Explore the vast range of titles available.

Breast cancer is the most common malignant disease worldwide, with over 2.26 million new cases in 2020. Its diagnosis is determined by a histological review of breast biopsy specimens, which can be labor-intensive, subjective, and error-prone. Artificial Intelligence (AI)—based tools can support cancer detection and classification in breast biopsies ensuring rapid, accurate, and objective diagnosis. We present here the development, external clinical validation, and deployment in routine use of an AI-based quality control solution for breast biopsy review. The underlying AI algorithm is trained to identify 51 different types of clinical and morphological features, and it achieves very high accuracy in a large, multi-site validation study. Specifically, the area under the receiver operating characteristic curves (AUC) for the detection of invasive carcinoma and of ductal carcinoma in situ (DCIS) are 0.99 (specificity and sensitivity of 93.57 and 95.51%, respectively) and 0.98 (specificity and sensitivity of 93.79 and 93.20% respectively), respectively. The AI algorithm differentiates well between subtypes of invasive and different grades of in situ carcinomas with an AUC of 0.97 for invasive ductal carcinoma (IDC) vs. invasive lobular carcinoma (ILC) and AUC of 0.92 for DCIS high grade vs. low grade/atypical ductal hyperplasia, respectively, as well as accurately identifies stromal tumor-infiltrating lymphocytes (TILs) with an AUC of 0.965. Deployment of this AI solution as a real-time quality control solution in clinical routine leads to the identification of cancers initially missed by the reviewing pathologist, demonstrating both clinical utility and accuracy in real-world clinical application.

Most deep learning (DL) studies have focused on neoplastic pathology, with the realm of inflammatory pathology remaining largely untouched. To investigate the use of DL for nonneoplastic gastric biopsies. Gold standard diagnoses were blindly established by 2 gastrointestinal pathologists. For phase 1, 300 classic cases (100 normal, 100 , 100 reactive gastropathy) that best displayed the desired pathology were scanned and annotated for DL analysis. A total of 70% of the cases for each group were selected for the training set, and 30% were included in the test set. The software assigned colored labels to the test biopsies, which corresponded to the area of the tissue assigned a diagnosis by the DL algorithm, termed area distribution (AD). For Phase 2, an additional 106 consecutive nonclassical gastric biopsies from our archives were tested in the same fashion. For Phase 1, receiver operating curves showed near perfect agreement with the gold standard diagnoses at an AD percentage cutoff of 50% for normal (area under the curve [AUC] = 99.7%) and (AUC = 100%), and 40% for reactive gastropathy (AUC = 99.9%). Sensitivity/specificity pairings were as follows: normal (96.7%, 86.7%), (100%, 98.3%), and reactive gastropathy (96.7%, 96.7%). For phase 2, receiver operating curves were slightly less discriminatory, with optimal AD cutoffs reduced to 40% across diagnostic groups. The AUCs were 91.9% for normal, 100% for , and 94.0% for reactive gastropathy. Sensitivity/specificity parings were as follows: normal (73.7%, 79.6%), (95.7%, 100%), reactive gastropathy (100%, 62.5%). A convolutional neural network can serve as an effective screening tool/diagnostic aid for gastritis.

Background/rationale Image-guided needle biopsies are commonly used to diagnose musculoskeletal tumors, but nondiagnostic (ND) results can delay diagnosis and treatment. It is important to understand which factors or diagnoses predispose to a ND result so that appropriate patient education or a possible change in the clinical plan can be made. Currently it is unclear which factors or specific lesions are more likely to lead to a ND result after image-guided needle biopsy. Questions/purposes We therefore identified specific factors and diagnoses most likely to yield ND results. We also asked whether an image-guided needle biopsy of bone and soft tissue lesions is an accurate and clinically useful tool. Methods We retrospectively reviewed data from a prospectively collected database for a case-control study of 508 image-guided needle biopsies of patients with suspected musculoskeletal tumors between 2003 and 2008. Results The interpretations of 453 of the 508 (89%) needle biopsies were accurate and clinically useful. Forty-five biopsies (9%) were ND and 10 (2%) were incorrect (IC). Bone lesions had a higher ND rate than soft tissue lesions (13% vs. 4%). The specific diagnosis with the highest ND rate was histiocytosis. Elbow and forearm locations had higher ND rates than average. Malignant tumors had a higher IC rate than benign tumors (5% vs. 0%); fibromyxoid sarcoma and rare subtypes of osteosarcoma had higher IC rates than other diagnoses. Repeat needle or open biopsies were performed in 71 (14%) patients. Bone lesions were more likely than soft tissue lesions to require repeat biopsies (18% vs. 9%). Conclusions A high rate of accuracy and clinical usefulness is possible with image-guided needle biopsies of musculoskeletal lesions. We believe these biopsies appropriate in selected circumstances but a key factor for appropriate use is an experienced musculoskeletal tumor team with frequent communication to correlate clinical, radiographic, and histologic information for each patient. Level of Evidence Level II, prognostic study. See the Guidelines for Authors for a complete description of levels of evidence.

