Dataset Information

Combining multimodal imaging and treatment features improves machine learning-based prognostic assessment in patients with glioblastoma multiforme.

ABSTRACT:

Background

For Glioblastoma (GBM), various prognostic nomograms have been proposed. This study aims to evaluate machine learning models to predict patients' overall survival (OS) and progression-free survival (PFS) on the basis of clinical, pathological, semantic MRI-based, and FET-PET/CT-derived information. Finally, the value of adding treatment features was evaluated.

Methods

One hundred and eighty-nine patients were retrospectively analyzed. We assessed clinical, pathological, and treatment information. The VASARI set of semantic imaging features was determined on MRIs. Metabolic information was retained from preoperative FET-PET/CT images. We generated multiple random survival forest prediction models on a patient training set and performed internal validation. Single feature class models were created including "clinical," "pathological," "MRI-based," and "FET-PET/CT-based" models, as well as combinations. Treatment features were combined with all other features.

Results

Of all single feature class models, the MRI-based model had the highest prediction performance on the validation set for OS (C-index: 0.61 [95% confidence interval: 0.51-0.72]) and PFS (C-index: 0.61 [0.50-0.72]). The combination of all features did increase performance above all single feature class models up to C-indices of 0.70 (0.59-0.84) and 0.68 (0.57-0.78) for OS and PFS, respectively. Adding treatment information further increased prognostic performance up to C-indices of 0.73 (0.62-0.84) and 0.71 (0.60-0.81) on the validation set for OS and PFS, respectively, allowing significant stratification of patient groups for OS.

Conclusions

MRI-based features were the most relevant feature class for prognostic assessment. Combining clinical, pathological, and imaging information increased predictive power for OS and PFS. A further increase was achieved by adding treatment features.

SUBMITTER: Peeken JC

PROVIDER: S-EPMC6346243 | biostudies-literature | 2019 Jan

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Combining multimodal imaging and treatment features improves machine learning-based prognostic assessment in patients with glioblastoma multiforme.

Peeken Jan C JC Goldberg Tatyana T Pyka Thomas T Bernhofer Michael M Wiestler Benedikt B Kessel Kerstin A KA Tafti Pouya D PD Nüsslin Fridtjof F Braun Andreas E AE Zimmer Claus C Rost Burkhard B Combs Stephanie E SE

Cancer medicine 20181218 1

<h4>Background</h4>For Glioblastoma (GBM), various prognostic nomograms have been proposed. This study aims to evaluate machine learning models to predict patients' overall survival (OS) and progression-free survival (PFS) on the basis of clinical, pathological, semantic MRI-based, and FET-PET/CT-derived information. Finally, the value of adding treatment features was evaluated.<h4>Methods</h4>One hundred and eighty-nine patients were retrospectively analyzed. We assessed clinical, pathological, ...[more]

PMID: 30561851

Similar Datasets

Project description:The synthesis of self-assembled peptide nanoparticles using a facile one-pot synthesis approach is gaining increasing attention, allowing therapy in combination with diagnosis. Their drawback is limited diagnostic potential, which can be improved after necessary modifications and efficacious functionalization. Herein, a cyclic heptapeptide having the Arg-Gly-Asp-Lys-Leu-Ala-Lys sequence was modified by conjugation of the ε-amino group of the terminal lysine residue with diethylenetriamine pentaacetic acid (DTPA) as a bifunctional chelating agent (BFC) for radiolabeling with a γ-emitting radionuclide (99mTc, half-life 6.01 h; energy 140 keV). Further, the free amino group of the middle lysine residue was successfully conjugated with near-infrared fluorescence (NIRF) dye Cyanine5.5 N-succinimidyl ester (Ex/Em = 670/701 nm) by a co-assembly method to form newly designed novel NIRF dye conjugated self-assembled peptide-DTPA (Cy5.5@SAPD) nanoparticles. The fluorescent nanoparticle formation was confirmed by using a fluorescence spectrophotometer (Ex/Em = 650/701 nm), and the transmission electron microscope (TEM) images showed a size of ∼ 40 nm with a lattice fringe distance of 0.294 nm. Cytotoxicity and confocal laser scanning microscopy (CLSM) studies showed that these nanoparticles possess a high affinity for the αvβ3-integrin receptor overexpressed on brain tumor glioblastoma with an EC50 = 20 μM. Moreover, these nanoparticles were observed to have potential to internalize into U87MG cells more prominently than HEK-293 cancer cells and induce apoptosis. The apoptosis assay showed 79.5% apoptotic cells after 24 h treatment of Cy5.5@SAPD nanoparticles. Additionally, these nanoparticles were also radiolabeled with 99mTc for the single photon emission computed tomography (SPECT) imaging study in tumor-bearing female Balb/c mice. The excellent imaging feature of Cy5.5@SAPD-99mTc nanoparticles as a multimodal (SPECT/NIRF) diagnostic probe, as well as noteworthy therapeutic potential was observed. The results suggested that our newly designed novel dual-targeting dual-imaging nanoparticles may serve as an admirable theranostic probe to treat brain tumor glioblastoma multiforme.

Dataset Information

Combining multimodal imaging and treatment features improves machine learning-based prognostic assessment in patients with glioblastoma multiforme.

Background

Methods

Results

Conclusions

Publications

Combining multimodal imaging and treatment features improves machine learning-based prognostic assessment in patients with glioblastoma multiforme.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets