A low-cost machine learning framework for predicting drug-drug interactions based on fusion of multiple features and a parameter self-tuning strategy.

Poly-drug therapy is now recognized as a crucial treatment, and the analysis of drug-drug interactions (DDIs) offers substantial theoretical support and guidance for its implementation. Predicting potential DDIs using intelligent algorithms is an emerging approach in pharmacological research. However, the existing supervised models and deep learning-based techniques still have several limitations. This paper proposes a novel DDI analysis and prediction framework called the Multi-View Semi-supervised Graph-based (MVSG) framework, which provides a comprehensive judgment by integrating multiple DDI features and functions without any time-consuming training process. Unlike conventional approaches, MVSG can search for the most suitable similarity (or distance) measurement among DDI data and construct graph structures for each feature. By employing a parameter self-tuning strategy, MVSG fuses multiple graphs according to the contributions of features' information. The actual anticancer drug data are extracted from the authoritative public database for evaluating the effectiveness of our framework, including 904 drugs, 7730 DDI records and 19 types of drug interactions. Validation results indicate that the prediction is more accurate when multiple features are adopted by our framework. In comparison to conventional machine learning techniques, MVSG can achieve higher performance even with less labeled data and without a training process. Finally, MVSG is employed to narrow down the search for potential valuable combinations.

Comparative study of ultrasound-guided microwave ablation and traditional surgery in the treatment of plasma cell mastitis: a multicenter study.

Background: Plasma cell mastitis (PCM) is a benign breast disease that is refractory and difficult to cure. We integrated microwave ablation into the treatment of PCM to compare the clinical value of ultrasound-guided microwave ablation and traditional surgery in the treatment of PCM. Methods: A total of 68 patients with PCM who were admitted to 3 centers (Zhejiang Cancer Hospital, Hebei Province Hospital of Traditional Chinese Medicine, and Yantai Affiliated Hospital of Binzhou Medical university) from January 2017 to June 2019 were selected. All patients were diagnosed with PCM after pathological and clinical manifestations. Among these, 38 cases were treated with ultrasound-guided microwave ablation, and 30 cases were treated with traditional surgery. The operation time, hospitalization time, incision healing, intraoperative blood loss, postoperative pain degree, evaluation of breast shape effect, time taken for postoperative lesion disappearance, effective rate, and recurrence were recorded in the follow-up, and the clinical efficacy was compared and observed. Results: The effective rate of the ablation group was 86.8% (33/38), that of the operation group was 46.7% (14/30), and the difference was statistically significant (95% CI: 2.311-24.618; P<0.05). The average time of the lesion completely disappearing was 75.55±43.59 days in the ablation group and 103.87±45.98 days in the operation group, and the difference was statistically significant (P<0.05). The hospital stays, operation time, and intraoperative blood loss of patients in the ablation group were less than those of the operation group, and the difference was statistically significant (95% CI: -10.69 to -6.27, 95% CI: -77.06 to -51.26, and 95% CI: -21.54 to -13.64; P<0.05). The postoperative pain scoring, operative incision healing at 14 days after the operation, and breast appearance evaluation after treatment in the ablation group were better than those of the operation group, and the difference was statistically significant (P<0.05). Conclusions: Compared with traditional surgery, ultrasound-guided microwave ablation is a more effective treatment option for PCM.

[Preliminary Study on Detection of Circulating Tumor Cells in Lung Cancer by EGFR/Vimentin/Folic Acid Magnetic Sphere].

BACKGROUND: Circulating tumor cells (CTC) play an important role in the screening and prognosis of lung cancer, but the low efficiency and specificity of CTC isolation obviously restrict its clinical application. The purpose of this study is to explore a new and efficient isolation method of CTC in patients with non-small cell lung cancer (NSCLC) in order to achieve the purpose of early diagnosis of NSCLC. METHODS: Three kinds of immunolipid magnetic spheres of epidermal growth factor receptor (EGFR), vimentin and folic acid (FA) were prepared by thin film method. After characterization, the sorting scheme of cell line was explored, the optimal sorting scheme of NSCLC CTC was constructed, and its clinical application value was studied. RESULTS: The average capture efficiency of EGFR, Vimentin and FA magnetic spheres used alone and in combination to lung cancer cell lines was 78%, 79%, 82% and 91%, respectively. In 60 patients with lung cancer, using 2 CTC per 7.5 mL blood as cutoff value, the positive rates of EGFR, Vimentin and FA magnets used alone and in combination were 65.0%, 33.3%, 93.3% and 100%, respectively. It was also found that the number of CTC detected by combined use of the three magnetic spheres was correlated with clinical stages (P<0.05). CONCLUSIONS: The combination of three kinds of magnetic spheres can separate EGFR+, Vimentin+, FA+ expressed CTC, which is beneficial to the downstream analysis of CTC correlation. This study provides a new method to improve the efficiency of NSCLC CTC capture, and verifies that the captured CTC counting method can be used in the auxiliary diagnosis of lung cancer.

The construction of EpCAM/vimentin-PLGA/lipid immunomagnetic microspheres and the isolation of circulating tumor cells from lung cancer.

