DTI-ALPS is not independently associated with overall survival beyond established prognostic factors in glioblastoma

Impaired glymphatic function has been associated with shorter survival in glioblastoma. However, the relationship between glymphatic dysfunction and survival has not yet been evaluated in the context of established prognostic factors, including extent of resection, adjuvant therapy, MGMT promoter methylation status, and clinical performance status.

Single-center retrospective study including patients undergoing surgery for unifocal glioblastoma IDHwt CNS WHO grade 4 (GBM), with preoperative DTI studies in comparison to healthy controls (HC). Glymphatic impairment was assessed using diffusion tensor imaging analysis along the perivascular space (DTI-ALPS), which is an accepted surrogate marker for glymphatic function. This study follows the STROBE guidelines for cohort studies.

We included 47 patients with GBM and 50 HC. The hemispherically averaged ALPS index was lower in GBM patients (1.15 vs. 1.27, p = 0.01). There were no differences between contralateral ALPS and HC (1.22 vs. 1.27, p = 0.29). Ipsilateral ALPS was lower compared to HC (1.05 vs. 1.27, p < 0.001) and to the contralateral hemisphere (1.1 vs. 1.22, p = 0.01). Univariable Cox regression showed no association between overall survival and ALPS parameters (ipsilateral HR 2.25 95% CI 0.49-10.31 p = 0.3, contralateral HR 2.96 95% CI 0.57-15.32 p = 0.2). In multivariate analysis significant associations were found for female and extent of resection but not for MGMT status, age, ipsi- or contralateral ALPS indices.

In patients with GBM, glymphatic function might be impaired as expressed by decreased ALPS index in the ipsilateral hemisphere. Ipsi- or contralateral ALPS index reduction did not predict overall survival independent of known predictors. Because DTI-ALPS has inherent limitations, further investigation of glymphatic dysfunction using different methods in glioblastoma is warranted.