepared using standard laboratory techniques, digitized, and pre-processed for analysis. Deep convolutional neural networks (CNNs) were developed to identify the regions of interest containing malignant cells and to segment tumor nuclei. Imaging-based features associated with spatial parameters were extracted from the segmented regions of interest (ROIs). Clinical datasets and pathologic biomarkers (estrogen receptor, progesterone receptor, and human epidermal growth factor 2) were collected from all study subjects. Pathologic, clinical, and imaging-based features were input into machine learning (ML) models to classify histologic grade, and model performances were tested against ground-truth labels at the patient-level. Classification performances were evaluated using receiver-operating characteristic (ROC) analysis. RESULTS: Multiparametric feature sets, containing both clinical and imaging-based features, demonstrated high classification performance. Using imaging-derived markers alone, the classification performance demonstrated an area under the curve (AUC) of 0.745, while modeling these features with other pathologic biomarkers yielded an AUC of 0.836. CONCLUSION: These results demonstrate an association between tumor nuclear spatial features and tumor grade. If further validated, these systems may be implemented into pathology CADs and can assist pathologists to expeditiously grade tumors at the time of diagnosis and to help guide clinical decisions.

Myth or Reality: Self-Assessment Is Central to Effective Curriculum in Anatomical Pathology Graduate Medical Education.

Self-assessment, a personal evaluation of one's professional attributes and abilities against a perceived norm, has frequently been cited as a necessary component of self-directed learning and the maintenance of competency within regulated health professions, including the medical professions. However, education research literature has consistently shown uninformed personal global assessment of performance to be inaccurate in a variety of contexts, and have limited value in a workplace-based curriculum. Incorporating known standards of performance with internal and external data on the performance improves a learner's ability to accurately self-assess. Selecting content suitable for self-assessment, providing explicit assessment standards, encouraging feedback-seeking behaviors, supporting a growth mindset, and providing quality feedback in a supportive context are all strategies that can support learner self-assessment, learner engagement in reflection, and action on feedback in Anatomical Pathology graduate medical education.

A review and comparison of breast tumor cell nuclei segmentation performances using deep convolutional neural networks.

Breast cancer is currently the second most common cause of cancer-related death in women. Presently, the clinical benchmark in cancer diagnosis is tissue biopsy examination. However, the manual process of histopathological analysis is laborious, time-consuming, and limited by the quality of the specimen and the experience of the pathologist. This study's objective was to determine if deep convolutional neural networks can be trained, with transfer learning, on a set of histopathological images independent of breast tissue to segment tumor nuclei of the breast. Various deep convolutional neural networks were evaluated for the study, including U-Net, Mask R-CNN, and a novel network (GB U-Net). The networks were trained on a set of Hematoxylin and Eosin (H&E)-stained images of eight diverse types of tissues. GB U-Net demonstrated superior performance in segmenting sites of invasive diseases (AJI = 0.53, mAP = 0.39 & AJI = 0.54, mAP = 0.38), validated on two hold-out datasets exclusively containing breast tissue images of approximately 7,582 annotated cells. The results of the networks, trained on images independent of breast tissue, demonstrated that tumor nuclei of the breast could be accurately segmented.

Overcoming the limitations of patch-based learning to detect cancer in whole slide images.

Whole slide images (WSIs) pose unique challenges when training deep learning models. They are very large which makes it necessary to break each image down into smaller patches for analysis, image features have to be extracted at multiple scales in order to capture both detail and context, and extreme class imbalances may exist. Significant progress has been made in the analysis of these images, thanks largely due to the availability of public annotated datasets. We postulate, however, that even if a method scores well on a challenge task, this success may not translate to good performance in a more clinically relevant workflow. Many datasets consist of image patches which may suffer from data curation bias; other datasets are only labelled at the whole slide level and the lack of annotations across an image may mask erroneous local predictions so long as the final decision is correct. In this paper, we outline the differences between patch or slide-level classification versus methods that need to localize or segment cancer accurately across the whole slide, and we experimentally verify that best practices differ in both cases. We apply a binary cancer detection network on post neoadjuvant therapy breast cancer WSIs to find the tumor bed outlining the extent of cancer, a task which requires sensitivity and precision across the whole slide. We extensively study multiple design choices and their effects on the outcome, including architectures and augmentations. We propose a negative data sampling strategy, which drastically reduces the false positive rate (25% of false positives versus 62.5%) and improves each metric pertinent to our problem, with a 53% reduction in the error of tumor extent. Our results indicate classification performances of image patches versus WSIs are inversely related when the same negative data sampling strategy is used. Specifically, injection of negatives into training data for image patch classification degrades the performance, whereas the performance is improved for slide and pixel-level WSI classification tasks. Furthermore, we find applying extensive augmentations helps more in WSI-based tasks compared to patch-level image classification.

Analysis of tumor nuclear features using artificial intelligence to predict response to neoadjuvant chemotherapy in high-risk breast cancer patients.

PURPOSE: Neoadjuvant chemotherapy (NAC) is used to treat patients with high-risk breast cancer. The tumor response to NAC can be classified as either a pathological partial response (pPR) or pathological complete response (pCR), defined as complete eradication of invasive tumor cells, with a pCR conferring a significantly lower risk of recurrence. Predicting the response to NAC, however, remains a significant clinical challenge. The objective of this study was to determine if analysis of nuclear features on core biopsies using artificial intelligence (AI) can predict response to NAC. METHODS: Fifty-eight HER2-positive or triple-negative breast cancer patients were included in this study (pCR n = 37, pPR n = 21). Multiple deep convolutional neural networks were developed to automate tumor detection and nuclear segmentation. Nuclear count, area, and circularity, as well as image-based first- and second-order features including mean pixel intensity and correlation of the gray-level co-occurrence matrix (GLCM-COR) were determined. RESULTS: In univariate analysis, the pCR group had fewer multifocal/multicentric tumors, higher nuclear intensity, and lower GLCM-COR compared to the pPR group. In multivariate binary logistic regression, tumor multifocality/multicentricity (OR = 0.14, p = 0.012), nuclear intensity (OR = 1.23, p = 0.018), and GLCM-COR (OR = 0.96, p = 0.043) were each independently associated with likelihood of achieving a pCR, and the model was able to successful classify 79% of cases (62% for pPR and 89% for pCR). CONCLUSION: Analysis of tumor nuclear features using digital pathology/AI can significantly improve models to predict pathological response to NAC.

Machine Learning Frameworks to Predict Neoadjuvant Chemotherapy Response in Breast Cancer Using Clinical and Pathological Features.

PURPOSE: Neoadjuvant chemotherapy (NAC) is used to treat locally advanced breast cancer (LABC) and high-risk early breast cancer (BC). Pathological complete response (pCR) has prognostic value depending on BC subtype. Rates of pCR, however, can be variable. Predictive modeling is desirable to help identify patients early who may have suboptimal NAC response. Here, we test and compare the predictive performances of machine learning (ML) prediction models to a standard statistical model, using clinical and pathological data. METHODS: Clinical and pathological variables were collected in 431 patients, including tumor size, patient demographics, histological characteristics, molecular status, and staging information. A standard multivariable logistic regression (MLR) was developed and compared with five ML models: k-nearest neighbor classifier, random forest (RF) classifier, naive Bayes algorithm, support vector machine, and multilayer perceptron model. Model performances were measured using a receiver operating characteristic (ROC) analysis and statistically compared. RESULTS: MLR predictors of NAC response included: estrogen receptor (ER) status, human epidermal growth factor-2 (HER2) status, tumor size, and Nottingham grade. The strongest MLR predictors of pCR included HER2+ versus HER2- BC (odds ratio [OR], 0.13; 95% CI, 0.07 to 0.23; P < .001) and Nottingham grade G3 versus G1-2 (G1-2: OR, 0.36; 95% CI, 0.20 to 0.65; P < .001). The area under the curve (AUC) for the MLR was AUC = 0.64. Among the various ML models, an RF classifier performed best, with an AUC = 0.88, sensitivity of 70.7%, and specificity of 84.6%, and included the following variables: menopausal status, ER status, HER2 status, Nottingham grade, tumor size, nodal status, and presence of inflammatory BC. CONCLUSION: Modeling performances varied between standard versus ML classification methods. RF ML classifiers demonstrated the best predictive performance among all models.

Whole genome sequence of Corynebacterium kroppenstedtii isolated from a case of recurrent granulomatous mastitis.

We describe the case of a 33-year-old woman with recurrent granulomatous mastitis associated with Corynebacterium kroppenstedtii. This organism has been increasingly associated with granulomatous mastitis, specifically the cystic neutrophilic histopathologic variant, although currently there is a paucity both of reported cases and genomic sequence data. We highlight the challenges in the diagnosis and treatment of this entity, in particular focusing on the various methods of microbiologic identification, including MALDI-TOF, 16 s rRNA PCR and whole-genome sequencing.