gms | German Medical Science

Objective: Monitoring glioma treatment usually focuses on MRI-based macroanatomical changes at the primary tumor site and its borders, although diffuse tumor cell invasion is known to exceed the locally apparent tumor. Tumor associated alterations of systemic extralesional white matter and cognitive decline therefore may precede locally apparent disease dynamics. We applied DTI and neuropsychological testing to investigate longitudinal microstructural alterations of normal appearing white matter (NAWM) and cognitive functions in glioma patients.

Methods: 12 patients (mean age: 43 +/- 14 years, 8 males, 8 high grade gliomas) and 12 individually age- and gender-matched healthy controls underwent DTI and a standardized neuropsychological examination pre- (t₀) and 4 +/- 2 months postoperatively (t₁). Seven patients received adjuvant radiochemotherapy postoperatively. DTI data were skeletonized and fractional anisotropy (FA) and mean-, radial- and axial diffusivity values (MD, RD and AD) of NAWM were derived voxel-wise after prior tumor masking, restricting DTI-analyses to extralesional white matter. DTI metrics and cognitive performance measures were used for between- (t₀) and within-group comparisons (t₀ vs. t₁). Effect sizes (d) were calculated for significant results and indicate small (0.2–0.5), medium (0.5–0.8) and strong effects (> 0.8).

Results: Cognitive performance was worse in treatment-naíve patients as compared to controls with regard to verbal memory and executive performance at t₀ (verbal recall: Z=-2.193, p<0.05; d=1.62; verbal recognition: Z=-2.252, p<0.05 d=1.71; cognitive flexibility: Z=-2.521, p<0.001, d=2.12), and further deteriorated, however not significantly, in all cognitive domains tested at t₁. In addition, patients revealed significantly lower FA and higher RD of NAWM than controls already at t₀ (FA: Z=-2.394, p<0.01, d=1.91; RD: Z=-3.059, p<0.001, d=3.76), indicating microstructural disintegration in NAWM even before treatment was initiated, and with a further significant deterioration at t₁ (FA: Z=-2.316, p<0.01, d=1.80; RD: Z=1.844, p<0.05, d=1.26).

Conclusion: DTI metrics were sensitive to longitudinal extratumoral microstructural alterations of NAWM, which may precede functionally and macroscopically apparent disease dynamics. We suggest DTI of extralesional white matter to be considered for treatment monitoring in diffuse glioma.