Incidentalomas detected by 18 FDG-PET are unusual (2-3% of all PET scans) but have a higher risk of cancer (14-50%) compared to background incidence [4]. [...]a focal nodule that is 18 FDG-PET-avid is an indication for FNA. CT and MRI features can not determine the risk of malignancy, except in very advanced cases that are unlikely to be incidental. [...]more data are available, incidentalomas seen on CT or MRI should undergo dedicated thyroid sonographic evaluation.

This proof-of-concept study was designed to determine if functional, pharmacodynamic profiles relevant to targeted therapy could be derived from live human melanoma samples using a novel automated platform. A series of 13 melanoma cell lines was briefly exposed to a BRAF inhibitor (PLX-4720) on a platform employing automated fluidics for sample processing. Levels of the phosphoprotein p-ERK in the mitogen-activated protein kinase (MAPK) pathway from treated and untreated sample aliquots were determined using a bead-based immunoassay. Comparison of these levels provided a determination of the pharmacodynamic effect of the drug on the MAPK pathway. A similar ex vivo analysis was performed on fine needle aspiration (FNA) biopsy samples from four murine xenograft models of metastatic melanoma, as well as 12 FNA samples from patients with metastatic melanoma. Melanoma cell lines with known sensitivity to BRAF inhibitors displayed marked suppression of the MAPK pathway in this system, while most BRAF inhibitor-resistant cell lines showed intact MAPK pathway activity despite exposure to a BRAF inhibitor (PLX-4720). FNA samples from melanoma xenografts showed comparable ex vivo MAPK activity as their respective cell lines in this system. FNA samples from patients with metastatic melanoma successfully yielded three categories of functional profiles including: MAPK pathway suppression; MAPK pathway reactivation; MAPK pathway stimulation. These profiles correlated with the anticipated MAPK activity, based on the known BRAF mutation status, as well as observed clinical responses to BRAF inhibitor therapy. Pharmacodynamic information regarding the ex vivo effect of BRAF inhibitors on the MAPK pathway in live human melanoma samples can be reproducibly determined using a novel automated platform. Such information may be useful in preclinical and clinical drug development, as well as predicting response to targeted therapy in individual patients.

