Brain tumour microstructure is associated with post-surgical cognition

Abstract

Brain tumour microstructure is potentially predictive of changes following treatment to cognitive functions subserved by the functional networks in which they are embedded. To test this hypothesis, intra-tumoural microstructure was quantified from diffusion-weighted MRI to identify which tumour subregions (if any) had a greater impact on participants' cognitive recovery after surgical resection. Additionally, we studied the role of tumour microstructure in the functional interaction between the tumour and the rest of the brain. Sixteen patients (22-56 years, 7 females) with brain tumours located in or near speech-eloquent areas of the brain were included in the analyses. Two different approaches were adopted for tumour segmentation from a multishell diffusion MRI acquisition: the first used a two-dimensional four group partition of feature space, whilst the second used data-driven clustering with Gaussian mixture modelling. For each approach, we assessed the capability of tumour microstructure to predict participants' cognitive outcomes after surgery and the strength of association between the BOLD signal of individual tumour subregions and the global BOLD signal. With both methodologies, the volumes of partially overlapped subregions within the tumour significantly predicted cognitive decline in verbal skills after surgery. We also found that these particular subregions were among those that showed greater functional interaction with the unaffected cortex. Our results indicate that tumour microstructure measured by MRI multishell diffusion is associated with cognitive recovery after surgery.

Keywords: Brain tumors; Diffusion MRI; Microstructure; Neurosurgery; Tumour microstructure.

Figure 1

Flowchart of data processing and analysis. Two methodological approaches (DTI -top- and NODDI -bottom-) were used to identify tumour subregions according to its microstructure. Segmented tumours were used to separately analyse the correlation between BOLD signal within each tumour subregion and the Global Signal from the healthy brain. Additionally, the presence of tumour subregions was also correlated with cognitive decline after surgery.

Figure 2

Illustration of the spatial distribution of the different voxel groups within the tumour in two different participants (first row: Participant 1 (S1), second row: Participant 2 (S2)). (Left) T1 weighted image. (Middle) p-q-derived tumour subregions. Voxel group I is displayed in red; Voxel group II: dark blue; Voxel group III: Light blue and Voxel group IV: yellow. (Right) NODDI-derived tumour subregions: C1 is displayed in light blue; C2: yellow; C3: red; C4: green.

Figure 3

(A) Association of mean cognitive decline after surgery with the percentage occupancy of each p-q-derived subregion (groups I, II, III and IV). (B) Association of mean cognitive change after surgery and the percentage occupancy of p-q-derived group III subregion for each of the 5 cognitive domains (attention, non-verbal skills, memory, verbal skills and executive function).

Figure 4

Distribution of decline in mean cognition (A) and verbal skills (B) across participants (represented by individual dots) depending on the presence or not of each NODDI-derived subregion (C1, C2, C3, C4 and C5).

Figure 5

Distribution of β association values across participants (represented by individual dots) between: 1) BOLD signal derived from grey matter outside the tumour (GS), and 2) BOLD signal derived from the tumour subregions and from the region contralateral to the tumour (tumour (contra)). In the latter case, β values were calculated using a GS estimation that excluded the corresponding voxels contralateral to the tumour (to avoid overlapping between the independent and dependent variables). White dots represent the median of the data in each case; the thick grey line indicates the interquartile range. (A) DTI approach with tumour subregions: groups I, II, III and IV. * indicates significant differences, p-FDR < 0.01. (B) NODDI approach with tumour subregions: C1, C2, C3, C4 and C5.

Figure 6

Distribution of group III subregions (left brains) and per-voxel functional coupling with GS (right brains). Participant numbers in the figure match those presented in Table 1.