Predicting nodal metastases in squamous cell carcinoma of the oral tongue using artificial intelligence.

OBJECTIVE: The presence of occult nodal metastases in patients with squamous cell carcinoma (SCC) of the oral tongue has implications for treatment. Upwards of 30% of patients will have occult nodal metastases, yet a significant number of patients undergo unnecessary neck dissection to confirm nodal status. This study sought to predict the presence of nodal metastases in patients with SCC of the oral tongue using a convolutional neural network (CNN) that analyzed visual histopathology from the primary tumor alone. METHODS: Cases of SCC of the oral tongue were identified from the records of a single institution. Only patients with complete pathology data were included in the study. The primary tumors were randomized into 2 groups for training and testing, which was performed at 2 different levels of supervision. Board-certified pathologists annotated each slide. HALO-AI convolutional neural network and image software was used to perform training and testing. Receiver operator characteristic (ROC) curves and the Youden J statistic were used for primary analysis. RESULTS: Eighty-nine cases of SCC of the oral tongue were included in the study. The best performing algorithm had a high level of supervision and a sensitivity of 65% and specificity of 86% when identifying nodal metastases. The area under the curve (AUC) of the ROC curve for this algorithm was 0.729. CONCLUSION: A CNN can produce an algorithm that is able to predict nodal metastases in patients with squamous cell carcinoma of the oral tongue by analyzing the visual histopathology of the primary tumor alone.

The antiviral effects of a MEK1/2 inhibitor promote tumor regression in a preclinical model of human papillomavirus infection-induced tumorigenesis.

Human papillomaviruses (HPVs) are a significant public health concern due to their widespread transmission, morbidity, and oncogenic potential. Despite efficacious vaccines, millions of unvaccinated individuals and those with existing infections will develop HPV-related diseases for the next two decades and beyond. The continuing burden of HPV-related diseases is exacerbated by the lack of effective therapies or cures for infections, highlighting the need to identify and develop antivirals. The experimental murine papillomavirus type 1 (MmuPV1) model provides opportunities to study papillomavirus pathogenesis in cutaneous epithelium, the oral cavity, and the anogenital tract. However, to date the MmuPV1 infection model has not been used to demonstrate the effectiveness of potential antivirals. We previously reported that inhibitors of cellular MEK/ERK signaling suppress oncogenic HPV early gene expression in three-dimensional tissue cultures. Herein, we adapted the MmuPV1 infection model to determine whether MEK inhibitors have anti-papillomavirus properties in vivo. We demonstrate that oral delivery of a MEK1/2 inhibitor promotes papilloma regression in immunodeficient mice that otherwise would have developed persistent infections. Quantitative histological analyses reveal that inhibition of MEK/ERK signaling reduces E6/E7 mRNA, MmuPV1 DNA, and L1 protein expression within MmuPV1-induced lesions. These data suggest that MEK1/2 signaling is essential for both early and late MmuPV1 replication events supporting our previous findings with oncogenic HPVs. We also provide evidence that MEK inhibitors protect mice from developing secondary tumors. Thus, our data suggest that MEK inhibitors have potent antiviral and anti-tumor properties in a preclinical mouse model and merit further investigation as papillomavirus antiviral therapies.

Inhibition of Cellular MEK/ERK Signaling Suppresses Murine Papillomavirus Type 1 Replicative Activities and Promotes Tumor Regression.

