Integrating deep transfer learning and radiomics features in glioblastoma multiforme patient survival prediction

2020 ◽  
Author(s):  
Wei Han ◽  
Lei Qin ◽  
Camden Bay ◽  
Xin Chen ◽  
Kun-Hsing Yu ◽  
...  
Keyword(s):  
Glioblastoma Multiforme ◽  
Transfer Learning ◽  
Patient Survival ◽  
Survival Prediction

scholarly journals A Novel Six-mRNA Signature Predicts Survival of Patients With Glioblastoma Multiforme

2021 ◽  
Vol 12 ◽  
Author(s):  
Zhentao Liu ◽  
Hao Zhang ◽  
Hongkang Hu ◽  
Zheng Cai ◽  
Chengyin Lu ◽  
...  
Keyword(s):  
Glioblastoma Multiforme ◽  
Cox Regression ◽  
Methylation Status ◽  
Survival Rates ◽  
Gene Expression Omnibus ◽  
The Cancer Genome Atlas ◽  
Survival Prediction ◽  
Lasso Regression ◽  
Cox Regression Analysis ◽  
Score Model

Glioblastoma multiforme (GBM) is a devastating brain tumor and displays divergent clinical outcomes due to its high degree of heterogeneity. Reliable prognostic biomarkers are urgently needed for improving risk stratification and survival prediction. In this study, we analyzed genome-wide mRNA profiles in GBM patients derived from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases to identify mRNA-based signatures for GBM prognosis with survival analysis. Univariate Cox regression model was used to evaluate the relationship between the expression of mRNA and the prognosis of patients with GBM. We established a risk score model that consisted of six mRNA (AACS, STEAP1, STEAP2, G6PC3, FKBP9, and LOXL1) by the LASSO regression method. The six-mRNA signature could divide patients into a high-risk and a low-risk group with significantly different survival rates in training and test sets. Multivariate Cox regression analysis confirmed that it was an independent prognostic factor in GBM patients, and it has a superior predictive power as compared with age, IDH mutation status, MGMT, and G-CIMP methylation status. By combining this signature and clinical risk factors, a nomogram can be established to predict 1-, 2-, and 3-year OS in GBM patients with relatively high accuracy.

scholarly journals Explainable Transformer-Based Neural Network forthe Prediction of Survival Outcomes in Non-SmallCell Lung Cancer (NSCLC)

2021 ◽  
Author(s):  
Gustavo Arango ◽  
Elly Kipkogei ◽  
Etai Jacob ◽  
Ioannis Kagiampakis ◽  
Arijit Patra
Keyword(s):  
Immune System ◽  
Precision Medicine ◽  
Cancer Patients ◽  
Patient Survival ◽  
Response To Therapy ◽  
Survival Models ◽  
Attractive Alternative ◽  
Survival Prediction ◽  
Molecular Features ◽  
Clinical Measurements

In this paper, we introduce the Clinical Transformer - a recasting of the widely used transformer architecture as a method for precision medicine to model relations between molecular and clinical measurements, and the survival of cancer patients. Although the emergence of immunotherapy offers a new hope for cancer patients with dramatic and durable responses having been reported, only a subset of patients demonstrate benefit. Such treatments do not directly target the tumor but recruit the patient immune system to fight the disease. Therefore, the response to therapy is more complicated to understand as it is affected by the patients physical condition, immune system fitness and the tumor. As in text, where the semantics of a word is dependent on the context of the sentence it belongs to, in immuno-therapy a biomarker may have limited meaning if measured independent of other clinical or molecular features. Hence, we hypothesize that the transformer-inspired model may potentially enable effective modelling of the semantics of different biomarkers with respect to patient survival time. Herein, we demonstrate that this approach can offer an attractive alternative to the survival models utilized incurrent practices as follows: (1) We formulate an embedding strategy applied to molecular and clinical data obtained from the patients. (2) We propose a customized objective function to predict patient survival. (3) We show the applicability of our proposed method to bioinformatics and precision medicine. Applying the clinical transformer to several immuno-oncology clinical studies, we demonstrate how the clinical transformer outperforms other linear and non-linear methods used in current practice for survival prediction. We also show that when initializing the weights of a domain-specific transformer by the weights of a cross-domain transformer, we further improve the predictions. Lastly, we show how the attention mechanism successfully captures some of the known biology behind these therapies

Intra-carotid chemo (ICC) followed by radiation (RT) with concomitant temozolamide (TMZ) and subsequent maintenance TMZ therapy in patients (pts) with glioblastoma multiforme (GBM)

2006 ◽  
Vol 24 (18_suppl) ◽  
pp. 1554-1554 ◽  
Author(s):  
G. Mohin ◽  
R. Davis ◽  
A. Meek ◽  
A. Rosiello ◽  
C. Roque ◽  
...  
Keyword(s):  
Glioblastoma Multiforme ◽  
Standard Treatment ◽  
Patient Survival ◽  
Malignant Tumors ◽  
University Hospital ◽  
Time To Progression ◽  
Grade 3 ◽  
Year 2000 ◽  
Experienced Grade ◽  
Intracarotid Chemotherapy

