A typical presentation of primary pulmonary epithelioid sarcoma misdiagnosed as non-small cell lung cancer

Primary pulmonary epithelioid sarcoma: a case report

Background

Epithelioid sarcoma most frequently occurs in the dermal or subcutaneous area of the distal extremities. To date, there have been three cases of primary pulmonary epithelioid sarcoma reported. We report a case of epithelioid sarcoma that is considered a primary lung tumor.

Case presentation

A 65-year-old asymptomatic Asian male patient underwent chest radiography during a routine health examination, and an abnormal mass was detected. His past medical history was unremarkable. He smoked 40 cigarettes every day and had slightly obstructive impairment on spirometry. He worked as an employee of a company and had no history of asbestos exposure. He underwent partial resection of the right lung by thoracoscopy. A histological examination of the tumor revealed a cellular nodule of epithelioid and spindle-shaped cells. Some of the tumor cells displayed rhabdoid features and reticular arrangement in a myxomatous stroma. Immunohistochemically, the tumor cells were positive for vimentin, smooth muscle actin (SMA), CD34, and epithelial membrane antigen (EMA); loss of the BAF47/INI1 protein in the tumor cells was also confirmed. A diagnosis of epithelioid sarcoma was established. Careful screening by whole-body positron emission tomography for another primary lesion after surgery did not detect any possible lesion. He had no cutaneous disease.

Conclusion

To our knowledge, this is the fourth case of a proximal-type epithelioid sarcoma considered as a primary lung tumor.

DeePathology: Deep Multi-Task Learning for Inferring Molecular Pathology from Cancer Transcriptome

Despite great advances, molecular cancer pathology is often limited to the use of a small number of biomarkers rather than the whole transcriptome, partly due to computational challenges. Here, we introduce a novel architecture of Deep Neural Networks (DNNs) that is capable of simultaneous inference of various properties of biological samples, through multi-task and transfer learning. It encodes the whole transcription profile into a strikingly low-dimensional latent vector of size 8, and then recovers mRNA and miRNA expression profiles, tissue and disease type from this vector. This latent space is significantly better than the original gene expression profiles for discriminating samples based on their tissue and disease. We employed this architecture on mRNA transcription profiles of 10750 clinical samples from 34 classes (one healthy and 33 different types of cancer) from 27 tissues. Our method significantly outperforms prior works and classical machine learning approaches in predicting tissue-of-origin, normal or disease state and cancer type of each sample. For tissues with more than one type of cancer, it reaches 99.4% accuracy in identifying the correct cancer subtype. We also show this system is very robust against noise and missing values. Collectively, our results highlight applications of artificial intelligence in molecular cancer pathology and oncological research. DeePathology is freely available at https://github.com/SharifBioinf/DeePathology.