Human papillomaviruses (HPVs) are a significant public health concern due to their widespread transmission, morbidity, and oncogenic potential. Despite efficacious vaccines, millions of unvaccinated individuals and those with existing infections will develop HPV-related diseases for the next two decades. The continuing burden of HPV-related diseases is exacerbated by the lack of effective therapies or cures for most infections, highlighting the need to identify and develop antivirals. The experimental murine papillomavirus type 1 (MmuPV1) model provides opportunities to study papillomavirus pathogenesis in cutaneous epithelium, the oral cavity, and the anogenital tract. However, to date the MmuPV1 infection model has not been used to demonstrate the effectiveness of potential antivirals. We previously reported that inhibitors of cellular MEK/ERK signaling suppress oncogenic HPV early gene expression in vitro . Herein, we adapted the MmuPV1 infection model to determine whether MEK inhibitors have anti-papillomavirus properties in vivo . We demonstrate that oral delivery of a MEK1/2 inhibitor promotes papilloma regression in immunodeficient mice that otherwise would have developed persistent infections. Quantitative histological analyses revealed that inhibition of MEK/ERK signaling reduces E6/E7 mRNAs, MmuPV1 DNA, and L1 protein expression within MmuPV1-induced lesions. These data suggest that MEK1/2 signaling is essential for both early and late MmuPV1 replication events supporting our previous findings with oncogenic HPVs. We also provide evidence that MEK inhibitors protect mice from developing secondary tumors. Thus, our data suggest that MEK inhibitors have potent anti-viral and anti-tumor properties in a preclinical mouse model and merit further investigation as papillomavirus antiviral therapies. Significance Statement: Persistent human papillomavirus (HPV) infections cause significant morbidity and oncogenic HPV infections can progress to anogenital and oropharyngeal cancers. Despite the availability of effective prophylactic HPV vaccines, millions of unvaccinated individuals, and those currently infected will develop HPV-related diseases over the next two decades and beyond. Thus, it remains critical to identify effective antivirals against papillomaviruses. Using a mouse papillomavirus model of HPV infection, this study reveals that cellular MEK1/2 signaling supports viral tumorigenesis. The MEK1/2 inhibitor, trametinib, demonstrates potent antiviral activities and promotes tumor regression. This work provides insight into the conserved regulation of papillomavirus gene expression by MEK1/2 signaling and reveals this cellular pathway as a promising therapeutic target for the treatment of papillomavirus diseases.

Humanized Patient-derived Xenograft Models of Disseminated Ovarian Cancer Recapitulate Key Aspects of the Tumor Immune Environment within the Peritoneal Cavity.

The importance of the immune microenvironment in ovarian cancer progression, metastasis, and response to therapies has become increasingly clear, especially with the new emphasis on immunotherapies. To leverage the power of patient-derived xenograft (PDX) models within a humanized immune microenvironment, three ovarian cancer PDXs were grown in humanized NBSGW (huNBSGW) mice engrafted with human CD34+ cord blood-derived hematopoietic stem cells. Analysis of cytokine levels in the ascites fluid and identification of infiltrating immune cells in the tumors demonstrated that these humanized PDX (huPDX) established an immune tumor microenvironment similar to what has been reported for patients with ovarian cancer. The lack of human myeloid cell differentiation has been a major setback for humanized mouse models, but our analysis shows that PDX engraftment increases the human myeloid population in the peripheral blood. Analysis of cytokines within the ascites fluid of huPDX revealed high levels of human M-CSF, a key myeloid differentiation factor as well as other elevated cytokines that have previously been identified in ovarian cancer patient ascites fluid including those involved in immune cell differentiation and recruitment. Human tumor-associated macrophages and tumor-infiltrating lymphocytes were detected within the tumors of humanized mice, demonstrating immune cell recruitment to tumors. Comparison of the three huPDX revealed certain differences in cytokine signatures and in the extent of immune cell recruitment. Our studies show that huNBSGW PDX models reconstitute important aspects of the ovarian cancer immune tumor microenvironment, which may recommend these models for preclinical therapeutic trials. Significance: huPDX models are ideal preclinical models for testing novel therapies. They reflect the genetic heterogeneity of the patient population, enhance human myeloid differentiation, and recruit immune cells to the tumor microenvironment.

A Deep Learning Convolutional Neural Network Can Differentiate Between Helicobacter Pylori Gastritis and Autoimmune Gastritis With Results Comparable to Gastrointestinal Pathologists.

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. OBJECTIVE.­: To determine whether a CNN can differentiate the following 2 gastric inflammatory patterns: autoimmune gastritis (AG) and HPG. 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. 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. CONCLUSIONS.­: A CNN can distinguish between cases of HPG and autoimmune gastritis with accuracy equal to GI pathologists.

Comparing Deep Learning and Immunohistochemistry in Determining the Site of Origin for Well-Differentiated Neuroendocrine Tumors.

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.

A Deep Learning Convolutional Neural Network Can Recognize Common Patterns of Injury in Gastric Pathology.

