We proceeded with a multivariate Cox regression model including significant indices (age, radiotherapy, chemotherapy, TSPO-PET signal of the tumor, T1 contrast–enhanced volume, T2 hyperintensity volume) as well as MGMT, which likely did not reach significance in univariate Cox regression due to the lower sample size compared to a previous study ( 48 ), and glucocorticoid medication due to borderline significance.
← all excerpts
Remote Neuroinflammation in Newly Diagnosed Glioblastoma Correlates with Unfavorable Clinical Outcome.
1
—
—