Reynolds, Jordan P

Foundation models are reshaping computational pathology by enabling scalable task-agnostic representations of histopathological whole-slide images (WSIs). Unlike earlier task-specific deep learning systems, pathology foundation models (PFMs) leverage massive whole-slide image repositories and self-supervised Vision Transformer architectures to achieve broad generalization and few-shot adaptability. Their evolution reflects a shift from weakly supervised approaches such as Clustering-Constrained Attention Multiple Instance Learning (CLAM) and hierarchical architectures such as Hierarchical Image Pyramid Transformer (HIPT) to large-scale efforts including foundation models, UNI, Virchow, Phikon, CONtrastive learning from Captions for Histopathology (CONCH), GigaPath, H-Optimus, Transformer-Based Pathology Image and Text Alignment Network (TITAN), and the Mayo Clinic Atlas. These models demonstrate impressive performance across diagnostic and prognostic benchmarks while also opening pathways for multimodal integration with genomics and clinical data. Yet significant barriers remain including inconsistent generalization across institutions, interpretability lagging behind clinical needs, and slow integration into routine laboratory workflows. Certain domains of anatomic pathology such as cytopathology, transplant pathology, frozen sections, and rare tumor subtypes remain particularly resistant to current models. Here, we review the development of PFMs, critically evaluate their strengths and limitations, and outline priorities for their safe and effective clinical translation. We argue that the next phase of PFM development will depend on rigorous benchmarking, pathologist-in-the-loop deployment, and multimodal fusion ensuring these models evolve from research tools into clinically robust systems.

Abstract Diversity, equity, and inclusion is a powerful goal which many of us strive toward in medicine, both in patient care and administrative leadership. As the world evolves, the practice of medicine must evolve with it. We are cognizant of the importance of the history of our medical specialties. If we do not acknowledge all parts of our history, we are doomed to repeat it. This special issue is unique and unlike anything that has previously been published in Diagnostic Cytopathology. This issue looks at some of the history of cytopathology. This historical review is followed by some of the present state of cytopathology. There are insights into global cytopathology. The final portion of this issue examines the critical need for cytotechnology schools in the United States. All of these areas are critical to the past, present, and future of cytopathology.

Abstract Metastatic castration-resistant prostate cancer (mCRPC) is an aggressive malignancy with poor outcomes. To investigate novel therapeutic strategies, we characterized three new metastatic prostate cancer patient derived-tumor xenograft (PDTX) models and developed 3D spheroids from each to investigate molecular targeted therapy combinations including CDK4/6 inhibitors (CDK4/6i) with AKT inhibitors (ATKi). Metastatic prostate cancer tissue was collected and three PDTX models were established and characterized using whole-exome sequencing. PDTX 3D spheroids were developed from these three PDTXs to show resistance patterns and test novel molecular-targeted therapies. CDK4/6i's were combined with AKTi's to assess synergistic antitumor response to prove our hypothesis that blockade of AKT overcomes drug resistance to CDK4/6i. This combination was evaluated in PDTX three-dimensional (3D) spheroids and in vivo experiments with responses measured by tumor volumes, PSA, and Ga-68 PSMA-11 PET-CT imaging. We demonstrated CDK4/6i's with AKTi's possess synergistic antitumor activity in three mCRPC PDTX models. These models have multiple unique pathogenic and deleterious genomic alterations with resistance to single-agent CDK4/6i's. Despite this, combination therapy with AKTi's was able to overcome resistance mechanisms. The IHC and Western blot analysis confirmed on target effects, whereas tumor volume, serum PSA ELISA, and radionuclide imaging demonstrated response to therapy with statistically significant SUV differences seen with Ga-68 PSMA-11 PET-CT. These preclinical data demonstrating antitumor synergy by overcoming single-agent CDK 4/6i as well as AKTi drug resistance provide the rational for a clinical trial combining a CDK4/6i with an AKTi in patients with mCRPC whose tumor expresses wild-type retinoblastoma 1.

