Poster C43, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

The role of white-matter tracts in language processing in patients with brain tumors

Svetlana Malyutina1, Olga Dragoy1, Elisaveta Gordeyeva1, Andrey Zyryanov1, Dmitry Kopachev2, Ekaterina Stupina1, Valeria Tolkacheva3, Igor Pronin2, Maria Ivanova1;1National Research University Higher School of Economics, 2Burdenko National Scientific and Practical Center for Neurosurgery, 3European Master in Clinical Linguistics

Contemporary hodotopic models emphasize the critical role of white-matter tracts, rather than just cortical areas, in language processing (Catani, Jones, & ffytche, 2005). Still, the specific roles of different tracts in particular components of language processing remain to be delineated. The neurosurgical population provides a unique opportunity for studying this. Unlike stroke-related damage, brain tumors grow over time, allowing for functional re-organization of language processing. So, language deficits in the neurosurgical population demonstrate which functions of white-matter tracts could not be re-mapped onto a different neural substrate during tumor growth. This allows to establish unique contributions of different tracts to language processing. This study investigated correlations between the integrity of left-hemisphere white-matter tracts and language processing in brain tumor patients pre-operatively. Participants were 30 Russian-speaking individuals with brain tumors in the left frontal, insular, or temporal lobe (WHO grade II/III) recruited at the Burdenko National Scientific and Practical Center for Neurosurgery. Before neurosurgery, every patient underwent language testing and diffusion brain MRI. Language testing was performed with the Russian Aphasia Test (RAT; Ivanova et al., 2016), testing comprehension and production at the phonological, lexico-semantic, syntactic and discourse level. Diffusion-weighted MR imaging was performed on a 3T scanner (64 directions, 2.5 mm isovoxel, b=1500 s/mm2, two repetitions with opposite phase-encoding directions). After pre-processing in FSL and ExploreDTI, major left-hemisphere tracts were manually delineated in TrackVis in a deterministic tensor-based model: arcuate fasciculus (long, anterior and posterior segment), frontal aslant tract, inferior longitudinal, inferior fronto-occipital, and uncinate fasciculi. To establish tract-function relations, scores on language subtests were correlated with tract volume. Overall, the patients had minor language deficits: the mean RAT score (on a scale from 5, no deficits, to 0, severe deficits) was 4.82 (SD 0.19, range 4.20-5.00). Sentence comprehension scores were significantly positively correlated with the volume of the posterior segment of the arcuate fasciculus. Word, sentence and discourse production scores were significantly positively correlated with the volume of both long and posterior segment of the arcuate fasciculus. Unexpectedly, word and pseudoword repetition scores were negatively correlated with the volume of the frontal aslant tract (higher tract volume was associated with lower scores). No other significant correlations were observed. The findings are consistent with previous studies showing the role of the left arcuate fasciculus in syntactic processing (Wilson et al., 2011; Grossman et al., 2013) and language production at different levels (Bello et al., 2008; Duffau et al., 2002; Marchina et al., 2011). However, this is one of the few studies separately considering the segments of the arcuate fasciculus: here, syntactic processing was associated with the posterior segment, whereas language production was correlated with both long and posterior segment. In this sample, damage to inferior longitudinal, inferior fronto-occipital, or uncinate fasciculi was not associated with any language deficits. Further testing of patients with damage to these tracts is warranted. The results are relevant to hodotopic models of language processing and to clinical prognosis of language deficits and planning of awake neurosurgeries. The study was supported by RFFI (grant no. 16-06-00400).

Topic Area: Language Disorders