Artificial intelligence in hepatocellular carcinoma diagnosis: a comprehensive review of current literature.

BACKGROUND AND AIM: Hepatocellular carcinoma (HCC) diagnosis mainly relies on its pathognomonic radiological profile, obviating the need for biopsy. The project of incorporating artificial intelligence (AI) techniques in HCC aims to improve the performance of image recognition. Herein, we thoroughly analyze and evaluate proposed AI models in the field of HCC diagnosis. METHODS: A comprehensive review of the literature was performed utilizing MEDLINE/PubMed and Web of Science databases with the end of search date being the 30th of September 2023. The MESH terms "Artificial Intelligence," "Liver Cancer," "Hepatocellular Carcinoma," "Machine Learning," and "Deep Learning" were searched in the title and/or abstract. All references of the obtained articles were also evaluated for any additional information. RESULTS: Our search resulted in 183 studies meeting our inclusion criteria. Across all diagnostic modalities, reported area under the curve (AUC) of most developed models surpassed 0.900. A B-mode US and a contrast-enhanced US model achieved AUCs of 0.947 and 0.957, respectively. Regarding the more challenging task of HCC diagnosis, a 2021 deep learning model, trained with CT scans, classified hepatic malignant lesions with an AUC of 0.986. Finally, a MRI machine learning model developed in 2021 displayed an AUC of 0.975 when differentiating small HCCs from benign lesions, while another MRI-based model achieved HCC diagnosis with an AUC of 0.970. CONCLUSIONS: AI tools may lead to significant improvement in diagnostic management of HCC. Many models fared better or comparable to experienced radiologists while proving capable of elevating radiologists' accuracy, demonstrating promising results for AI implementation in HCC-related diagnostic tasks.

Expression profile of CYP1A1 and CYP1B1 enzymes in endometrial tumors.

The cytochrome P450 CYP1A1 and CYP1B1 enzymes are phase I extrahepatic enzymes involved in the activation of pro-carcinogenic compounds to carcinogenic metabolites. Although differential overexpression of CYP1A1 and CYP1B1 has been documented at the messenger RNA (mRNA) and protein level, studies that have examined CYP1 expression by enzyme activity assays are limited. In the current study, the expression of CYP1A1 and CYP1B1 was investigated in a panel of human tumors of endometrial origin by quantitative reverse transcriptase PCR (qRT-PCR), Western blotting, and enzyme activity assays. The data revealed that approximately 36 % (5/14) and 43 % (6/14) of the endometrial tumors overexpressed CYP1A1 and CYP1B1 mRNA, whereas in 57 % of the endometrial tumors, CYP1 mRNA levels were downregulated. The mean mRNA levels of CYP1B1 and CYP1A1 in endometrial tumors did not show a significant difference compared to normal tissues (p > 0.05). Western blotting confirmed the qRT-PCR results and CYP1A1 and CYP1B1 proteins were shown to be downregulated in 7/14 (50 %) of the tumors and overexpressed in 4/14 (29 %) of the tumors. As regards to enzyme activity, 21 % (3/14) of the endometrial samples revealed elevated CYP1 activity levels across the tumor counterparts. Overall, the data suggest a putative downregulation of CYP1A1 and CYP1B1 expression in endometrial tumors, whereas overexpression of active CYP1 enzymes in 21 % of the tumors highlights the potential use of the latter enzymes as chemotherapeutic targets in endometrial cancer.

All you need to know about gastrinoma today | Gastrinoma and Zollinger-Ellison syndrome: A thorough update.

Zollinger-Ellison syndrome (ZES) is a distinct syndrome characterized by hyperchlorhydria-induced peptic ulcer disease and chronic diarrhea. It is the result of a gastrin-excess state caused by a duodenal or pancreatic neuroendocrine tumor referred to as gastrinoma. This gastrin-secreting neuroendocrine tumor is usually sporadic in nature, or part of multiple endocrine neoplasia type 1 syndrome. The high rate of malignancy associated with gastrinomas substantiates the need for early diagnosis. In order to diagnose ZES with laboratory tests, patients under antacid medication are required to stay off proton pump inhibitors for at least one week and H2 receptor antagonists for 48 h. Fasting serum gastrin level measurement serves as an initial and fundamental diagnostic test, boasting a sensitivity of 99%. Gastrinoma patients will present with a gastrin level greater than 100 pg/mL, while a serum gastrin level higher than 1000 pg/mL, in the presence of gastric pH <2, is considered diagnostic. Since more common causes of hypergastrinemia exist in the setting of hypochlorhydria, ruling those out should precede ZES consideration. Such causes include atrophic gastritis, Helicobacter pylori (H. pylori)-associated pangastritis, renal failure, vagotomy, gastric outlet obstruction and retained antrum syndrome. The secretin stimulation test and the calcium gluconate injection test represent classic adjuvant diagnostic techniques, while alternative approaches are currently being introduced and evaluated. Specifically, the secretin stimulation test aids in differentiating ZES cases from other hypergastrinemic states. Its principle is based on secretin stimulation of gastrinoma cells to secrete gastrin, while inhibiting normal G cells. The rapid intravenous infusion of 4 µg/kg secretin over 1 min is followed by gastrin level evaluation at specific intervals post-infusion. Localization of the primary tumor and its metastases is the next diagnostic step when gastrinoma-associated ZES is either suspected or biochemically confirmed. Endoscopic ultrasound has showcased sensitivity as high as 83% for pancreatic gastrinomas and is considered the primary modality in such cases, although its tumor detection rates are substantially lower in duodenal lesions. Gallium-68 radiotracers, especially DOTATOC with positron emission tomography, are currently setting the standard in tumor localization, enhancing traditional imaging techniques and showcasing high sensitivity and specificity. Although gastrinomas have been reported in various anatomic locations, the vast majority arise in a specific site named the "gastrinoma triangle", involving parts of the duodenum, pancreas and extra-hepatic biliary system. Proton pump inhibitors serve as the cornerstone of symptomatic ZES treatment. Surgery is routinely performed in localized sporadic ZES, irrespective of imaging results. ZES in multiple endocrine neoplasia type 1 requires work-up for evaluation and treatment of hyperparathyroidism, while surgery might be an option for selected cases. In cases of advanced and metastatic disease, there is a variety of potential treatments, ranging for somatostatin analogs to chemotherapeutic drugs, liver-directed therapies and liver transplantation, while neither hepatic metastases, nor locally invasive disease necessarily preclude surgical management. This article thoroughly and critically reviews available literature and provides an extensive and multidimensional overview of ZES, along with current controversies regarding management of this disease.

