An artificial intelligence prediction model outperforms conventional guidelines in predicting lymph node metastasis of T1 colorectal cancer.

Background: According to guidelines, a lot of patients with T1 colorectal cancers (CRCs) undergo additional surgery with lymph node dissection after being treated by endoscopic resection (ER) despite the low incidence of lymph node metastasis (LNM). Aim: The aim of this study was to develop an artificial intelligence (AI) model to more effectively identify T1 CRCs at risk for LNM and reduce the rate of unnecessary additional surgery. Methods: We retrospectively analyzed 651 patients with T1 CRCs. The patient cohort was randomly divided into a training set (546 patients) and a test set (105 patients) (ratio 5:1), and a classification and regression tree (CART) algorithm was trained on the training set to develop a predictive AI model for LNM. The model used 12 clinicopathological factors to predict positivity or negativity for LNM. To compare the performance of the AI model with the conventional guidelines, the test set was evaluated according to the Japanese Society for Cancer of the Colon and Rectum (JSCCR) and National Comprehensive Cancer Network (NCCN) guidelines. Finally, we tested the performance of the AI model using the test set and compared it with the JSCCR and NCCN guidelines. Results: The AI model had better predictive performance (AUC=0.960) than the JSCCR (AUC=0.588) and NCCN guidelines (AUC=0.850). The specificity (85.8% vs. 17.5%, p<0.001), balanced accuracy (92.9% vs. 58.7%, p=0.001), and the positive predictive value (36.3% vs. 9.0%, p=0.001) of the AI model were significantly better than those of the JSCCR guidelines and reduced the percentage of the high-risk group for LNM from 83.8% (JSCCR) to 20.9%. The specificity of the AI model was higher than that of the NCCN guidelines (85.8% vs. 82.4%, p=0.557), but there was no significant difference between the two. The sensitivity of the NCCN guidelines was lower than that of our AI model (87.5% vs. 100%, p=0.301), and according to the NCCN guidelines, 1.2% of the 105 test set patients had missed diagnoses. Conclusion: The AI model has better performance than conventional guidelines for predicting LNM in T1 CRCs and therefore could significantly reduce unnecessary additional surgery.

Nonsebaceous Lymphadenoma of the Salivary Glands: A Potential Overdiagnosis Pitfall.

Nonsebaceous lymphadenoma is a rare benign salivary tumor. It is easily misdiagnosed as lymphoepithelial carcinoma, leading to overtreatment. Some patients experience sequelae after undergoing cervical lymph node resection and adjuvant treatment, so it is critical to distinguish these entities. We describe the histopathological and immunohistochemical features of this rare entity in 3 cases and discuss the differential diagnosis and histogenesis. Nonsebaceous lymphadenoma can be distinguished from lymphoepithelial carcinoma by the following histological features: There is a lymph node-like form at low magnification, with prominent proliferating epithelial nests but no destructive growth pattern; variable numbers of tubuloglandular components are always seen in proliferating epithelial nests, which transition to cystically dilated salivary ducts; no lesion necrosis exists; and mitotic figures are absent or rare. No patients experienced recurrence during the 8- to 69-month (mean, 29 months) follow-up.

Unusual morphologic features of low-grade endometrial stromal sarcoma: A case report.

BACKGROUND: Endometrial stromal tumours are uncommon tumours of the uterus. They mainly occur in perimenopausal women. Tumours with typical clinicopathological features do not usually pose diagnostic problems. However, rare clinicopathological features can occur, and clinicians without significant experience may have difficulty diagnosing these tumours and managing these patients. METHODS: Herein, we report a case of endometrial stromal sarcoma that occurred in a 25-year-old woman. The pathological features, immunophenotype, treatment and prognosis were discussed. RESULTS: The tumour revealed morphological heterogeneity, and there were similar proliferative-type endometrial stromal cells, an extensive amount of mature adipose tissue, and prominent rhabdomyoblastic and smooth muscle cells. Histopathological and immunohistochemical studies confirmed low-grade endometrial stromal sarcoma with smooth muscle, adipocytic and rhabdomyoblastic differentiation (approximately 60% were differentiated tissues). The final treatment of the tumour was total abdominal hysterectomy with bilateral salpingo-oophorectomy. There was no evidence of recurrence for 109 months postoperatively. CONCLUSIONS: We found that low-grade endometrial stromal tumours with extensive adipocytic and prominent rhabdomyoblastic differentiation are misdiagnosed because they are infrequent. They must be differentiated from rhabdomyosarcoma with accurate identification of adipocytes, and long-term follow-up is needed.