Pathology studies using convolutional neural networks (CNNs) have focused on neoplasms, while studies in inflammatory pathology are rare. We previously demonstrated a CNN that differentiates reactive gastropathy, Helicobacter pylori gastritis (HPG), and normal gastric mucosa.CONTEXT.—Pathology studies using convolutional neural networks (CNNs) have focused on neoplasms, while studies in inflammatory pathology are rare. We previously demonstrated a CNN that differentiates reactive gastropathy, Helicobacter pylori gastritis (HPG), and normal gastric mucosa.To determine whether a CNN can differentiate the following 2 gastric inflammatory patterns: autoimmune gastritis (AG) and HPG.OBJECTIVE.—To determine whether a CNN can differentiate the following 2 gastric inflammatory patterns: autoimmune gastritis (AG) and HPG.Gold standard diagnoses were blindly established by 2 gastrointestinal (GI) pathologists. One hundred eighty-seven cases were scanned for analysis by HALO-AI. All levels and tissue fragments per slide were included for analysis. The cases were randomized, 112 (60%; 60 HPG, 52 AG) in the training set and 75 (40%; 40 HPG, 35 AG) in the test set. A HALO-AI correct area distribution (AD) cutoff of 50% or more was required to credit the CNN with the correct diagnosis. The test set was blindly reviewed by pathologists with different levels of GI pathology expertise as follows: 2 GI pathologists, 2 general surgical pathologists, and 2 residents. Each pathologist rendered their preferred diagnosis, HPG or AG.DESIGN.—Gold standard diagnoses were blindly established by 2 gastrointestinal (GI) pathologists. One hundred eighty-seven cases were scanned for analysis by HALO-AI. All levels and tissue fragments per slide were included for analysis. The cases were randomized, 112 (60%; 60 HPG, 52 AG) in the training set and 75 (40%; 40 HPG, 35 AG) in the test set. A HALO-AI correct area distribution (AD) cutoff of 50% or more was required to credit the CNN with the correct diagnosis. The test set was blindly reviewed by pathologists with different levels of GI pathology expertise as follows: 2 GI pathologists, 2 general surgical pathologists, and 2 residents. Each pathologist rendered their preferred diagnosis, HPG or AG.At the HALO-AI AD percentage cutoff of 50% or more, the CNN results were 100% concordant with the gold standard diagnoses. On average, autoimmune gastritis cases had 84.7% HALO-AI autoimmune gastritis AD and HP cases had 87.3% HALO-AI HP AD. The GI pathologists, general anatomic pathologists, and residents were on average, 100%, 86%, and 57% concordant with the gold standard diagnoses, respectively.RESULTS.—At the HALO-AI AD percentage cutoff of 50% or more, the CNN results were 100% concordant with the gold standard diagnoses. On average, autoimmune gastritis cases had 84.7% HALO-AI autoimmune gastritis AD and HP cases had 87.3% HALO-AI HP AD. The GI pathologists, general anatomic pathologists, and residents were on average, 100%, 86%, and 57% concordant with the gold standard diagnoses, respectively.A CNN can distinguish between cases of HPG and autoimmune gastritis with accuracy equal to GI pathologists.CONCLUSIONS.—A CNN can distinguish between cases of HPG and autoimmune gastritis with accuracy equal to GI pathologists.

Background: Determining the site of origin for metastatic well-differentiated neuroendocrine tumors (WDNETs) is challenging, and immunohistochemical (IHC) profiles do not always lead to a definitive diagnosis. We sought to determine if a deep-learning convolutional neural network (CNN) could improve upon established IHC profiles in predicting the site of origin in a cohort of WDNETs from the common primary sites. Materials and Methods: Hematoxylin and eosin (H&E)-stained tissue microarrays (TMAs) were created using 215 WDNETs arising from the known primary sites. A CNN trained and tested on 60% (n = 130) and 40% (n = 85) of these cases, respectively. One hundred and seventy-nine cases had TMA tissue remaining for the IHC analysis. These cases were stained with IHC markers pPAX8, CDX2, SATB2, and thyroid transcription factor-1 (markers of pancreas/duodenum, ileum/jejunum/duodenum, colorectum/appendix, and lung WDNET sites of origin, respectively). The CNN diagnosis was deemed correct if it designated a majority or plurality of the tumor area as the known site of origin. The IHC diagnosis was deemed correct if the most specific marker for a particular site of origin met an H-score threshold determined by two pathologists. Results: When all cases were considered, the CNN correctly identified the site of origin at a lower rate compared to IHC (72% vs. 82%, respectively). Of the 85 cases in the CNN test set, 66 had sufficient TMA material for IHC stains, thus 66 cases were available for a direct case-by-case comparison of IHC versus CNN. The CNN correctly identified 70% of these cases, while IHC correctly identified 76%, a finding that was not statistically significant (P = 0.56). Conclusion: A CNN can identify WDNET site of origin at an accuracy rate close to the current gold standard IHC methods.

Multikinase inhibitors (MKIs) targeting VEGF receptors and other receptor tyrosine kinases have shown considerable activity in clinical trials of thyroid cancer. Thyroid cancer frequently exhibits activation of the RAS/RAF/MEK/ERK pathway. In other types of cancer, paradoxical ERK activation has emerged as a potential resistance mechanism to RAF-inhibiting drugs including MKIs such as sorafenib and pazopanib. We therefore queried whether the MEK inhibitor trametinib, could augment the activity of pazopanib in thyroid cancer cell lines. Trametinib potently inhibited growth in vitro (GI50 1.1-4.8 nM), whereas pazopanib had more limited in vitro activity, as anticipated (GI50 1.4-7.1 µM). We observed progressive upregulation of ERK activity with pazopanib treatment, an effect abrogated by trametinib. For xenografts (bearing either KRASG12R or BRAFV600E mutations), the combination of trametinib and pazopanib led to sustained shrinkage in tumor volume by 50% or more, compared to pre-treatment baseline. Trametinib also was highly effective as a single agent, compared to pazopanib alone. These preclinical findings support the evaluation of trametinib, alone or in combination with pazopanib or other kinase inhibitors, in thyroid cancer clinical trials. We highlight the importance of pharmacodynamic assessment of the ERK pathway for patients enrolled in trials involving MKIs.

