Epithelial-myoepithelial carcinoma (EMC) is a rare salivary gland tumor of presumed intercalated duct origin with a low risk of metastasis and mortality. Factors shown to affect behavior include positive margins, vascular invasion, necrosis, and myoepithelial anaplasia. The latter category and dedifferentiated EMCs have been separated on the basis of presumed myoepithelial versus ductal origin, respectively. Three additional cases of typical EMC with transition to high-grade carcinoma are presented. Two of the tumors were stained with CAM5.2, 34betaE12, cytokeratin 14, p63, S100, calponin, smooth muscle actin, and muscle-specific actin. All tumors showed a gradual transition to a high-grade carcinoma from an EMC, each composed of clear cells even in the high-grade regions. One case also showed a discrete area with ductal lumina and another had plasmacytoid morphology. Squamous differentiation was seen in all cases as well. A consistent immunostaining pattern was not noted. Areas with focal lumina were diffusely positive for CAM5.2 only. Areas with clear cells showed patchy S100 positivity only, whereas cytokeratin 14 and 34betaE12-stained squamous pearls. The case with plasmacytoid morphology was diffusely positive for p63. No immunoexpression was noted with smooth muscle actin, muscle-specific actin, or calponin. It was not possible to convincingly separate the high-grade component in these cases into ductal dedifferentiated EMC versus myoepithelial. Recently, there has been a move to abandon the term "dedifferentiation" in favor of "high-grade transformation" in other salivary gland malignancies. We report these 3 such cases, review the literature and propose that these lesions be regarded as "EMC with high-grade transformation."

Artificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent validation. With this in mind, we organized the PANDA challenge—the largest histopathology competition to date, joined by 1,290 developers—to catalyze development of reproducible AI algorithms for Gleason grading using 10,616 digitized prostate biopsies. We validated that a diverse set of submitted algorithms reached pathologist-level performance on independent cross-continental cohorts, fully blinded to the algorithm developers. On United States and European external validation sets, the algorithms achieved agreements of 0.862 (quadratically weighted κ, 95% confidence interval (CI), 0.840–0.884) and 0.868 (95% CI, 0.835–0.900) with expert uropathologists. Successful generalization across different patient populations, laboratories and reference standards, achieved by a variety of algorithmic approaches, warrants evaluating AI-based Gleason grading in prospective clinical trials.

The Gleason score is the most important prognostic marker for prostate cancer patients, but it suffers from significant observer variability. Artificial intelligence (AI) systems based on deep learning can achieve pathologist-level performance at Gleason grading. However, the performance of such systems can degrade in the presence of artifacts, foreign tissue, or other anomalies. Pathologists integrating their expertise with feedback from an AI system could result in a synergy that outperforms both the individual pathologist and the system. Despite the hype around AI assistance, existing literature on this topic within the pathology domain is limited. We investigated the value of AI assistance for grading prostate biopsies. A panel of 14 observers graded 160 biopsies with and without AI assistance. Using AI, the agreement of the panel with an expert reference standard increased significantly (quadratically weighted Cohen's kappa, 0.799 vs. 0.872; p = 0.019). On an external validation set of 87 cases, the panel showed a significant increase in agreement with a panel of international experts in prostate pathology (quadratically weighted Cohen's kappa, 0.733 vs. 0.786; p = 0.003). In both experiments, on a grouplevel, AI-assisted pathologists outperformed the unassisted pathologists and the standalone AI system. Our results show the potential of AI systems for Gleason grading, but more importantly, show the benefits of pathologist-AI synergy. Members of the ISUP Pathology Imagebase Expert Panel are listed below Acknowledgements.

Goblet cell carcinoid is an enigmatic and rare tumor involving the appendix almost exclusively. Since its identification in 1969, understanding of this disease has evolved greatly, but issues regarding its histogenesis, nomenclature and management are still conjectural. The published English language literature from 1966 to 2009 was retrieved via PubMed and reviewed. Various other names have been used for this entity such as adenocarcinoid, mucinous carcinoid, crypt cell carcinoma, and mucin-producing neuroendocrine tumor, although none have been found to be completely satisfactory or universally accepted. The tumor is thought to arise from pluripotent intestinal epithelial crypt-base stem cells by dual neuroendocrine and mucinous differentiation. GCCs present in the fifth to sixth decade and show no definite sex predominance. The most common clinical presentation is acute appendicitis, followed by abdominal pain and a mass. Fifty percent of the female patients present with ovarian metastases. The histologic hallmark of this entity is the presence of clusters of goblet cells in the lamina propria or submucosa stain for various neuroendocrine markers, though the intensity is often patchy. Atypia is usually minimal, but carcinomatous growth patterns may be seen. These may be of signet ring cell type or poorly differentiated adenocarcinoma. Recently molecular studies have shown these tumors to lack the signatures of adenocarcinoma but they have some changes similar to that of ileal carcinoids (allelic loss of chromosome 11q, 16q and 18q). The natural history of GCC is intermediate between carcinoids and adenocarcinomas of the appendix. The 5-year overall survival is 76%. The most important prognostic factor is the stage of disease. Appendectomy and right hemicolectomy are the main modalities of treatment, followed by adjuvant chemotherapy in select cases. There is some debate about the surgical approach for these tumors, and a summary of published series and recommendations are provided.

Our study validates the reliability of Wieneke scoring system in predicting malignancy in pediatric ACTs. It is simple and easy to use and therefore useful in day-to-day practice.