Malignant Transformation of Metaplastic Thymoma into High-Grade Sarcomatoid Carcinoma: A Case Report.

The correlation of histogenesis between metaplastic thymoma and thymic sarcomatoid carcinoma is unclear. We report a case of metaplastic thymoma transformed into high-grade sarcomatoid carcinoma. A 64 × 54 × 32 mm anterior mediastinal mass in a 61-year-old woman microscopically consisted mainly of classic metaplastic thymoma, with the center dominated by high-grade sarcomatoid carcinoma. In some areas, both epithelial and spindle cell components of the metaplastic thymoma showed increased cellular atypia, mitotic activity, and focal necrosis and gradually transformed into the polygonal/pleomorphic and spindle cell components of sarcomatoid carcinoma. Immunohistochemically, the characteristics of the polygonal/pleomorphic sarcomatoid cells were similar to those of the epithelial component of metaplastic thymoma, while the spindle sarcomatoid cells were more similar to the spindle cells component of metaplastic thymoma. The Ki-67 index was less than 5% in the metaplastic thymoma areas but up to 70% in the sarcomatoid carcinoma area. Radical operation and postoperative radiotherapy were performed. Multifocal relapses at the pleura occurred 13 months after surgery.

GATA3 is a useful immunohistochemical marker for distinguishing sarcomatoid malignant mesothelioma from lung sarcomatoid carcinoma and organizing pleuritis.

Sarcomatoid malignant mesothelioma (SMM) tends to occur in the pleura and is morphologically similar to lung sarcomatoid carcinoma (LSC) and organizing pleuritis (OP). Because SMM often does not express mesothelial markers, it is very difficult to distinguish from LSC and OP. GATA-binding protein 3 (GATA3) is a specific immunohistochemical (IHC) marker of breast and urothelial carcinoma. We routinely find that GATA is expressed in MM; however, GATA3 expression in SMM and its reference value for distinguishing SMM from LSC and OP remain unclear. Here, we used IHC methods to detect the expression of GATA3 and classic mesothelial markers in 17 SMM, 12 LSC, and 7 OP cases. We detected the following expression rates in SMM versus LSC cases: GATA3 (70.6% vs. 16.7%, p = 0.008), calretinin (52.9% vs. 8.3%, p = 0.019), Wilms tumor (WT)-1 (64.7% vs. 0%, p = 0.000), D2-40 (47.1% vs. 16.7%, p = 0.126), CK5/6 (35.3% vs. 25.0%, p = 0.694), and pan-cytokeratin (CKpan) (88.2% vs. 100.0%, p = 0.498). The specificities of calretinin, WT-1, and GATA3 in distinguishing SMM from LSC were 91.7%, 100%, and 83.3%, respectively, and combinations of any two of these three markers exhibited 100% specificity for SMM. Notably, the sensitivity of calretinin+/WT1+ staining for SMM was only 23.5%, which increased to 64.7% after including GATA3. Furthermore, all OP cases showed partial or diffuse expression of CKpan, WT-1, and D2-40 but no GATA3 and calretinin expression. In conclusion, GATA3 is an IHC marker with excellent sensitivity and specificity for SMM, and the combined consideration of GATA3, calretinin, and WT-1 was best for distinguishing SMM from LSC. Moreover, CKpan, WT-1, and D2-40 had no value for distinguishing SMM from OP, and GATA3 and calretinin were the most specific markers for distinguishing these two lesions.