1554 Background: Glioblastoma multiforme is one of the most resistant malignant tumors. Surgery provides only temporary, palliative relief. Radiation therapy affords additional, though short, benefit. Systemic, intravenous and/or oral chemotherapy with nitrosoureas and TMZ, increases survival only slightly. Median survival (MS) after all these modalities incorporated remains unimpressive, less than 14 months. Intracarotid chemotherapy with cisplatin and VP-16, a standard treatment at Stony Brook University Hospital since 1999, prior to RT, results in MS of 20 months. A year later we added TMZ as a part of standard treatment with or without ICC in patients with GBM. Provided here is a retrospective analysis of the treatment consisting of surgery, ICC, RT with concomitant TMZ followed by 2 years of TMZ maintenance. Methods: Fifteen patients with pathologically confirmed, newly diagnosed GBM were treated from year 2000 to 2005; 11 men; median age 53 (range 25-68); 80% with PS over 70. They underwent a surgical procedure (5 near total) followed by ICC with cisplatin 60 mg/m2 and VP-16 40 mg/m2 every 3 weeks for total of 3 cycles. Subsequently they received TMZ 75 mg/m2/day orally, concomitantly with RT 6120–6300 cGY, followed a month later by maintenance therapy with TMZ 200 mg/m2/ daily for 5 days out of each month for 2 years or until progression. Results: Fourteen pts.are being evaluated (one too early), all have survived at least 12 months.One-year PFS is 79%. Median time to progression is 19.5 months; MS is 25 months with a range of 12-48 months. Toxicity is limited to nausea and vomiting mainly grade 1-2; only 1 patient experienced grade 3 vomiting with subsequent TMZ maintenance dose reduction by about 10%. Conclusions: Surgery followed by ICC, then combination of RT + TMZ followed by maintenance TMZ therapy seems to be more effective in improving patient survival than surgery, ICC and RT as reported previously by us in pts with GBM (Cancer 2000; 10:2350–6). This treatment has been well tolerated. No significant financial relationships to disclose.

Role of IκBα as a negative regulator of EGFR and a molecular determinant of prognosis in glioblastoma multiforme

2009 ◽  
Vol 27 (15_suppl) ◽  
pp. 2028-2028
Author(s):  
M. Bredel ◽  
J. Renfrow ◽  
A. Yadav ◽  
A. Alvarez ◽  
D. Lin ◽  
...  
Keyword(s):  
Glioblastoma Multiforme ◽  
Patient Survival ◽  
Alkylating Agents ◽  
The United States ◽  
Multidimensional Analysis ◽  
Negative Regulator ◽  
Egfr Signaling ◽  
Glioblastoma Cells ◽  
Molecular Determinant ◽  
Functional Analyses

2028 Background: Glioblastoma multiforme is a complex disease that involves the deregulation of overlapping signaling pathways. Constitutive activation of the transcription factor nuclear factor-κB (NF-κB) has been broadly associated with various human cancers, including glioblastomas, and their therapy resistance and may be due to cross-coupling with other oncogenic pathways, such as epidermal growth factor signaling. Methods: Multidimensional analysis involving gene and transcript data for the endogenous NF-κB modulator IκBα/NFKBIA and clinical patient profiles of 482 glioblastomas/high-grade gliomas from multiple institutions in the United States and The Cancer Genome Atlas Pilot Project. Functional analyses using LN229, U87, and U118 glioblastoma cells, and human embryonic kidney 293T cells with transgene phenotypes for IκBα. IκBα promoter and coding sequence and promoter methylation analyses in a resistance model of 15 glioblastomas cell lines with in vitro and/or in vivo resistance to O6-alkylating agents. Results: We have identified a regulatory circuit between NF-κB and EGFR signaling in glioblastomas, where IκBα binds to EGFR and attenuates EGFR signaling by immobilizing its kinase domain into an inactive conformation. We found the NFKBIA gene at 14q13.2 deleted in 25% of glioblastomas and its occurrence mutually exclusive with EGFR amplification. Loss of NFKBIA associates independently with patient survival. Functional analyses uncover a bona fide tumor suppressor role for IκBα in glioblastoma cells, where it functions to constrain tumorigenic and migratory potential, and induce spontaneous cellular senescence, and apoptosis in response to treatment. IκBα expression is an independent predictor of patient prognosis in multiple glioblastoma populations. Glioblastomas with initially high IκBα expression significantly repress IκBα upon tumor recurrence, suggesting an acquired mechanism to evade its tumor-suppressive and/or chemo-sensitizing effects during tumor progression. Conclusions: IκBα is a molecular determinant of biological tumor behavior and patient survival in glioblastoma multiforme. Deletion of NFKBIA could present an alternate mechanism to activate EGFR signaling in EGFR non-amplified glioblastomas. No significant financial relationships to disclose.

Sign in / Sign up

Export Citation Format

Share Document