No abstract

Background Accurately predicting the survival rate of breast cancer patients is a major issue for cancer researchers. Machine learning (ML) has attracted much attention with the hope that it could provide accurate results, but its modeling methods and prediction performance remain controversial. The aim of this systematic review is to identify and critically appraise current studies regarding the application of ML in predicting the 5-year survival rate of breast cancer. Methods In accordance with the PRISMA guidelines, two researchers independently searched the PubMed (including MEDLINE), Embase, and Web of Science Core databases from inception to November 30, 2020. The search terms included breast neoplasms, survival, machine learning, and specific algorithm names. The included studies related to the use of ML to build a breast cancer survival prediction model and model performance that can be measured with the value of said verification results. The excluded studies in which the modeling process were not explained clearly and had incomplete information. The extracted information included literature information, database information, data preparation and modeling process information, model construction and performance evaluation information, and candidate predictor information. Results Thirty-one studies that met the inclusion criteria were included, most of which were published after 2013. The most frequently used ML methods were decision trees (19 studies, 61.3%), artificial neural networks (18 studies, 58.1%), support vector machines (16 studies, 51.6%), and ensemble learning (10 studies, 32.3%). The median sample size was 37256 (range 200 to 659820) patients, and the median predictor was 16 (range 3 to 625). The accuracy of 29 studies ranged from 0.510 to 0.971. The sensitivity of 25 studies ranged from 0.037 to 1. The specificity of 24 studies ranged from 0.008 to 0.993. The AUC of 20 studies ranged from 0.500 to 0.972. The precision of 6 studies ranged from 0.549 to 1. All of the models were internally validated, and only one was externally validated. Conclusions Overall, compared with traditional statistical methods, the performance of ML models does not necessarily show any improvement, and this area of research still faces limitations related to a lack of data preprocessing steps, the excessive differences of sample feature selection, and issues related to validation. Further optimization of the performance of the proposed model is also needed in the future, which requires more standardization and subsequent validation.

Objective The sphingolipid de novo synthesis pathway is considered a promising target for pharmacological intervention in atherosclerosis. However, its potential is hampered by the fact that the substance’s atherogenic mechanism is not completely understood. To unravel the complex mechanisms, we utilized the sphingomyelin synthase 2 (Sms2) gene knockout approach to test our hypothesis that selectively decreasing plasma lipoprotein SM, can play an important role in preventing atherosclerosis. Methods and Results We prepared Sms2 and Apoe double knockout (KO) mice. They showed a significant decrease in plasma lipoprotein SM levels (35%, P<0.01) and a significant increase in ceramide and dihydroceramide levels (87.5 and 27%, P<0.01, respectively), but no significant changes in other tested sphingolipids, cholesterol, and triglyceride. Non-HDL lipoproteins from the double KO mice showed a reduction of SM but not cholesterol and displayed a less tendency toward aortic sphingomyelinase-mediated lipoprotein aggregation in vitro and retention in aortas in vivo, compared to controls. More important, at age 19 weeks, Sms2 KO/Apoe KO mice showed a significant reduction in atherosclerotic lesions of the aortic arch and root (52%, P<0.01), compared to controls. We also found that the Sms2 KO/Apoe KO brachiocephalic artery (BCA) contained significantly less SM, ceramide, free cholesterol, and cholesteryl ester (35, 32, 58, and 60%, P<0.01, respectively), than that of Apoe KO BCA. Conclusions Decreasing plasma SM levels through decreasing SMS2 activity could become a promising treatment for atherosclerosis.

Fine needle aspiration (FNA) is increasingly being supplanted by core needle biopsy. However, breast surgeons continue to rely on FNA at our institution. This retrospective study evaluated breast FNA for its diagnostic accuracy and breast cancer biomarker testing utility. All breast FNAs performed at Massachusetts General Hospital 2009-2015 were reviewed. Cytology diagnoses were compared with subsequent tissue or clinical diagnoses. Immunohistochemistry and fluorescence in situ hybridization (FISH) results using formalin-fixed paraffin-embedded (FFPE) cell blocks and histologic tissue blocks were compared. 1654 consecutive breast FNAs were included. Breast FNA demonstrated the following diagnostic performance: positive predictive value of malignant cytology diagnosis 100 %, negative predictive value of benign cytology diagnosis 97.5 %, complete sensitivity 91.6 %, and specificity 95.5 %. Concordance rates for estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 (HER2) immunohistochemistry, and HER2 FISH were 98.2 % (κ = 0.95, p < 0.001), 100.0 % (κ = 1.000, p < 0.001), 83.1 % (κ = 0.69, p < 0.001), and 93.5 % (κ = 0.785, p < 0.001), respectively. Review of consecutive breast FNAs in a large cohort confirmed the excellent accuracy of this biopsy technique for breast lesion diagnosis. FNA FFPE cell blocks collected in the course of routine clinical care are adequate, practical, and reliable for breast cancer biomarker testing.