Recent studies indicate that tumor suppressor genes can be epigenetically silenced through promoter hypermethylation. To further understand epigenetic alterations in cholangiocarcinoma, we have studied the methylation profiles of 12 candidate tumor suppressor genes (APC, E-cadherin/CDH1, MGMT, RASSF1A, GSTP, RAR-β, p14ARF, p15INK4b, p16INK4a, p73, hMLH1 and DAPK) in 72 cases of cholangiocarcinoma, including equal number cases of intrahepatic cholangiocarcinoma and extrahepatic cholangiocarcinoma. A total of 10 cases of benign biliary epithelia were included as controls. The methylation status of tumor suppressor genes was analyzed using methylation-specific PCR. We found that 85% of all cholangiocarcinomas had methylation of at least one tumor suppressor gene. The frequency of tumor suppressor gene methylation in cholangiocarcinoma was: RASSF1A (65%), p15INK4b (50%), p16INK4a (50%), APC (46%), E-cadherin/CDH1 (43%), p14ARF (38%), p73 (36%), MGMT (33%), hMHL1 (25%), GSTP (14%), RAR-β (14%) and DAPK (3%). Although single tumor suppressor gene methylation can be seen in benign biliary epithelium, methylation of multiple tumor suppressor genes is only seen in cholangiocarcinoma. About 70% (50/72) of the cholangiocarcinomas had three or more tumor suppressor genes methylated and 52% (38/72) of cases had four or more tumor suppressor genes methylated. Concerted methylation of multiple tumor suppressor genes was closely associated with methylation of RASSF1A, p16 and/or hMHL1. Methylation of RASSF1A was more common in extrahepatic cholangiocarcinoma than intrahepatic cholangiocarcinoma (83 vs 47%, P=0.003) while GSTP was more frequently seen in intrahepatic compared to extrahepatic cholangiocarcinoma (31 vs 6%, P=0.012). Our study indicates that methylation of promoter CpG islands of tumor suppressor genes is a common epigenetic event in cholangiocarcinoma. Based on distinct methylation profiles, intrahepatic cholangiocarcinoma and extrahepatic cholangiocarcinoma are two closely related but biologically unique neoplastic processes. Taking advantage of the unique concurrent methylation profile of multiple genes in cholangiocarcinoma may facilitate the distinction of cholangiocarcinoma from benign biliary epithelium in clinical settings.

To report an unusual occurrence of recurrent hyperparathyroidism due to papillary thyroid carcinoma. We describe the clinical history, physical examination findings, laboratory values, imaging findings, and pathologic findings of a woman who developed recurrent hyperparathyroidism 13 years after successful parathyroidectomy. A 59-year-old woman presented to our clinic with recurrent primary hyperparathyroidism. In 1994, she presented with nephrolithiasis and underwent resection of a right superior parathyroid adenoma that resulted in clinical and biochemical cure. Her clinical course had been followed at periodic intervals, and she had been symptom-free and normocalcemic. In 2007, she again developed nephrolithiasis and was documented to have recurrent hyperparathyroidism. Imaging studies suggested a parathyroid adenoma near the right inferior pole of the thyroid. The patient had reoperative neck exploration. No obvious parathyroid adenoma was found and a right thyroid lobectomy was performed, which resulted in normalization of intraoperative intact parathyroid hormone levels, and the incision was closed. Final pathology demonstrated no parathyroid adenoma, but instead, a 1-cm papillary thyroid carcinoma that stained positive for parathyroid hormone. More than 6 months after surgery, she remains clinically and biochemically cured. Recurrent hyperparathyroidism occurs secondary to multiple causes. This case demonstrates the challenge a surgeon faces in managing recurrent disease and highlights a rare phenomenon of papillary thyroid cancer causing recurrent hyperparathyroidism.