The Papanicolaou Society of Cytopathology (PSC) reporting system classifies pancreatobiliary samples into six categories (I–VI), providing guidance for personalized management. As the World Health Organization (WHO) has been preparing an updated reporting system for pancreatobiliary cytopathology, this systematic review aimed to evaluate the risk of malignancy (ROM) of each PSC category, also the sensitivity and specificity of pancreatic FNA cytology using the current PSC system. Five databases were investigated with a predefined search algorithm. Inclusion and exclusion criteria were applied to select the eligible studies for subsequent data extraction. A study quality assessment was also performed. Eight studies were included in the qualitative analysis. The ROM of the PSC categories I, II, III, IV, V, VI were in the ranges of 8–50%, 0–40%, 28–100%, 0–31%, 82–100%, and 97–100%, respectively. Notably, the ROM IVB (“neoplastic—benign”) subcategory showed a 0% ROM. Four of the included studies reported separately the ROMs for the IVO subcategory (“neoplastic—other”; its overall ROM ranged from 0 to 34%) with low (LGA) and high-grade atypia (HGA). ROM for LGA ranged from 4.3 to 19%, whereas ROM for HGA from 64 to 95.2%. When the subcategory IVO with HGA was considered as cytologically positive, together with the categories V and VI, there was a higher sensitivity of pancreatic cytology, at minimal expense of the specificity. Evidence suggests the proposed WHO international system changes—shifting the IVB entities into the “benign/negative for malignancy” category and establishing two new categories, the “pancreatic neoplasm, low-risk/grade” and “pancreatic neoplasm, high-risk/grade”—could stratify pancreatic neoplasms more effectively than the current PSC system.

The Paris System (TPS) for Reporting Urinary Cytology is a standardized, evidence-based reporting system, comprising seven diagnostic categories: nondiagnostic, negative for high-grade urothelial carcinoma (NHGUC), atypical urothelial cells (AUC), suspicious for high-grade urothelial carcinoma (SHGUC), HGUC, low-grade urothelial neoplasm (LGUN), and other malignancies. This study aimed to calculate the pooled risk of high-grade malignancy (ROHM) of each category and demonstrate the diagnostic accuracy of urine cytology reported with TPS. Four databases (PubMed, Embase, Scopus, Web of Science) were searched. Specific inclusion and exclusion criteria were applied, while data were extracted and analyzed both qualitatively and quantitatively. The pooled ROHM was 17.70% for the nondiagnostic category (95% CI, 0.0650; 0.3997), 13.04% for the NHGUC (95% CI, 0.0932; 0.1796), 38.65% for the AUC (95% CI, 0.3042; 0.4759), 12.45% for the LGUN (95% CI, 0.0431; 0.3101), 76.89 for the SHGUC (95% CI, 0.7063; 0.8216), and 91.79% for the HGUC and other malignancies (95% CI, 0.8722; 0.9482). A summary ROC curve was created and the Area Under the Curve (AUC) was 0.849, while the pooled sensitivity was 0.669 (95% CI, 0.589; 0.741) and false-positive rate was 0.101 (95% CI, 0.063; 0.158). In addition, the pooled DOR of the included studies was 21.258 (95% CI, 14.336; 31.522). TPS assigns each sample into a diagnostic category linked with a specific ROHM, guiding clinical management.

Greater than 2000 cases of vaping‐associated pulmonary injury and 39 deaths reported in the United States have heightened public awareness and assumed major public health significance over recent months. Currently, knowledge of the effects of various vaping agents on the respiratory tract is in its infancy. Additional understanding of the agents implicated in these lung injuries as well as a correlation between bronchoalveolar lavage and lung biopsy samples is required.

Introduction: The increasing use of endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) cytology to examine pancreatic neoplasms has led to an increase in the diagnosis of metastases to the pancreas. Renal cell carcinoma (RCC) is the most common metastasis to the pancreas. Our study examines 33 cases of metastatic RCC to the pancreas sampled by EUS-FNA from four large tertiary care hospitals. Materials and Methods: We searched the cytopathology database for RCC metastatic to the pancreas diagnosed by EUS-FNA from January 2005 to January 2015. Patient age, history of RCC, nephrectomy history, follow-up postnephrectomy, radiological impression, and EUS-FNA cytologic diagnosis were reviewed. Results: Thirty-three patients were identified. The average age was 67.5 years (range, 49–84 years). Thirty-two patients had a previous documented history of RCC. One patient had the diagnosis of pancreatic metastasis at the same time of the kidney biopsy. Thirty-one patients had been treated with nephrectomy. Twenty-seven patients were being monitored annually by computed tomography or magnetic resonance imaging. Twenty-five patients had multiple masses by imaging, but 8 patients had a single mass in the pancreas at the time of EUS-FNA. EUS-FNA of 20 cases showed classic morphology of RCC. Thirteen cases had either “atypical” clinical-radiologic features or morphologic overlaps with primary pancreatic neoplasms or other neoplasms. Cell blocks were made on all 13 cases and immunochemical stains confirmed the diagnosis. Conclusions: EUS-FNA cytology is useful for the diagnosis of metastatic RCC to the pancreas. Cytomorphology can be aided with patient history, imaging analyses, cell blocks, and immunochemical stains.

