Search

Primary malignant cardiac tumors rarely occur, and cardiac synovial sarcoma (SS) is especially rare among such tumors. Herein, we present the case of a 35-year-old female with primary cardiac SS treated with surgery, chemotherapy, and radiotherapy. She presented with chest symptoms and underwent imaging examinations. A cardiac tumor was suspected, and an open biopsy was performed. The pathological findings suggested cardiac SS. Next, we performed a resection, and the tumors persisted at a macroscopic level. Immunohistochemistry was negative for SS18-SSX and positive for the SSX C-terminus and cytokeratin CAM5.2, a reduction of SMARCB1/INI1 was observed, and fluorescence in situ hybridization showed positive SS18 split staining. Owing to the FNCLCC grade 3 tumor and R2 margins, adjuvant chemotherapy with ifosfamide, doxorubicin, and radiotherapy was initiated, and the patient was diagnosed with cardiac SS. The differences in patients with cardiac SS compared with general SS include male predominance, larger tumor size, and poorer prognosis. Pathological findings of immunohistochemistry and fluorescence in situ hybridization were found to be more reliable than imaging findings for a correct diagnosis. Additionally, because incomplete resection is frequently performed, adjuvant therapy, including chemotherapy and radiation therapy, may be performed. The findings indicate that multiple therapies, including surgery, chemotherapy, and radiotherapy, are essential treatment strategies for improving the prognosis of patients with cardiac SS.

Abstract Background Herein, for the first time, we present a case with mixed invasive micropapillary and neuroendocrine mammary neoplasm. Case The patient, a 65‐year‐old postmenopausal woman, had become aware of a tumor in her right breast 11 months prior to presentation at our hospital. The cut surface of the mastectomy specimen contained a well‐circumscribed, multinodular, red‐brown tumor, measuring 15x15x15 cm. Histopathologically, this solid cystic lesion consisted of medullary growth of cancer cells accompanied by a well‐developed vascular network as well as conspicuous hemorrhage. Cancer cell nests of various sizes displayed an “inside‐out” structure surrounded by empty spaces. Most cancer cells were polygonal, though a few were short fusiform‐shaped, and possessed finely granular, eosinophilic cytoplasm and ovoid, fine‐granular nuclei. Eighteen mitotic figures were observed in 10 high‐power fields. Macrometastases, up to 13x8 mm in size, with the same morphological features as the original tumor site, were identified in 3 of 15 dissected right axillary nodes. Immunohistochemically, primary and metastatic cancer cells were diffusely positive for chromogranin A and the estrogen receptor (Allred's total score: 8) and focally reactive for synaptophysin and the progesterone receptor (total score: 5). HER2 and cytokeratin 5/6 were negative, and the MIB‐1 labelling index was 36.2%. MUC1 and EMA lined the stroma‐facing surfaces of the cell membranes, indicating reversed polarity. Conclusion Our current patient, who had an invasive breast carcinoma with concomitant neuroendocrine and micropapillary features, developed multiple nodal metastases in association with a large‐diameter tumor showing a luminal B‐like immuno‐profile. Accordingly, meticulous clinical follow‐up remains essential for this uncommon case.

Purpose: The purpose of this study was to develop and evaluate lung cancer segmentation with a pretrained model and transfer learning. The pretrained model was constructed from an artificial dataset generated using a generative adversarial network (GAN).Materials and Methods: Three public datasets containing images of lung nodules/lung cancers were used: LUNA16 dataset, Decathlon lung dataset, and NSCLC radiogenomics. The LUNA16 dataset was used to generate an artificial dataset for lung cancer segmentation with the help of the GAN and 3D graph cut. Pretrained models were then constructed from the artificial dataset. Subsequently, the main segmentation model was constructed from the pretrained models and the Decathlon lung dataset. Finally, the NSCLC radiogenomics dataset was used to evaluate the main segmentation model. The Dice similarity coefficient (DSC) was used as a metric to evaluate the segmentation performance.Results: The mean DSC for the NSCLC radiogenomics dataset improved overall when using the pretrained models. At maximum, the mean DSC was 0.09 higher with the pretrained model than that without it.Conclusion: The proposed method comprising an artificial dataset and a pretrained model can improve lung cancer segmentation as confirmed in terms of the DSC metric. Moreover, the construction of the artificial dataset for the segmentation using the GAN and 3D graph cut was found to be feasible.