Digging into the NGS Information from a Large-Scale South European Population with Metastatic/Unresectable Pancreatic Ductal Adenocarcinoma: A Real-World Genomic Depiction.

Despite ongoing oncological advances, pancreatic ductal adenocarcinoma (PDAC) continues to have an extremely poor prognosis with limited targeted and immunotherapeutic options. Its genomic background has not been fully characterized yet in large-scale populations all over the world. Methods: Replicating a recent study from China, we collected tissue samples from consecutive Greek patients with pathologically-confirmed metastatic/unresectable PDAC and retrospectively investigated their genomic landscape using next generation sequencing (NGS). Findings: From a cohort of 409 patients, NGS analysis was successfully achieved in 400 cases (56.50% males, median age: 61.8 years). Consistent with a previous study, KRAS was the most frequently mutated gene in 81.50% of tested samples, followed by TP53 (50.75%), CDKN2 (8%), and SMAD4 (7.50%). BRCA1/2 variants with on-label indications were detected in 2%, and 87.50% carried a variant associated with off-label treatment (KRAS, ERBB2, STK11, or HRR-genes), while 3.5% of the alterations had unknown/preliminary-studied actionability (TP53/CDKN2A). Most of HRR-alterations were in intermediate- and low-risk genes (CHEK2, RAD50, RAD51, ATM, FANCA, FANCL, FANCC, BAP1), with controversial actionability: 8% harbored a somatic non-BRCA1/2 alteration, 6 cases had a high-risk alteration (PALB2, RAD51C), and one co-presented a PALB2/BRCA2 alteration. Elevated LOH was associated with HRR-mutated status and TP53 mutations while lowered LOH was associated with KRAS alterations. Including TMB/MSI data, the potential benefit from an NGS-oriented treatment was increased from 1.91% to 13.74% (high-MSI: 0.3%, TMB > 10 muts/MB: 12.78%). TMB was slightly increased in females (4.75 vs. 4.46 muts/MB) and in individuals with age > 60 (4.77 vs. 4.40 muts/MB). About 28.41% showed PD-L1 > 1% either in tumor or immune cells, 15.75% expressed PD-L1 ≥ 10%, and only 1.18% had PD-L1 ≥ 50%. This is the largest depiction of real-world genomic characteristics of European patients with PDAC, which offers some useful clinical and research insights.

Modified docetaxel-cisplatin in combination with capecitabine as first-line treatment in metastatic gastric cancer. a phase II study.

The combination of docetaxel, cisplatin and fluorouracil is considered to be one of the reference regimens for advanced gastric cancer, but due to its major myelotoxicity, its use in clinical practice has become limited. This prospective phase II study evaluated the activity and toxicity of a modified regimen with lower doses of docetaxel and cisplatin combined with oral capecitabine instead of fluorouracil for patients with advanced gastric cancer. Treatment consisted of docetaxel at 60 mg/m(2) i.v. followed by cisplatin at 60 mg/m(2), both administered on day one, every three weeks. Capecitabine at 2 g/m(2) per day was administered in two divided doses for 14 days (days 2-15). Thirty six patients were enrolled in the study. The median age was 64 years and performance status (ECOG) was 0-1. All patients had advanced disease, 78% with liver metastases, 100% with intra-abdominal lymph node metastases and 67% with peritoneal implants. Out of the 36 patients, 13 had undergone gastric resection, 13 had received adjuvant chemotherapy with irinotecan-leucovorin-fluorouracil, while seven patients had undergone adjuvant radiotherapy. The remaining 23 patients presented with advanced inoperable disease. Among 36 evaluable for response cases, there were 16 (44.4%) (Confidence Internal (CI) 95%=28-60%), partial responses. Stable disease was recorded in 12 (33.3%), resulting in an overall disease control rate of 78% (CI 95%=69-87%), while 8 (22.3%) patients progressed on chemotherapy. The median response duration was 6 (range=3-8) months. The median time-to-progression was 5 (range=3-6) months and the median survival (after the administration of a second-line chemotherapy in 12 patients), was 12 (range=5-24) months. Myelotoxocity was the main toxicity, with grade 3-4 neutropenia occurring in 18 (50%) and febrile neutropenia in six (16%) patients. Granulocyte-Colony Stimulating Factor (G-CSF) support was given to 16 (44.4%) patients, while grade 3 thrombocytopenia was recorded in two (6%). In conclusion, this modified regimen of docetaxel-cisplatin-capecitabine appears to have comparable efficacy with that reported for the reference regimen, with acceptable toxicity when G-CSF support is provided. However, because due to the small size of the study, further investigation is warranted.