We aimed to investigate the feasibility of machine learning (ML) algorithm to distinguish pseudoprogression (PsPD) from progression (PD) in patients with glioblastoma (GBM). We recruited the patients diagnosed as primary GBM who received gross total resection (GTR) and concurrent chemoradiotherapy in two institutions from April 2010 to April 2017 and presented suspicious contrast-enhanced lesion on brain magnetic resonance imaging (MRI) during follow-up. Patients from two institutions were allocated to training (N = 59) and testing (N = 19) datasets, respectively. We developed a convolutional neural network combined with a long short-term memory ML structure. MRI data, which was 9 axial post-contrast T1-weighted images in our study, and clinical features were incorporated (Model 1). In the testing set, the trained Model 1 resulted in AUC of 0.83, AUPRC of 0.87, and F1-score of 0.74 using optimal threshold. The performance was superior to that of Model 2 (CNN-LSTM model with MRI data alone) and Model 3 (random forest model with clinical feature alone). The developed algorithm involving MRI data and clinical features could help making decision during follow-up of patients with GBM treated with GTR and concurrent CCRT.

Histone modifications and DNA methylation are two major factors in epigenetic phenomenon. Unlike the histone deacetylase inhibitors, which are known to exert radiosensitizing effects, there have only been a few studies thus far concerning the role of DNA methyltransferase (DNMT) inhibitors as radiosensitizers. The principal objective of this study was to evaluate the effects of DNMT inhibitors on the radiosensitivity of human cancer cell lines, and to elucidate the mechanisms relevant to that process.

Background: Despite aggressive treatment with radiation therapy and concurrent adjuvant temozolomide (TMZ), glioblastoma multiform (GBM) still has a dismal prognosis. We aimed to identify strategies to improve the therapeutic outcome of combined radiotherapy and TMZ in GBM by targeting pro-survival signaling from the epidermal growth factor receptor (EGFR). Methods: Glioma cell lines U251, T98G were used. Colony formation, DNA damage repair, mode of cell death, invasion, migration and vasculogenic mimicry as well as protein expression were determined. Results: U251 cells showing a low level of methyl guanine transferase (MGMT) were highly responsive to the radiosensitizing effect of TMZ compared to T98G cells having a high level of MGMT. Treatment with a dual inhibitor of Class I PI3K/mTOR, PI103; a HSP90 inhibitor, 17-DMAG; or a HDAC inhibitor, LBH589, further increased the cytotoxic effect of radiation therapy plus TMZ in U251 cells than in T98G cells. However, treatment with a mTOR inhibitor, rapamycin, did not discernibly potentiate the radiosensitizing effect of TMZ in either cell line. The mechanism of enhanced radiosensitizing effects of TMZ was multifactorial, involving impaired DNA damage repair, induction of autophagy or apoptosis, and reversion of EMT (epithelial-mesenchymal transition). Conclusions: Our results suggest possible strategies for counteracting the pro-survival signaling from EGFR to improve the therapeutic outcome of combined radiotherapy and TMZ for high-grade gliomas.

Introduction: Gene amplification is an important mechanism for activating oncogenes in malignant tumors. Although amplification of HER2, C-MYC, CCND1 and FGFR1 has been reported in breast cancers, their role in the progression of in situ to invasive breast carcinoma is unclear. To investigate this question we compared the amplification frequencies of these genes in pure ductal carcinoma in situ (DCIS), DCIS associated with invasive carcinoma, and invasive carcinoma. Methods: We performed fluorescence in situ hybridization of the selected genes on tissue microarrays composed of 179 pure DCIS and 438 invasive carcinomas. Two hundred and sixteen of the latter had DCIS components, and in those cases we compared gene amplification in the intraductal and invasive components of each carcinoma. Results: The rate of amplification of FGFR1 was higher in invasive carcinomas than in the pure DCIS, but the opposite was true for HER2 amplification. These findings applied consistently to high-grade tumors, but not to low/intermediate-grade tumors. The amplification status of HER2, C-MYC, CCND1 and FGFR1 was generally similar in the matched invasive and DCIS components of the same tumors. However, FGFR1 amplification was more common in the invasive components than in the DCIS components. In survival analyses, FGFR1 amplification was found to be an independent prognostic factor for poor disease-free survival for all patients with invasive carcinoma and for the hormone receptor-positive subgroup. Conclusion: Amplification of HER2, C-MYC and CCND1 seems to play a role in the early development of breast cancer, but not in its progression. However, the increased frequency of FGFR1 amplification in invasive carcinomas compared with pure DCIS and in the invasive components of individual tumors, and its association with decreased disease-free survival, suggests a role for FGFR1 amplification in the progression of breast cancer including in situ-to-invasive transition, as well as initiation.