This study aims to investigate the isolation effect of the epithelial cell adhesion molecule (EpCAM) and the vimentin antibody PLGA/lipid magnetic microsphere on the isolation and identification of lung cancer circulating tumor cells (CTC) in the isolation and identification system of lung cancer CTC. The synthesis of the magnetic microsphere was achieved by the composite package of Fe3O4 magnetite nanoparticles with poly (lactic-co-glycolic acid) carboxylic acid endcap (PLGA-COOH) and octadecyl quaternized carboxymethyl chitosan (OQC), immunomagnetic microspheres (IMS) was prepared by linking an EpCAM antibody and a vimentin antibody. Blood samples of tumor-bearing nude mice (A549 lung cancer cells) were collected. Through a separation technique, the CTCs were captured by the EpCAM immunomagnetic microspheres (EpCAM-MS) and vimentin immunomagnetic microspheres (Vim-MS), and the cells were then counted and compared with the pathological condition of the tumor tissues. The results showed that self-prepared EpCAM-MS and Vim-MS could effectively capture lung cancer CTC and match the pathological findings.

MicroRNA-148b is a potential prognostic biomarker and predictor of response to radiotherapy in non-small-cell lung cancer.

miR-148b has been found to be aberrantly expressed in various tumor types. It has recently been reported to be involved in regulating radioresistance in non-small cell lung cancer (NSCLC) cells. However, its expression level and association with radiosensitivity in human patient samples have not been investigated. Real-time PCR was used to evaluate the expression levels of miR-148b. Χ (2) test was performed to analyze the association between miR-148b expression levels and clinicopathological factors or radiosensitivity. Kaplan-Meier survival curve was constructed to estimate the overall survival (OS), and the differences in survival were compared using the log-rank test. Cox regression analysis was conducted to determine the prognostic value of miR-148b. The relative level of miR-148b was significantly decreased in NSCLC tissues compared with matched non-cancerous tissues (P < 0.0001), and it was higher in tissues of patients who are good responders compared to those who are poor responders to radiotherapy (P < 0.0001). Lower expression of miR-148b was positively associated with high tumor stage (P = 0.0407) and radioresistance (P = 0.0002), and it predicted poor survival in patients with (P = 0.0129) or without (P = 0.0094) radiotherapy treatment. miR-148b was an independent prognostic factor for NSCLC as demonstrated by Cox proportional hazards risk analysis. miR-148b may serve as a prognostic biomarker of poor survival and a novel predictor of response to radiotherapy treatment in NSCLC.

MicroRNA-200a Targets EGFR and c-Met to Inhibit Migration, Invasion, and Gefitinib Resistance in Non-Small Cell Lung Cancer.

Lung cancer, especially non-small cell lung cancer (NSCLC), is the major cause of cancer death worldwide. Mutations in epidermal growth factor receptor (EGFR) and hepatocyte growth factor receptor (c-Met), both of which are receptor tyrosine kinases, have been identified in a considerable percentage of NSCLC patients. EGFR and c-Met share the same downstream pathways and cooperate not only in promoting metastasis but also in conferring resistance to tyrosine kinase inhibitor (TKI) therapies in NSCLC. MicroRNAs (miRNAs) are a family of small non-coding RNAs, usually 21-25 nucleotides long, and are critical in regulating gene expression. Abnormal miRNA expression has been implicated in the initiation and progression in many forms of cancers, including lung cancer. In this study, we found that miR-200a is downregulated in NSCLC cells, where it directly targets the 3'-UTR of both EGFR and c-Met mRNA. Overexpression of miR-200a in NSCLC cells significantly downregulates both EGFR and c-Met levels and severely inhibits cell migration and invasion. Moreover, in NSCLC cell lines that are resistant to gefitinib, a drug often used in TKI therapies to treat NSCLC, miR-200a expression is able to render the cells much more sensitive to the drug treatment.

Endothelial PAS domain-containing protein 1 confers TKI-resistance by mediating EGFR and MET pathways in non-small cell lung cancer cells.

Mutations in epidermal growth factor receptor (EGFR) rendering it constitutively active is one of the major causes for metastatic non-small-cell lung cancer (NSCLC), and EGFR-targeted therapies utilizing tyrosine kinase inhibitors (TKIs) are often used clinically as the first-line treatment. But approximately half of NSCLC patients develop resistance to these therapies, where the MET proto-oncogene is amplified by EGFR through the hypoxia-inducible factor (HIF)-1α. Here we report that endothelial PAS domain-containing protein 1 (EPAS1), with 48% sequence identity to HIF-1α, specifically binds to TKI-resistant T790M EGFR, but not to wild-type EGFR, in NSCLC cell lines. Expression of EPAS1 enhances amplification of MET when simultaneously expressed with T790M EGFR but not with wild-type EGFR, and this enhancement is independent of ligand binding domain of EGFR. MET amplification requires EPAS1, since EPAS1 knock-down reduced MET levels. When NSCLC cells expressing T790M EGFR were treated with TKIs, reduced EPAS1 levels significantly enhanced the drug effect, whereas over-expression of EPAS1 increased the drug resistant effect. This EPAS1-dependent TKI-resistance was abolished by knocking-down MET, suggesting that EPAS1 does not cause TKI-resistance itself but functions to bridge EGFR and MET interactions. Our findings suggest that EPAS1 is a key factor in the EGFR-MET crosstalk in conferring TKI-resistance in NSCLC cases, and could be used as a potential therapeutic target in TKI-resistant NSCLC patients.