Date: Thu, 28 Mar 2024 13:45:49 -0500 (CDT) Message-ID: <421095156.42539.1711651549488@tcia-wiki-rh-1.ad.uams.edu> Subject: Exported From Confluence MIME-Version: 1.0 Content-Type: multipart/related; boundary="----=_Part_42538_924867614.1711651549488" ------=_Part_42538_924867614.1711651549488 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Content-Location: file:///C:/exported.html
To date, no study has attempted to integrate imaging biomarkers and tumo= r gene expression into a statistical model that could constitute a more rob= ust predictor of patient outcome than either biomarkers or gene expression = alone. his study explored whether such a model could allow reliable predict= ion of patient survival and time to tumor recurrence based on a combination= of magnetic resonance imaging (MRI) features and tumor gene expression.
The study aimed to incorporate glioblastoma imaging features and genomic= biomarkers into statistical models in order to reliably predict patient ou= tcomes. MRI images of 70 glioblastoma multiforme (GBM) patients were review= ed by six neuroradiologists using the VASARI= scoring system; 620 angiogenesis genes were tested. Patient outcome wa= s measured as duration of survival and time to recurrence. Eight MRI featur= es were associated with survival (unadjusted p-value <0.05). Ependymal e= xtension (feature F19) was correlated with the shortest survival (p=3D0.001= 2). Expression of ANG and TGFB2 genes correlated with shorter survival. CCL= 5 and TNF genes correlated with longer survival. A statistical model incorp= orating F19 with expression of the above genes correctly predicted survival= for 82% of patients. Left hemispheric tumor location (feature F2) correlat= ed with the longest time to recurrence (p=3D0.0084). An optimal linear regr= ession model was constructed to include F2 and expression of the STAT1, ARH= GAP24, and SSTR2 genes. This study demonstrated that a subset of VASARI ima= ging features does correlate with survival and time to recurrence. Linear r= egression models incorporating one or multiple imaging features and tumor g= ene expression can reliably predict patient outcome.
Preliminary analysis has been derived from Round 1 of the VASARI Research Project and presented at t= he following conferences:
The study is still ongoing and is now being reviewed with the inclusion = of Round 2 data as well.
The following shared lists have been created to easily obtain the subset= of The Cancer Genome Atlas (TCGA)-GBM images= relevant to this study.
Note: See Section 3.7 of TCIA User Guid= e for help with Shared Lists.
Corresponding gene, survival, and recurrence data was obtained from TCGA Data Portal. The following text file contains the full list of= sample IDs from the data portal which were used in the preliminary analysi= s: