Augmented interpretation of HER2, ER, and PR in breast cancer by artificial intelligence analyzer: enhancing interobserver agreement through a reader study of 201 cases.

BACKGROUND: Accurate classification of breast cancer molecular subtypes is crucial in determining treatment strategies and predicting clinical outcomes. This classification largely depends on the assessment of human epidermal growth factor receptor 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) status. However, variability in interpretation among pathologists pose challenges to the accuracy of this classification. This study evaluates the role of artificial intelligence (AI) in enhancing the consistency of these evaluations. METHODS: AI-powered HER2 and ER/PR analyzers, consisting of cell and tissue models, were developed using 1,259 HER2, 744 ER, and 466 PR-stained immunohistochemistry (IHC) whole-slide images of breast cancer. External validation cohort comprising HER2, ER, and PR IHCs of 201 breast cancer cases were analyzed with these AI-powered analyzers. Three board-certified pathologists independently assessed these cases without AI annotation. Then, cases with differing interpretations between pathologists and the AI analyzer were revisited with AI assistance, focusing on evaluating the influence of AI assistance on the concordance among pathologists during the revised evaluation compared to the initial assessment. RESULTS: Reevaluation was required in 61 (30.3%), 42 (20.9%), and 80 (39.8%) of HER2, in 15 (7.5%), 17 (8.5%), and 11 (5.5%) of ER, and in 26 (12.9%), 24 (11.9%), and 28 (13.9%) of PR evaluations by the pathologists, respectively. Compared to initial interpretations, the assistance of AI led to a notable increase in the agreement among three pathologists on the status of HER2 (from 49.3 to 74.1%, p < 0.001), ER (from 93.0 to 96.5%, p = 0.096), and PR (from 84.6 to 91.5%, p = 0.006). This improvement was especially evident in cases of HER2 2+ and 1+, where the concordance significantly increased from 46.2 to 68.4% and from 26.5 to 70.7%, respectively. Consequently, a refinement in the classification of breast cancer molecular subtypes (from 58.2 to 78.6%, p < 0.001) was achieved with AI assistance. CONCLUSIONS: This study underscores the significant role of AI analyzers in improving pathologists' concordance in the classification of breast cancer molecular subtypes.

PD-1 inhibitor plus oncolytic vaccinia virus is a safe and effective treatment option for metastatic renal cell carcinoma.

BACKGROUND: Although a combination of immune checkpoint inhibitors (ICIs) is recommended as the first line treatment option for metastatic renal cell carcinoma (mRCC), several immune-related adverse events (irAEs) occur, especially hepatitis. We explored the therapeutic benefits and safety profile of combining oncolytic vaccinia virus, JX-594, with a programmed cell death protein-1 (PD-1) inhibitor. METHODS: We used early-stage and advanced-stage orthotopic murine mRCC models developed by our group. PD-1 inhibitor monotherapy or a PD-1 inhibitor combined with either JX-594 or a cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) inhibitor were systemically injected through the peritoneum. An immunofluorescence analysis was performed to analyze the tumor immune microenvironment (TIME). irAEs were assessed in terms of hepatitis. RESULTS: In the early-stage mRCC model mice, the combination of JX-594 and a PD-1 inhibitor significantly decreased the primary tumor size and number of lung nodules, compared with the ICI combination, but the JX-594 and PD-1 inhibitor combination and ICI combination did not differ significantly in the advanced-stage mRCC model mice. The JX-594 and PD-1 inhibitor combination induced tumor-suppressing TIME changes in both the early- and advanced-stage mRCC models. Furthermore, mice treated with the ICI combination had significantly greater hepatic injuries than those treated with the JX-594 and PD-1 inhibitor combination which was evaluated in early-stage mRCC model. CONCLUSIONS: The JX-594 and PD-1 inhibitor combination effectively reduced primary tumors and the metastatic burden, similar to ICI combination therapy, through dynamic remodeling of the TIME. Furthermore, hepatitis was significantly decreased in the JX-594 and PD-1 inhibitor combination group, suggesting the potential benefit of that combination for reducing ICI-induced toxicity.

Quantitative proteomics and immunohistochemistry uncover NT5DC2 as a diagnostic biomarker for papillary urothelial carcinoma.

The accurate diagnosis of papillary urothelial carcinoma (PUC) is frequently challenging due to benign mimickers. Other than morphology-based diagnostic criteria, reliable biomarkers for differentiating benign and malignant papillary urothelial neoplasms remain elusive, so we sought to discover new markers to address this challenge. We first performed tandem mass spectrometry-based quantitative proteomics using diverse papillary urothelial lesions, including PUC, urothelial papilloma (UP), inverted urothelial papilloma, and cystitis cystica. We prioritized potential diagnostic biomarkers using machine learning, and subsequently validated through immunohistochemistry (IHC) in two independent cohorts. Metabolism, transport, cell cycle, development, and immune response functions were differentially enriched between malignant and benign papillary neoplasms. RhoB and NT5DC2 were shortlisted as optimal candidate markers for PUC diagnosis. In our pilot study using IHC, NT5DC2 was subsequently selected as its expression consistently differed in PUC (p = 0.007). Further validation of NT5DC2 using 49 low-grade (LG) urothelial lesions, including 15 LG-PUCs and 17 UPs, which are the most common mimickers, concordantly revealed lower IHC expression levels in LG-PUC (p = 0.0298). Independent external validation with eight LG-PUCs and eight UPs confirmed the significant downregulation of NT5DC2 in LG-PUC (p = 0.0104). We suggest that NT5DC2 is a potential IHC biomarker for differentiating LG-PUC from its benign mimickers, especially UP.

Glomerulonephritis following COVID-19 infection or vaccination: a multicenter study in South Korea.

BACKGROUND: Despite the widespread impact of the severe acute respiratory syndrome coronavirus 2 (coronavirus disease 2019, COVID-19) and vaccination in South Korea, our understanding of kidney diseases following these events remains limited. We aimed to address this gap by investigating the characteristics of glomerular diseases following the COVID-19 infection and vaccination in South Korea. METHODS: Data from multiple centers were used to identify de novo glomerulonephritis (GN) cases with suspected onset following COVID-19 infection or vaccination. Retrospective surveys were used to determine the COVID-19-related histories of patients who were initially not implicated. Bayesian structural time series and autoregressive integrated moving average models were used to determine causality. RESULTS: Glomerular diseases occurred shortly after the infection or vaccination. The most prevalent postinfection GN was podocytopathy (42.9%), comprising primary focal segmental glomerulosclerosis and minimal change disease, whereas postvaccination GN mainly included immunoglobulin A nephropathy (IgAN; 57.9%) and Henoch-Schönlein purpura nephritis (HSP; 15.8%). No patient progressed to end-stage kidney disease. Among the patients who were initially not implicated, nine patients with IgAN/HSP were recently vaccinated against COVID-19. The proportion of glomerular diseases changed during the pandemic in South Korea, with an increase in acute interstitial nephritis and a decrease in pauci-immune crescentic GN. CONCLUSION: This study showed the characteristics of GNs following COVID-19 infection or vaccination in South Korea. Understanding these associations is crucial for developing effective patient management and vaccination strategies. Further investigation is required to fully comprehend COVID-19's impact on GN.