GEAMP, a novel gastroesophageal junction carcinoma cell line derived from a malignant pleural effusion

Itraconazole Exerts Its Antitumor Effect in Esophageal Cancer By Suppressing the HER2/AKT Signaling Pathway

Itraconazole, an FDA-approved antifungal, has antitumor activity against a variety of cancers. We sought to determine the effects of itraconazole on esophageal cancer and elucidate its mechanism of action. Itraconazole inhibited cell proliferation and induced G₁-phase cell-cycle arrest in esophageal squamous cell carcinoma and adenocarcinoma cell lines. Using an unbiased kinase array, we found that itraconazole downregulated protein kinase AKT phosphorylation in OE33 esophageal adenocarcinoma cells. Itraconazole also decreased phosphorylation of downstream ribosomal protein S6, transcriptional expression of the upstream receptor tyrosine kinase HER2, and phosphorylation of upstream PI3K in esophageal cancer cells. Lapatinib, a tyrosine kinase inhibitor that targets HER2, and siRNA-mediated knockdown of HER2 similarly suppressed cancer cell growth in vitro. Itraconazole significantly inhibited growth of OE33-derived flank xenografts in mice with detectable levels of itraconazole and its primary metabolite, hydroxyitraconazole, in esophagi and tumors. HER2 total protein and phosphorylation of AKT and S6 proteins were decreased in xenografts from itraconazole-treated mice compared to xenografts from placebo-treated mice. In an early phase I clinical trial (NCT02749513) in patients with esophageal cancer, itraconazole decreased HER2 total protein expression and phosphorylation of AKT and S6 proteins in tumors. These data demonstrate that itraconazole has potent antitumor properties in esophageal cancer, partially through blockade of HER2/AKT signaling.

Interpretable deep learning uncovers cellular properties in label-free live cell images that are predictive of highly metastatic melanoma

Deep learning has emerged as the technique of choice for identifying hidden patterns in cell imaging data but is often criticized as "black box." Here, we employ a generative neural network in combination with supervised machine learning to classify patient-derived melanoma xenografts as "efficient" or "inefficient" metastatic, validate predictions regarding melanoma cell lines with unknown metastatic efficiency in mouse xenografts, and use the network to generate in silico cell images that amplify the critical predictive cell properties. These exaggerated images unveiled pseudopodial extensions and increased light scattering as hallmark properties of metastatic cells. We validated this interpretation using live cells spontaneously transitioning between states indicative of low and high metastatic efficiency. This study illustrates how the application of artificial intelligence can support the identification of cellular properties that are predictive of complex phenotypes and integrated cell functions but are too subtle to be identified in the raw imagery by a human expert. A record of this paper's transparent peer review process is included in the supplemental information. VIDEO ABSTRACT.