Background:The cytologic diagnosis of atypia of undetermined significance or follicular lesion of undetermined significance (AUS/FLUS) is controversial because of variation in how it is applied in practice, as well as uncertainty about patient management. We aimed to assess the percentage of thyroid fine-needle aspiration biopsies (FNABs) with AUS/FLUS diagnoses in different North American and European practice settings (e.g. community, academic, etc.), assess whether patients were managed according to current guidelines, and determine patient outcomes.Materials and Methods:A detailed questionnaire survey was posted in secure websites used separately by cytopathologists and cytotechnologists. The questionnaire was posted from August 1 through December 31, 2013.Results:Endocrinologists and cytopathologists performed 51.7% and 37.1% of thyroid FNABs, respectively. The Bethesda reporting system for thyroid FNAB was used in 90% of practices. The rate of AUS/FLUS varied widely among institutions, with 46.1% of represented institutions reporting AUS/FLUS rates of 3–10%. The median follow-up rate of patients with an initial AUS/FLUS diagnosis was 70% (range, 10–100%). For the majority of represented institutions (86.4%), patients with initial AUS/FLUS diagnosis had follow-up with endocrinologists. Of repeat AUS/FLUS thyroid FNABs, a median of 52% was considered benign, and 18% were suspicious of or positive for malignancy (median, 10% and 7.5%, respectively).Conclusions:Reporting of the AUS/FLUS category varied widely among different institutions. The median follow-up rate was lower than published guidelines. The most common follow-up diagnosis was benign thyroid nodule. Improved standardization of cytologic criteria should be adopted to reduce such variation.

Background:Bronchial brush (BB) cytology carries low sensitivity for detecting neuroendocrine carcinomas (NECs), including typical carcinoid (TC) tumors of the lung. We aimed to investigate the detection of neuroendocrine tumors including TC through BB routine cytology cell block (CB), immunohistochemistry (IHC), and fluorescencein situhybridization (FISH).Materials and Methods:A SNOMED search showed 187 lung biopsy or resection specimens from 2008 through 2011 containing neuroendocrine or carcinoid in the diagnosis. Residual BB specimens retained in PreservCyt were used to prepare a ThinPrep slide for FISH analysis. CBs were stained with H and E and IHC for chromogranin and synaptophysin.Results:Of the 187 cases, 16 had residual BB material available within 1 year of diagnosis and were used in CB preparation for IHC and FISH slides. Cytologic evaluation determined 1 case positive for malignancy (small cell lung carcinoma [SCLC]), 1 suspicious for adenocarcinoma, and 14 negative for malignancy. On the basis of histologic diagnosis, FISH was performed. SCLC showed polysomy (86% abnormal cells); 2 TC tumors showed a gain of 7p12 (15% abnormal cells) and a gain of 5q15 (72% abnormal cells), respectively. Two cases had CBs with positive immunoreactivity for chromogranin and synaptophysin. The sensitivity for detection of NEC was 18.8%, 15.4%, and 25% for cytologic evaluation, CB, and FISH, respectively.Conclusion:Neuroendocrine tumors, including TC are difficult to detect with BB cytologic evaluation, most likely because tumor cells lack in the specimen. Assessment of further studies is needed to explore the role of cytology and ancillary methods for detection of these tumors.

BACKGROUNDThe cytologic diagnosis of urothelial carcinoma (UC) of the upper urothelial tract (UT) is challenging. Using the UroVysion probe set adds diagnostic value for the detection of bladder cancer in voided urine. In instrumented UT specimens, the authors encountered positive UT cytology and fluorescence in situ hybridization (FISH) cases that did not demonstrate subsequent UT carcinoma.METHODSThe performance of cytology and FISH in the presence or absence of concomitant bladder cancer within 2 years was compared in 61 patients (112 samples) from 2003 through 2009. The mean follow‐up was 3.2 years. The authors also compared the performance of near‐tetrasomy versus hypertetrasomy. Biopsy confirmation of UTUC in 21 patients was considered the gold standard.RESULTSCytology alone was found to be poorly sensitive (38%) but highly specific (89%) for the detection of UTUC. FISH was found to increase the sensitivity of cytology and decrease specificity. Tetrasomy FISH resulted in many false‐positive cases. Other false‐positive FISH results were likely due to the presence of bladder cancer cells contaminating the UT specimen.CONCLUSIONSCaution should be used when evaluating instrumented urine specimens of the UT from patients with a previous history of bladder carcinoma, and tetrasomy FISH results should not be interpreted as abnormal because it significantly lowers the specificity of the test. The combination of cytology and FISH appears to have good specificity while maintaining good sensitivity in evaluating UTUC when using modified scoring criteria for the appropriate patient population. Cancer (Cancer Cytopathol) 2014;122:459–467. © 2014 American Cancer Society.