CONTEXT.­: Most deep learning (DL) studies have focused on neoplastic pathology, with the realm of inflammatory pathology remaining largely untouched. OBJECTIVE.­: To investigate the use of DL for nonneoplastic gastric biopsies. DESIGN.­: Gold standard diagnoses were blindly established by 2 gastrointestinal pathologists. For phase 1, 300 classic cases (100 normal, 100 Helicobacter pylori, 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. RESULTS.­: 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 H pylori (AUC = 100%), and 40% for reactive gastropathy (AUC = 99.9%). Sensitivity/specificity pairings were as follows: normal (96.7%, 86.7%), H pylori (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 H pylori, and 94.0% for reactive gastropathy. Sensitivity/specificity parings were as follows: normal (73.7%, 79.6%), H pylori (95.7%, 100%), reactive gastropathy (100%, 62.5%). CONCLUSIONS.­: A convolutional neural network can serve as an effective screening tool/diagnostic aid for H pylori gastritis.

High levels of tumor-associated neutrophils are associated with improved overall survival in patients with stage II colorectal cancer.

Conflicting reports regarding whether high tumor-associated neutrophils (TAN) are associated with outcomes in colorectal cancer (CRC) exist. Previous investigators have counted TAN using non-neutrophil-specific immunohistochemistry (IHC) stains. We examined whether TAN levels as determined by multi-field manual counting would predict prognosis. IRB approval was obtained and two pathologists, blinded to stage/outcome, counted TAN in 20 high power fields (HPF) per specimen. TAN score was defined as the mean of these counts. High TAN was defined as at or greater than the median score for that stage. Demographics, tumor characteristics, and overall survival (OS) were obtained from the records and examined for association with TAN score. IHC for arginase expression was performed in a subset of samples. 221 patients were included. Stage II patients with high TAN scores had an OS of 232 months as compared to those with low TAN (OS = 85 months, p = 0.03). The survival benefit persisted in multivariable analysis (HR 0.48, CI 0.25-0.91, p = 0.026) controlling for age and sex. Women had increased survival as compared to men, and there were no significant prognostic associations with TAN count in stage III/IV patients, although there were only 12 stage IV patients. Arginase staining did not provide additional information. Stage II colorectal cancer patients with high TAN live nearly 3 times longer than those with low TAN. Women with stage II disease and high TAN counts appear to be driving the survival benefit seen in the stage II patients and have increased overall survival in all stages.

Musashi-2 (MSI2) supports TGF-ß signaling and inhibits claudins to promote non-small cell lung cancer (NSCLC) metastasis.

Non-small cell lung cancer (NSCLC) has a 5-y survival rate of ∼16%, with most deaths associated with uncontrolled metastasis. We screened for stem cell identity-related genes preferentially expressed in a panel of cell lines with high versus low metastatic potential, derived from NSCLC tumors of Kras(LA1/+);P53(R172HΔG/+) (KP) mice. The Musashi-2 (MSI2) protein, a regulator of mRNA translation, was consistently elevated in metastasis-competent cell lines. MSI2 was overexpressed in 123 human NSCLC tumor specimens versus normal lung, whereas higher expression was associated with disease progression in an independent set of matched normal/primary tumor/lymph node specimens. Depletion of MSI2 in multiple independent metastatic murine and human NSCLC cell lines reduced invasion and metastatic potential, independent of an effect on proliferation. MSI2 depletion significantly induced expression of proteins associated with epithelial identity, including tight junction proteins [claudin 3 (CLDN3), claudin 5 (CLDN5), and claudin 7 (CLDN7)] and down-regulated direct translational targets associated with epithelial-mesenchymal transition, including the TGF-ß receptor 1 (TGFßR1), the small mothers against decapentaplegic homolog 3 (SMAD3), and the zinc finger proteins SNAI1 (SNAIL) and SNAI2 (SLUG). Overexpression of TGFßRI reversed the loss of invasion associated with MSI2 depletion, whereas overexpression of CLDN7 inhibited MSI2-dependent invasion. Unexpectedly, MSI2 depletion reduced E-cadherin expression, reflecting a mixed epithelial-mesenchymal phenotype. Based on this work, we propose that MSI2 provides essential support for TGFßR1/SMAD3 signaling and contributes to invasive adenocarcinoma of the lung and may serve as a predictive biomarker of NSCLC aggressiveness.