EGFR and BRAF mutations in inverted sinonasal papilloma - a more complex landscape?

Prediction of anticancer drug sensitivity using an interpretable model guided by deep learning

BACKGROUND

The prediction of drug sensitivity plays a crucial role in improving the therapeutic effect of drugs. However, testing the effectiveness of drugs is challenging due to the complex mechanism of drug reactions and the lack of interpretability in most machine learning and deep learning methods. Therefore, it is imperative to establish an interpretable model that receives various cell line and drug feature data to learn drug response mechanisms and achieve stable predictions between available datasets.

RESULTS

This study proposes a new and interpretable deep learning model, DrugGene, which integrates gene expression, gene mutation, gene copy number variation of cancer cells, and chemical characteristics of anticancer drugs to predict their sensitivity. This model comprises two different branches of neural networks, where the first involves a hierarchical structure of biological subsystems that uses the biological processes of human cells to form a visual neural network (VNN) and an interpretable deep neural network for human cancer cells. DrugGene receives genotype input from the cell line and detects changes in the subsystem states. We also employ a traditional artificial neural network (ANN) to capture the chemical structural features of drugs. DrugGene generates final drug response predictions by combining VNN and ANN and integrating their outputs into a fully connected layer. The experimental results using drug sensitivity data extracted from the Cancer Drug Sensitivity Genome Database and the Cancer Treatment Response Portal v2 reveal that the proposed model is better than existing prediction methods. Therefore, our model achieves higher accuracy, learns the reaction mechanisms between anticancer drugs and cell lines from various features, and interprets the model's predicted results.

CONCLUSIONS

Our method utilizes biological pathways to construct neural networks, which can use genotypes to monitor changes in the state of network subsystems, thereby interpreting the prediction results in the model and achieving satisfactory prediction accuracy. This will help explore new directions in cancer treatment. More available code resources can be downloaded for free from GitHub ( https://github.com/pangweixiong/DrugGene ).

Frequent EGFR exon 20 insertion in the so-called peripheral-type squamous cell neoplasm of uncertain malignant potential: a variant of bronchiolar adenoma or under-recognised entity?

INTRODUCTION

Herein we describe a series of rare peripheral pulmonary neoplasms temporarily termed "peripheral type squamous cell neoplasm of uncertain malignant potential (PSCN-UMP)" and investigate their relationship to bronchiolar adenoma (BA) and squamous cell carcinoma (SCC).

MATERIALS AND METHODS

The histologic and immunohistochemical features of 10 PSCN-UMPs and six BAs were compared. Whole exome sequencing (WES) and bioinformatics analysis were performed to further compare the genetic features of PSCN-UMPs, BAs, and NSCLCs.

RESULTS

All PSCN-UMPs were peripherally located and histologically characterised by the lepidic, nested, and papillary proliferation of relatively bland squamous cells, accompanied by entrapped hyperplastic reactive pneumocytes. The basal squamous cells coexpressed TTF1 and squamous markers. Both cellular components exhibited bland morphology and a low proliferative activity. The six BAs met the morphologic and immunophenotypic features of proximal-type BA. Genetically, driver mutations, including frequent EGFR exon 20 insertions, were found in PSCN-UMPs, while the KRAS mutation, BRAF mutation, and ERC1::RET fusion were detected in BAs. PSCN-UMPs also shared some alterations with BAs in mutational signatures, while copy number variants (CNV) were enriched in MET and NKX2-1 in PSCN-UMP and MCL1, MECOM, SGK1, and PRKAR1A in BA.

CONCLUSION

PSCN-UMPs exhibited the proliferation of bland squamous cells accompanied by entrapped pneumocytes and frequent EGFR exon 20 insertions, which showed distinct features from BAs and SCCs. Recognition of this specific entity will help to expand the morphologic and molecular spectrum of peripheral lung squamous neoplasms.

Etiology of sinonasal inverted papilloma: An update

Objective

Sinonasal inverted papilloma (IP) and its clinical features have been widely studied, but there are few studies delving into its etiology and risk factors. A narrative review was conducted to summarize a contemporary understanding of the potential etiologies of IP, including immunologic/inflammatory, viral, genetic, and environmental causes.

Study Design

Review.

Methods

A MEDLINE search was conducted through August 11, 2021, focusing on studies investigating the etiology and risk factors for sinonasal IP and its malignant transformation.

Results

High‐ and low‐risk human papillomavirus have been connected with the formation of IP, but conflicting evidence exists regarding their role. Occupational and industrial exposures may also contribute to IP formation, while smoking may increase the odds of malignant progression. Exon 20 mutations in EGFR are an active area of research in IP with mixed evidence. Finally, several cell cycle and angiogenic factors such as Ki67, VEGF, and Akt/mTOR have been implicated in the development and progression of IP.

Conclusion

There continues to be conflicting evidence around the development of IP, but significant progress has been made in recent years. Further study is needed for all these potential etiologies to elucidate risk factors and therapeutic strategies.