Orbital apocrine hidrocystoma. Report of two cases.

INTRODUCTION: We report the clinical features and the management of two cases of orbital hidrocystoma in the setting of an enlarging orbital mass. CASES DESCRIPTION: A 48-year-old man presented with a mass in the right upper medial orbital quadrant, firmly attached to the supraorbital incisure. A 70-year-old man had a well demarcated lesion in the upper lateral orbital quadrant adherent to the lacrimal gland. There was no history of previous orbital trauma. In both cases histopathology confirmed a diagnosis of apocrine hidrocystoma. Following surgery, the first patient complained of mild hypoesthesia in the territory of the supraorbital nerve that resolved spontaneously within 3 weeks. Surgery was uneventful in the other patient. No recurrence was seen during the follow up. CONCLUSIONS: Apocrine hidrocystomas have been rarely described in the orbit, but should be considered in the differential diagnosis of orbital cystic masses. Recurrence is rare following complete surgical excision.

AI and High-Grade Glioma for Diagnosis and Outcome Prediction: Do All Machine Learning Models Perform Equally Well?

Radiomic models outperform clinical data for outcome prediction in high-grade gliomas (HGG). However, lack of parameter standardization limits clinical applications. Many machine learning (ML) radiomic models employ single classifiers rather than ensemble learning, which is known to boost performance, and comparative analyses are lacking in the literature. We aimed to compare ML classifiers to predict clinically relevant tasks for HGG: overall survival (OS), isocitrate dehydrogenase (IDH) mutation, O-6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation, epidermal growth factor receptor vIII (EGFR) amplification, and Ki-67 expression, based on radiomic features from conventional and advanced magnetic resonance imaging (MRI). Our objective was to identify the best algorithm for each task. One hundred fifty-six adult patients with pathologic diagnosis of HGG were included. Three tumoral regions were manually segmented: contrast-enhancing tumor, necrosis, and non-enhancing tumor. Radiomic features were extracted with a custom version of Pyradiomics and selected through Boruta algorithm. A Grid Search algorithm was applied when computing ten times K-fold cross-validation (K=10) to get the highest mean and lowest spread of accuracy. Model performance was assessed as AUC-ROC curve mean values with 95% confidence intervals (CI). Extreme Gradient Boosting (xGB) obtained highest accuracy for OS (74,5%), Adaboost (AB) for IDH mutation (87.5%), MGMT methylation (70,8%), Ki-67 expression (86%), and EGFR amplification (81%). Ensemble classifiers showed the best performance across tasks. High-scoring radiomic features shed light on possible correlations between MRI and tumor histology.

Histological growth patterns and molecular analysis of resected colorectal lung metastases.

Lung is the site of metastasis in about 15-25 % of colorectal cancer (CRC) patients. Lung metastasectomy of CRC represents a standard therapy in patients with resectable metastases. In this study we investigated both histological patterns of metastases and mutations in MAPkinase pathway genes and their relationship to prognosis. The study included 74 patients that underwent metastasectomy of colorectal lung metastasis (CLM). In patients that underwent surgical resection of more than one metastasis in the same operation the largest was chosen. In patients that had undergone multiple lung metastasectomy only the sample from the first metastasectomy was included. Histologically metastases were scored according to amount and distribution of necrosis and fibrosis and three patterns were identified: "pattern A", metastasis with extensive, confluent central necrosis surrounded by a rim of neoplastic glands; "pattern B", metastasis characterized by a proliferation of neoplastic glands in a dense stroma with focal necrosis mainly intraglandular; "pattern C", metastasis with a mixed A and B morphology. In all samples direct sequencing of exon 2 of KRAS and NRAS genes and exon 15 of BRAF genes was carried out.Histological patterns weren't related to metastasis size or other clinical features however pattern C metastases showed a significant worst disease free survival (DFS). KRAS mutations were observed in 39 % of patients. Mutations in KRAS codon 13 resulted significantly associated with synchronous metastasis and poor prognosis. No mutations were identified in exon 2 NRAS gene whilst 1.4 % harboured a mutation in BRAF. To our knowledge this is the first study that investigates in a large series of CLM histological growth patterns, molecular alterations and their relationship to prognosis. Our data suggest a prognostic role in CLM of KRAS specific mutations and histopathological patterns.

Deep Learning Can Differentiate IDH-Mutant from IDH-Wild GBM.

Isocitrate dehydrogenase (IDH) mutant and wildtype glioblastoma multiforme (GBM) often show overlapping features on magnetic resonance imaging (MRI), representing a diagnostic challenge. Deep learning showed promising results for IDH identification in mixed low/high grade glioma populations; however, a GBM-specific model is still lacking in the literature. Our aim was to develop a GBM-tailored deep-learning model for IDH prediction by applying convoluted neural networks (CNN) on multiparametric MRI. We selected 100 adult patients with pathologically demonstrated WHO grade IV gliomas and IDH testing. MRI sequences included: MPRAGE, T1, T2, FLAIR, rCBV and ADC. The model consisted of a 4-block 2D CNN, applied to each MRI sequence. Probability of IDH mutation was obtained from the last dense layer of a softmax activation function. Model performance was evaluated in the test cohort considering categorical cross-entropy loss (CCEL) and accuracy. Calculated performance was: rCBV (accuracy 83%, CCEL 0.64), T1 (accuracy 77%, CCEL 1.4), FLAIR (accuracy 77%, CCEL 1.98), T2 (accuracy 67%, CCEL 2.41), MPRAGE (accuracy 66%, CCEL 2.55). Lower performance was achieved on ADC maps. We present a GBM-specific deep-learning model for IDH mutation prediction, with a maximal accuracy of 83% on rCBV maps. Highest predictivity achieved on perfusion images possibly reflects the known link between IDH and neoangiogenesis through the hypoxia inducible factor.

Gastroesophageal reflux symptoms and microscopic esophagitis in a cohort of consecutive patients affected by atrophic body gastritis: a pilot study.

AIM: In patients affected by atrophic body gastritis (ABG) gastro-oesophageal reflux (GER) related symptoms have been reported, despite the presence of hypochlorhydria. OBJECTIVE: Objectives of this single-centre study was to assess in ABG the occurrence of GER-related symptoms and their relationship with histopathologic oesophageal findings. MATERIALS AND METHODS: Fifty-four consecutive patients (20.4%male, 57.6 ± 14 years) undergoing to follow-up for ABG, underwent assessment of GER-related symptoms and gastroscopy with multiple gastric and oesophageal biopsies to investigate the presence of microscopic esophagitis (ME). RESULTS: At least one typical GER symptoms were reported in 24.1% with 9.2% of patients complaining of heartburn and 18.5% regurgitation. One or more atypical GERD symptoms were reported in 44.4% of patients. Two symptomatic ABG patients presented oesophageal lesions at endoscopy (one with erosive esophagitis (LA-C) and one with Barrett's oesophagus (C2M2)), 49% reported a mild ME and 24.5% a severe ME. No significant differences regarding GERD prevalence were found among patients with or without ME, but cough was the only symptom significantly more frequent in patients with ME (38.95% vs. 7.7%, p = .042). CONCLUSIONS: These data showed that GERD is present in a quarter of ABG patients, suggesting that hypochlorhydria not exclude per se arising of oesophageal symptoms. In ABG we found that ME is a frequent finding but its clinical relevance remains to be investigated with further studies.