Objective Sphingomyelin Sphingolipid de novo biosynthesis is related with nonalcoholic fatty liver disease (NAFLD) or hepatic steatosis. However, the mechanism is still unclear. Sphingomyelin synthase (SMS), utilizing ceramide as one of the substrates to produce sphingomyelin, sits at the crossroads of sphingolipid biosynthesis. SMS has two isoforms: SMS1 and SMS2. SMS2 is the major isoform in the liver. Approach and Results To investigate the relationship between liver SMS2 activity-mediated sphingolipid changes and hepatic steatosis, we utilized two mouse models, SMS2 liver-specific transgenic (LTg) and SMS2 knockout (KO) mice. We found that SMS2LTg livers have lower ceramide and higher sphingomyelin, while SMS2 KO livers have higher ceramide and lower sphingomyelin. We also found that liver SMS2 overexpression promoted fatty acid uptaking and liver steatosis, while SMS2 deficiency had an opposite effect, in comparison with their respective controls. Importantly, the exogenous ceramide supplementation to Huh7 cells, a human hepatoma cell line, reduced the expression of PPAR γ2 and its target genes, CD36 and FSP27. PPAR γ reporter analysis confirmed this phenomenon. Moreover, PPARγ antagonist treatment significantly decreased triglyceride accumulation in SMS2LTg liver. Conclusions We attributed these effects to ceramide which can suppress of PPARγ2, thus reducing expression of CD36 and FSP27, and reducing liver steatosis. After all, SMS2 inhibition in the liver could diminish liver steatosis.

Sphingomyelin synthase 1 (SMS1) and SMS2 are two isoforms of SMS, the last enzyme for sphingomyelin (SM) biosynthesis. To evaluate the role of SMS in vivo in terms of plasma lipoprotein metabolism, we generated recombinant adenovirus vectors containing human SMS1 cDNA (AdV-SMS1), SMS2 cDNA (AdV-SMS2), or the reporter LacZ cDNA (AdV-LacZ) as a control. On day 7 after intravenous infusion of 2 3 10 11 particles of both AdV-SMS1 and AdV-SMS2 into mice, liver SMS1 and SMS2 mRNA levels as well as SMS activity were significantly increased (2.5-, 2.7-, 2.1-, and 2.3-fold, respectively; P , 0.001). Lipoprotein analysis indicated that AdV-SMS1 and AdV-SMS2 treatment caused no changes of total SM and cholesterol levels but significantly decreased HDL-SM and HDL-cholesterol (42% and 38%, and 27% and 25%, respectively; P , 0.05). It also significantly increased non-HDL-SM and non-HDLcholesterol levels (50% and 35%, and 64% and 61%, respectively; P , 0.05) compared with AdV-LacZ controls. SDS--PAGE showed a significant increase in apolipoprotein B (apoB; P , 0.01) but no changes in apoA-I levels. Moreover, we found that non-HDL from both AdV-SMS1-and AdV-SMS2-treated mice was significantly aggregated after treatment with a mammalian sphingomyelinase, whereas lipoproteins from control animals did not aggregate. To investigate the mechanism of HDL changes, we measured liver scavenger receptor class B type I (SR-BI) levels by Western blot. We found that AdV-SMS1 and AdV-SMS2 mouse liver homogenates contained 50% and 55% higher SR-BI levels than in controls, whereas no change was observed in hepatic ABCA1 levels. An HDL turnover study revealed an increase of plasma clearance rates for ]HDL in both AdV-SMS1 and AdV-SMS2 mice compared with controls. In conclusion, adenovirus-mediated SMS1 and SMS2 overexpression increased lipoprotein atherogenic potential. Such an effect may contribute to the increased plasma SM levels observed in animal models of atherosclerosis and in human patients with coronary artery disease.-Dong, J., J.

No abstract

High infiltration of M2-polarized macrophages in the primary tumor indicates unfavorable prognosis and poor overall survival in the patients with triple-negative breast cancer (TNBC). Thus, reversing M2-polarized tumor-associated macrophages in the tumors has been considered as a potential therapeutic strategy for TNBC. Sphingomyelin synthase 2 (SMS2) is the key enzyme for sphingomyelin production, which plays an important role in plasma membrane integrity and function. In this study we investigated whether SMS2 inhibitor or SMS2 gene knockout could reduce macrophages M2 polarization and tumor progression in a mouse model of TNBC. We showed that SMS2 mRNA expression was linked to immunosuppressive tumor microenvironment and poor prognosis in TNBC patients. The knockout of SMS2 or application of 15w (a specific SMS2 inhibitor) markedly decreased the generation of M2-type macrophages in vitro, and reduced the tumor weight and lung metastatic niche formation in a 4T1-TNBC mouse model. We further demonstrated that the in vivo antitumor efficacy of 15w was accompanied by a multifaceted remodeling of tumor immune environment reflecting not only the suppression of M2-type macrophages but also diminished levels of regulatory T cells and myeloid-derived suppressor cells leading to a dramatically improved infiltration of antitumor CD8 + T lymphocytes. Collectively, our results reveal a novel and important role of SMS2 in the protumorigenic function and may offer a new strategy for macrophage-targeted anticancer therapy.