Characterization of MRI findings by pTERT mutation status and the prognostic significance in GBM patients with recurrent lesions

Aim

To investigate the MRI characteristics of recurrent glioblastoma (GBM) in relation to telomerase reverse transcriptase-promoter (pTERT) mutation status and their prognostic significance.

Materials and Methods

A retrospective analysis was conducted on 80 patients with primary IDH wild-type GBM. MRI features, including conventional and functional imaging, were compared between pTERT mutant and wild-type groups. Cox regression and receiver operating characteristic (ROC) analyses were performed to evaluate survival outcomes and predictive models.

Results

Compared to patients with wild-type pTERT (n = 24), those with mutant pTERT (n = 56) were older (p = 0.008) and had shorter overall survival (OS) (p = 0.003). They also exhibited a higher incidence of massive-like enhancement (p = 0.036) and new subependymal enhancement (p = 0.034), lower relative apparent diffusion coefficient (rADC) (p = 0.006), and higher relative cerebral blood volume (rCBV) (p = 0.048) and choline/N-acetyl aspartate ratio (p = 0.033). Multivariate analysis identified new subependymal enhancement (HR = 2.348, p = 0.018) and massive-like enhancement (HR = 2.348, p = 0.018) as independent risk factors for shorter OS. The areas under the ROC curve (AUCs) for predicting poor OS using pTERT mutation status, conventional MRI, and combination models were 0.700, 0.695, and 0.764, respectively. Among the 31 patients with available functional MRI data, the AUCs for prediction using functional MRI, conventional MRI + pTERT mutation status, and the combination model were 0.783, 0.710, and 0.888, respectively.

Conclusions

MRI features, particularly subependymal and massive-like enhancements, along with pTERT mutation status, serve as valuable prognostic markers for survival outcomes in recurrent GBM patients. The combination model incorporating MRI characteristics and genetic status enhances predictive accuracy.