PreprintPDF Available

Brain network disruption predicts memory and attention deficits after surgical resection of glioma

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Surgical resection with adjuvant chemotherapy and radiotherapy are effective treatments to delay brain tumour progression and improve survival. Nevertheless, a large proportion of patients have treatment-induced cognitive deficits that dramatically reduce their life quality. A major problem in basic and clinical neuroscience is that the dispersed and interlocking nature of cognitive circuits makes predicting functional impairments challenging. Here we investigated tumour interactions with brain networks in relation to cognitive recovery after surgical resection and during chemo-radiotherapy treatment. 17 patients with low- and high-grade glioma (aged 22-56 years) were longitudinally MRI-scanned and cognitively assessed using a tablet-based screening tool before and after surgery, and during a 12-months recovery period. Using structural MRI and Neurite Orientation Dispersion and Density Imaging (NODDI) derived from diffusion-weighted images, we respectively estimated tumour overlap and Neurite Density (as an in-vivo proxy measure of axon and dendrite concentration) with brain networks and functional maps derived from normative data in healthy participants. We found that neither total lesion volume nor tumour location based on traditional lobular divisions were associated with memory or attention deficits. However, tumour and lesion overlap with the Default Mode Network (DMN), Attention Network and attention-related regions located in frontal and parietal cortex was associated with memory and attention deficits. This association was above and beyond the contributions of preoperative cognitive status and tumour volume (Linear Mixed Model, Pfdr<0.05). On the other hand, Neurite Density derived was reduced not only within the tumour, but also beyond the tumour boundary, revealing a distal effect that can have global consequences on brain networks. High preoperative Neurite Density outside the tumour, but within the Frontoparietal Network was associated with better memory and attention recovery. Moreover, postoperative and follow-up Neurite Density within the DMN, Frontoparietal and Attention Networks was also associated with memory and attention improvements (Pfdr<0.05). We conclude that gliomas located on brain networks that are fundamental for cognitive processing mediate cognitive deficits in patients with brain tumours and that, despite being focal lesions, they exert a distal effect on Neurite Density in these networks that is also associated with cognitive recovery. Our work provides insights into the brain reorganisation that occurs due to the presence of a tumour and its potential capability to predict treatment-induced cognitive deficits. A better understanding of the impact of treatment on these brain circuits would contribute to designing multimodal biomarkers for enhancing patient stratification and tailored rehabilitation to better preserve cognitive function.
Content may be subject to copyright.
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Brain network disruption predicts memory and attention
deficits after surgical resection of glioma
Romero-Garcia R1, §,*, Hart MG2,*, Owen M1, Assem M3, Coelho P4, McDonald A5, Woodberry E5, Price
SJ2, Burke GAA6, Santarius T2,7, Erez Y3, Suckling J1,8,9
* These authors contributed equally
1. Department of Psychiatry, University of Cambridge, 2. Department of Neurosurgery, Addenbrooke’s
Hospital, Cambridge, 3. MRC Cognition and Brain Sciences Unit, University of Cambridge, 4. Neurophys
Limited, 5. Department of Neuropsychology, Cambridge University Hospitals NHS Foundation Trust, 6.
Department of Paediatric Haematology, Oncology and Palliative Care, Addenbrooke’s hospital,
Cambridge 7. Physiology, Development and Neuroscience, University of Cambridge, 8. Behavioural
and Clinical Neuroscience Institute, University of Cambridge; 9. Cambridge and Peterborough NHS
Foundation Trust.
§ Correspondence to: Rafael Romero-Garcia PhD, University of Cambridge, Department of Psychiatry,
Herchel Smith Building for Brain and Mind Sciences, Robinson Way, Cambridge CB2 0SZ, UK Email:
rr480@cam.ac.uk
Keywords: Brain Tumours, structural MRI, diffusion MRI, Neurosurgery, Brain Networks, NODDI, Low-
Grade Glioma, Cognitive function, memory, attention, patients
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
ABSTRACT
Surgical resection with adjuvant chemotherapy and radiotherapy are effective treatments to delay
brain tumour progression and improve survival. Nevertheless, a large proportion of patients have
treatment-induced cognitive deficits that dramatically reduce their life quality. A major problem in
basic and clinical neuroscience is that the dispersed and interlocking nature of cognitive circuits makes
predicting functional impairments challenging. Here we investigated tumour interactions with brain
networks in relation to cognitive recovery after surgical resection and during chemo-radiotherapy
treatment. 17 patients with low- and high-grade glioma (aged 22-56 years) were longitudinally MRI-
scanned and cognitively assessed using a tablet-based screening tool before and after surgery, and
during a 12-months recovery period. Using structural MRI and Neurite Orientation Dispersion and
Density Imaging (NODDI) derived from diffusion-weighted images, we respectively estimated tumour
overlap and Neurite Density (as an in-vivo proxy measure of axon and dendrite concentration) with
brain networks and functional maps derived from normative data in healthy participants. We found
that neither total lesion volume nor tumour location based on traditional lobular divisions were
associated with memory or attention deficits. However, tumour and lesion overlap with the Default
Mode Network (DMN), Attention Network and attention-related regions located in frontal and
parietal cortex was associated with memory and attention deficits. This association was above and
beyond the contributions of preoperative cognitive status and tumour volume (Linear Mixed Model,
Pfdr<0.05). On the other hand, Neurite Density derived was reduced not only within the tumour, but
also beyond the tumour boundary, revealing a distal effect that can have global consequences on brain
networks. High preoperative Neurite Density outside the tumour, but within the Frontoparietal
Network was associated with better memory and attention recovery. Moreover, postoperative and
follow-up Neurite Density within the DMN, Frontoparietal and Attention Networks was also associated
with memory and attention improvements (Pfdr<0.05). We conclude that gliomas located on brain
networks that are fundamental for cognitive processing mediate cognitive deficits in patients with
brain tumours and that, despite being focal lesions, they exert a distal effect on Neurite Density in
these networks that is also associated with cognitive recovery. Our work provides insights into the
brain reorganisation that occurs due to the presence of a tumour and its potential capability to predict
treatment-induced cognitive deficits. A better understanding of the impact of treatment on these
brain circuits would contribute to designing multimodal biomarkers for enhancing patient
stratification and tailored rehabilitation to better preserve cognitive function.
INTRODUCTION
Every year more than 300,000 people worldwide face the diagnosis of a brain tumour. As a result of
their tumours, a large proportion of patients develop cognitive deficits ranging from 29% in patients
with non-irradiated low-grade glioma, to about 90% in groups of patients with diverse brain tumours
(Klein et al., 2002; Meyers et al., 2004).
Patients who undergo surgical resection rather than biopsy have an improved overall survival (Jakola
et al., 2012), and extending the resection beyond the abnormality seen on MRI further improves
prognosis (Yordanova et al., 2011). However, the extent of resection is only a worthwhile prognostic
factor in the management of the tumour if subsequent cognitive functioning can be maintained, and
therefore any impairment of brain structure and function is a significant risk factor for a reduction in
quality of life. Consequently, preoperative anticipation of post-surgical cognitive alterations
represents a major interest for the clinical community.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Despite being recognised as a fundamental outcome measure of treatment success, cognitive deficits
still remain one of the most unpredictable aspects of patients’ prognosis (Taphoorn and Klein, 2004).
Unfortunately, clinicians have often underestimated the impact of surgery on functional outcome
(Ferroli et al., 2015; Sagberg et al., 2017). Radiotherapy (Douw et al., 2009) and chemotherapy (Wefel
and Schagen, 2012) (Douw et al., 2009)(Douw et al., 2009)[8][8][8][8]can also have a profound impact
on several cognitive domains such as attention, executive functioning and information processing
speed. Tumour- and treatment-induced cognitive impairment not only has a dramatic impact on
patients’ quality of life but it has also been recognised as a significant prognostic factor in patient
survival (Klein et al., 2003).
Structural MRI provides unique insights into the brain that are currently essential for brain tumour
diagnosis and monitoring. In routine clinical care, MRI is the standard technique for demarcating
abnormal tumoural regions, and it is the main method used to assess treatment response. However,
typical MRI sequences used for clinical evaluation are limited by low biological specificity that reduces
their capability to differentiate tumour types, infiltration, as well as macromolecular and histological
compositions. For example, although gadolinium contrast enhancement is currently used with MRI to
determine the malignancy of tumours, up to one-third of high-grade gliomas are non-enhancing (Scott
et al., 2002).
Recently developed MRI sequences that estimate tissue microstructure integrity have revolutionised
in-vivo investigation of the human brain, but their applications to neuro-oncology have been limited.
The capability of modern MRI protocols has been demonstrated for estimates of tumour infiltration
(Li et al., 2018), grading (Fan et al., 2006), tumour heterogeneity (Just, 2014), tumour progression
(Mohsen et al., 2013) and patient survival (Peng et al., 2013). The incorporation of recent MRI
protocols into the clinical routine may be critical to prevent treatment-induced neurocognitive
dysfunction in paediatric (Ajithkumar et al., 2017) and adult brain tumours (Ahles et al., 2012).
Notwithstanding, most previous brain tumour studies have used tensor models to explore
microstructural changes under the assumption of Gaussian diffusion processes. In contrast, the
Neurite Orientation Dispersion and Density Imaging (NODDI) technique is a recently developed multi-
shell sequence that uses varying gradients strengths to provide more specific measurements of tissue
microstructure than traditional Diffusion Tensor Imaging (DTI) (Zhang et al., 2012). Intra-cellular
diffusion estimation derived from NODDI as a marker of Neurite Density has been validated using
histochemical analysis in mouse models (Wang et al., 2019) and it has shown sensitivity to detect
axonal degeneration in neurological conditions such as Alzheimer’s (Colgan et al., 2016) and
Parkinson’s Disease (Kamagata et al., 2016). Caverzasi et al. (2016) showed that even for tumour
lesions that appear to be homogeneous on corresponding fluid-attenuated inversion-recovery (FLAIR)
images, NODDI has the potential to identify some infiltrative tumour components. Despite these
promising findings, no study has yet used NODDI to evaluate the impact of tumours and their
treatment on cognition.
Low-grade gliomas are slow-growing and infiltrative tumours involving glial cells. The restricted
proliferation rate reduces the lesion momentum”, which is defined broadly as the growth kinetics
and aggressiveness of a tumour’s evolution. It has been suggested that the chronic and slowly
progressive nature of these tumours allows more time for neuroplastic reorganisation which improves
not only survival rates, but also has a protective effect on neurocognitive functioning (Wefel et al.,
2016). Nevertheless, it is well established that although tumours and surgical resection represent focal
lesions within the brain, tumours have an overall impact on cognitive performance (Anderson et al.,
1990) that may be mediated by a disruption of distant neuronal circuits. These long-range effects
produced by focal brain tumour have been also observed at functional level, with gliomas reducing
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
long-range functional connectivity (Harris et al., 2014; Maesawa et al., 2015; van Dellen et al., 2012)
and globally disrupting the functional integrity (Hart et al., 2019).
A major challenge when predicting the consequences of surgery and radio-chemotherapy is that
higher-order cognitive functions are sub-served by communication across spatially extended neural
circuits that cannot be completely captured using traditional localizational models, and which
therefore require whole-brain approaches. Large-scale brain networks represented by connectomics
offer a holistic framework to analysing the brain as a circuit of interacting components by modelling
brain regions as nodes and connections between regions as edges that are critically related to
cognition (Bressler and Menon, 2010). In neuro-oncology, markers derived from brain network
approaches have demonstrated to be sensitive to the presence of low-grade glioma (Xu et al., 2013),
plasticity differences between low- and high-grade glioma (van Dellen et al., 2012), and surgically
induced alterations (Huang et al., 2014). Consequently, the use of brain network data to explore the
potential for cognitive disruption induced by brain tumours and their treatment could be of major
clinical relevance.
In this longitudinal study, we prospectively investigated whether the interaction between tumours
and normative brain networks derived from healthy participants can predict cognitive recovery
postoperatively and during a follow-up period of 12 months. We hypothesised that surgically-induced
cognitive deficits are associated with disruption of brain networks that have been previously identified
as fundamental for cognition.
METHODS
Sample
This study is a single centre prospective cohort design approved by the Cambridge Central Research
Ethics Committee (protocol number 16/EE/0151). Patients were identified at adult neuro-oncology
multidisciplinary team (MDT) meetings at Addenbrooke's Hospital (Cambridge, UK), and a consultant
neurosurgeon directly involved in the study identified potential patients based on the outcome of the
MDT discussion. Inclusion criteria included: (i) Participant is willing and able to give informed consent
for participation in the study, (ii) maging is evaluated by the MDT and judged to have typical
appearances of a diffuse low-grade glioma (DLGG), (iii) Stealth MRI is obtained (routine
neuronavigation MRI scan performed prior to surgery), (iv) WHO performance status 0 or 1, (v) 18
Age ≤ 80, (vi) tumour located in or near eloquent areas of the brain thought to be important
for speech and executive functions and (vii) patient undergoing awake surgical resection of DLGG. This
last inclusion criterion was adopted to collect additional intraoperative electrocorticography data that
has not been considered in the present manuscript. Participants were excluded if any of the following
apply: (i) There is uncertainty about the radiological diagnosis, (ii) concomitant anti-cancer therapy
(except for dexamethasone treatment), (iii) history of previous malignancy (except for adequately
treated basal and squamous cell carcinoma or carcinoma in-situ of the skin) within 5 years, and (iv)
previous severe head injury.
We recruited 17 patients aged 22-56 years (8 females) with different grades of glioma: WHO-I N=2,
WHO-II N=6, WHO-III N=7, WHO-IV N=2. Resection was completed in 4 patients, partial resection in 7
patients, and resection was non-measurable in 6 patients. 12 patients were also treated with
chemotherapy or radiotherapy post-surgery. Each patient was scanned up to four times: before
surgery (preop), after surgery (postop), and during the follow-up period at 3 and 12 months (month-
3 and month-12). See Table 1 for demographic details of participants.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Age
Gender
Handed-
ness
Presenta-
tion
Hemisphere
Location
Tumour Grade /
Pathology
41
Female
Left
Seizures
Left
Frontal
Grade II
Oligodendroglioma
32
Male
Right
Seizures
Right
Insula
Grade II
Astrocytoma
26
Male
Right
Seizures
Left
Temporal /
Insula
Grade IV
Glioblastoma
49
Female
Right
Incidental
Right
Insula
Grade II
Oligodengroglioma
55
Female
Right
Recurrence
Left
Frontal / SFG
/ frontal pole
Grade II
Oligodendroglioma
22
Female
Left
Seizures
Right
Frontal / IFG
Grade I
Ganglioglioma
29
Male
Right
Seizures
Right
Frontal / SFG
& MFG
Grade III
Astrocytoma
29
Male
Right
Seizures
Right
Frontal / MFG
Grade III
Astrocytoma
50
Male
Left
Seizures
Left
Temporal /
ITG
Grade IV
Glioblastoma
38
Female
Right
Seizures
Right
Frontal / MFG
Grade II
Oligodendroglioma
29
Male
Right
Seizures
Left
Frontal / SFG
/ frontal pole
Grade II
Astrocytoma
33
Female
Right
Headaches
Left
Temporal /
MTG
Grade III
Astrocytoma
27
Female
Right
Seizures
Left
Superior
Temporal
Gyrus
Grade I
Ganglioglioma
56
Female
Right
Seizures
Left
Superior
Temporal
Gyrus
Grade II
Astrocytoma
32
Male
Right
Seizures
Left
Superior
Temporal
Gyrus
Grade III
Astrocytoma
27
Male
Right
Seizures
Left
SFG/SMA &
Pre-central
Grade IV
Glioblastoma
30
Male
Right
Seizures
Left
Inferior
frontal
Grade III
Astrocytoma
Table 1. Demographic Information. SFG, Superior Frontal Gyrus; MFG, Middle Frontal Gyrus; IFG,
Inferior Frontal Gyrus; ITG, Inferior Temporal Gyrus; MTG, Middle Temporal Gyrus; SMA,
Supplementary Motor Area; RT, radiotherapy
MRI and NODDI data acquisition and pre-processing
MRI data were acquired using a Siemens Magnetom Prisma-fit 3 Tesla MRI scanner and 16-channel
receive-only head coil (Siemens AG, Erlangen, Germany). A T1-weighted MRI [magnetization-prepared
rapid gradient-echo (MPRAGE) sequence] was acquired using the following parameters: repetition
time (TR) = 2300 ms, echo time (TE) = 2.98 ms, flip angle (FA) = 9 deg, 1 mm3 isotropic voxel resolution
and a Field of View (FoV) = 256×240 mm2, 192 contiguous slices and acquisition time of 9 minutes and
14 seconds.
During the same scanning session, we additionally acquired a recently developed MRI multi-shell
diffusion technique, Neurite Orientation Dispersion and Density Imaging (NODDI) with 30 gradient
directions with b-value=800 mm/s, 60 gradient directions with b-value=2000 mm/s and ten
unweighted B0 images. Other acquisition parameters were: TR = 8200 ms, TE = 95 ms, 2.5 mm3 voxel
resolution, 60 slices, FOV = 240 mm and acquisition time of 15 minutes and 19 seconds. Images were
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
corrected for B0 field inhomogeneity, Gibbs artefacts and eddy-current distortions using MRtrix 3
(https://www.mrtrix.org/) and FSL 5.0 (http://fsl.fmrib.ox.ac.uk). All scans and processed data were
visually inspected by an experienced researcher (RRG).
NODDI Matlab Toolbox (http://mig.cs.ucl.ac.uk/index.php?n=Tutorial.NODDImatlab) was used to
quantify the microstructural complexity of dendrites and axons in vivo (Zhang et al., 2012). Compared
with traditional Diffusion Tensor Imaging (DTI), the multi-compartment tissue model implemented in
this toolbox disentangles two key contributing factors of Fractional Anisotropy: the Gaussian
contribution from water molecules located in the extracellular space (defined as the space around
neurites), and the restricted non-Gaussian diffusion that takes place in the intra-cellular space that is
bounded by axonal and dendritic membranes. The apparent intra-cellular volume fraction that
represents the fraction of dendrites and axons was used here as a measurement of Neurite Density.
Lesion masking and image transformation to standard space
Lesion masks were created using a semi-automated procedure. An experienced neurosurgeon (MH)
initially did a manual delineation of lesion masks for each participant on the preoperative T1 image
slices that included the tumour, and the resection site and damaged tissue on the follow-up T1 images.
Resulting masks were refined by the Unified Segmentation with Lesion toolbox
(https://github.com/CyclotronResearchCentre/USwithLesion) that uses tissue probability maps to
create a posterior tumour/lesion probability map. Inter-regional distances to the tumour boundary as
defined in the tumour mask were estimated as the geodesic distance of the shortest path constrained
by the white matter.
For each scan, the first B0 image of the diffusion sensitive sequence was linearly coregistered to the
T1 image using FSL FLIRT. The resulting inverse transformation was used to map the Neurite Density
map into the T1 image space. Each T1 image was non-linearly coregistered to standard space using
FSL FNIRT but excluding the tumour/lesion mask from the non-linear step of the wrapping to avoid
distorting the spatial distribution of the tumour/lesion. The resulting transformation was additionally
used to map the lesion mask and the Neurite Density map from T1 space to standard template space.
The ICBM 2009a symmetric brain, an unbiased non-linear average of the MNI152, was used here as a
standard template for normalisation of Neurite Density using the contralateral values hemisphere that
contained the tumour.
Networks atlas and meta-analysis maps based on normative data
Functional meta-activation maps were downloaded from Neurosynth (https://neurosynth.org/), a
platform for large-scale, automated synthesis of functional MRI that included data from 507,891
activation maps reported in 14,371 studies. Neurosynth generates statistical inference maps (i.e., Z
and p-value maps) which display the likelihood of a given term being used in a study if activation is
observed at a particular voxel. We used four terms that we hypothesised to be related to four of the
cognitive domains assessed by this study: attention, memory, perception and calculation (Figure 1,
top left corner). This resulted in four meta-analysis uniformity maps that were binarised at Z=0 (i.e.
negative values ignored).
We additionally exploited the map of large-scale networks defined in Yeo et al. (2011). This atlas was
created by clustering functionally coupled regions in 1000 young, healthy adults. Analogous to meta-
analytic maps derived from Neurosynth, regions delimited on the 7-Network liberal version of the Yeo
atlas were used as Regions of Interest (ROIs) for calculating tumour overlap and Neurite Density
(Figure 1, top right corner).
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Figure 1. Flowchart of pipeline analysis. After transforming tumour and lesion masks to standard space, tumour
(for preoperative images) and lesion (for postoperative and follow-up images) spatial overlap and Neurite
Density was calculated for each of the seven networks defined by Yeo et al. (2011) (top right) and each of the
four Neurosynth meta-analysis maps considered here (attention, memory, perception and calculation; top left).
For each assessment (postoperative, 3 months and 12 months), overlap and Neurite Density were compared
with cognitive recovery of each domain across participants.
Tumour overlapping and Neurite Density estimation
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
As location of the tumour varied across patients, we calculated, for each ROI and participant, the
Tumour Overlap Index” as the proportion of regional volume (in mm3) that spatially overlapped with
the tumour (for preoperative images) or the lesion (for postoperative and follow-up assessments)
according to the tumour/lesion masks, after being transformed to standard space.
Median Neurite Densities in each ROI defined on the 7-network Yeo atlas and the 4 Neurosynth maps
were calculated for each participant. Only voxels of the ROI that were not overlapping with the
tumour/lesion mask were included in the analysis to reduce the impact of tumour volume on Neurite
Density estimation. See Figure 1 for an illustrative flowchart.
Cognitive assessment
Immediately after each scanning session, cognitive performance was evaluated using the OCS-bridge
tablet-based screening tool (https://ocs-bridge.com/), which is specifically designed for patient
populations. OCS-Bridge automatically compares a patient’s score to the distribution (‘norm’) for
people of the same age, sex and educational history sampled from the general population. It consists
of 10 screening tasks designed to assess 6 major cognitive domains: attention, memory, perception,
praxis, language and number processing. For each individual task, a z-score value was computed by
subtracting the mean and dividing by the standard deviation of task scores across participants. Tasks
defined by an inverted scale where high values represent low cognitive performance were flipped by
multiplying the z-score by -1. By subtracting the preoperative z-score from the z-score of each
subsequent task, we defined the longitudinal trajectory of each assessment (Δ) where negative scores
represent worse performance than before surgery (i.e. cognitive deficit) and positive scores are
associated with increased performance than before surgery (i.e. cognitive recovery). For each of the
6 domains, score was calculated as the average z-score of all tasks associated to that domain: attention
(space/object neglect and accuracy, perseveration and organisation index, sustained attention, target
detection, consistency and space bias), perception (visual field, perception and extinction, visual
acuity, line bisection, object perception), memory (free verbal and episodic memory, orientation,
forward and backward digit span, prospective and retrospective memory), language (picture naming,
semantics, reading), praxis (hand and finger positions) and number processing (calculation and
number writing).
Statistical analysis
When the Kolmogorov-Smirnov normality test rejected the null hypothesis of data being sampled from
a normal distribution, Spearman’s rank was used to test associations between imaging and cognitive
data. Pearson’s correlation was used otherwise. Statistical tests that incorporated more than one
assessment from the same patients violates the assumption of independent data. Under this scenario,
we tested the association between variables using a Linear Mixed Effect Model (LMM) with random
intercept and slope that incorporated total tumour volume and total neurite density (only for NODDI
analyses) as covariates:
              
where Cog represents cognitive performance of a given domain (memory, attention, perception and
number processing), X is the predictor variable based on imaging data (i.e. tumour overlap or Neurite
Density for a given network or Neurosynth map), vol is the total volume of the tumour (for
preoperative assessments) or the lesion (for postoperative and follow-up assessments), ND is the total
Neurite Density of the participant (only included for predictions based on Neurite Density) and  
 represents the random intercept and slope. All statistical tests were corrected for
multiple comparisons using BenjaminiHochberg False Discovery Rate (FDR < 0.05).
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
RESULTS
Cognitive recovery trajectories
OCS-Bridge cognitive assessment was completed by 17 patients before surgery, 8 after surgery, 7 after
3 months and 4 after 12 months. Surgical resection and treatment had an impact on cognitive
performance in most participants. Attention, Perception, Memory and Number processing showed a
variety of trajectories, including progressive impairment, impairment followed by recovery, no change
and improvement after surgery and during recovery (Figure 2). Language and Praxis tests showed no
sensitivity to capture deficits in our cohort except for two participants. Consequently, these two
domains will not be considered in further analyses. Cognitive performance was not independent
across cognitive domains. Most of the four domains considered in subsequent analyses (Attention,
Perception, Memory and Number processing) showed a weak positive association that was significant
only for Memory and Attention (R2= 0.26, P<0.05), but that did not survive FDR correction for multiple
comparisons (Figure S1).
Figure 2. Cognitive performance across assessments. Mean z-scores of each participant normalised to
preoperative performance in the six cognitive domains assessed by OCS-Bridge. Each colour represents an
individual participant. Note that lines overlap at zero for language and praxis in most patients.
Impact of tumour volume and tumour overlap on cognitive recovery
The influence of the tumour and its treatment on cognitive recovery was initially evaluated by
comparing the total lesion volume with changes in cognitive performance. We found that although
attention and memory deficits showed a trend of negative correlation, only number processing
deficits were significantly correlated with lesion volume (ρ=-0.66, LMM, Pfdr<0.05, Figure 3).
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Figure 3. Association between lesion volume and cognitive recovery across patients. Lesion volume quantifies
the amount of tissue covered by the tumour (preoperative assessment) or that has been resected or damaged
(postoperative and follow-up assessments). Cognitive recovery refers to average cognitive performance in each
domain normalised to preoperative scores (positive scores correspond with post-surgical recovery and negative
scores to post-surgical deficits). Colours illustrate each individual participant with lines connecting them.
Tumours were located on frontal (5 Left Hemisphere -LH- and 4 Right hemisphere -RH-), temporal (6
LH) and insular (2 RH) cortices (Figure S2). Postoperative and follow-up performance on each of the
four cognitive domains assessed by OCS-bridge (attention, memory, perception, and number
processing) was not significantly different in participants with frontal, temporal and insular tumours
(Figure S3; LMM; all P>0.05). We found significant associations between attention and memory
changes immediately post-surgery with tumours located on specific brain networks. Postoperative
attention deficits were correlated with tumours overlapping with Ventral Attention network (ρ=-0.45,
Pfdr<0.05) and with attention-related regions (ρ=-0.74, Pfdr<0.05; as defined by Neurosynth meta-
analysis; Figure 4A). Similarly, postoperative memory deficits were significantly associated with
tumours overlapping with the Dorsal Attention Network (ρ=-0.46, Pfdr<0.05) and DMN (ρ=-0.60,
Pfdr<0.05). When follow-up assessments were included in the analyses (3 months and 12 months), we
found a significant association between Attention deficits and tumours overlapping with DMN (ρ=-
0.57, Pfdr<0.05) and attention-related regions (ρ=-0.53, Pfdr<0.05). Memory deficits were also
correlated with tumours overlapping with DMN (ρ=-0.59, Pfdr<0.05; Figure 4B). All associations were
tested using LMM after regressing out total volume effects.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Figure 4. Cognitive deficits as a function of tumour overlapping with Yeo networks and Neurosynth maps. (a)
Postoperative cognitive deficits as a function of tumour overlap with Yeo networks and Neurosynth maps.
Cognitive performance was normalised to pre-surgical values and overlap values were defined as the number of
voxels (equivalent to mm3) of the tumour mask that overlapped with each network. (b) Cognitive deficits during
recovery (postoperative, 3 months and 12 months) as a function of lesion overlap with Yeo networks and
Neurosynth maps. Lines link assessments that correspond to the same participant. None of the other domains
or networks showed any significant association that survived correction for multiple comparisons.
Long-range effect of tumours on Neurite Density
The impact of tumours and their treatment on brain structure were also explored using a Neurite
Density marker derived from diffusion imaging based on NODDI. We found that average Neurite
Density outside the tumour was negatively associated with tumour volume (R2= 0.28, P=0.03, Figure
S4), suggesting that the impact of the tumour on structural integrity may not be restricted to tumoral
regions alone. In support of this hypothesis, we found a distance-effect on Neurite Density as a
function of distance to the tumour boundary. Peri-tumoural regions located between 0 and 20 mm
from the tumour boundary had up to half of the Neurite Density compared to contralateral regions
(Figure 5).
Figure 5. Neurite Density as a function of distance to the tumour boundary. Values are normalised to the
contralateral hemisphere (i.e values lower than one represent regions with reduced neurite density). Zero
distance (x=0) corresponds to average Neurite Density values within each tumour. Each colour represents an
individual participant.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Associations between Neurite Density within brain networks and cognitive recovery
Preoperative Neurite Density within Frontoparietal Network in both hemispheres (i.e. not normalised)
was positively correlated with attention and memory deficits during recovery, suggesting that Neurite
Density has an initial protective effect on outcome (ρ=0.30, Pfdr<0.05 and ρ=0.85, Pfdr<0.05,
respectively; Figure 6A). Moreover, Neurite Density during recovery was also associated with cognitive
deficits. When postoperative and follow-up Neurite Density values were considered, Attention and
Memory deficit were correlated with Neurite Density within Ventral Attention (ρ=0.53, Pfdr<0.05 for
attention), Frontoparietal (ρ=0.54, Pfdr<0.005 for attention and ρ=0.79, Pfdr<0.05 for memory) and
DMN (ρ=-0.59, Pfdr<0.005 for attention and ρ =-0.84, Pfdr<0.05 for memory; Figure 6B). All associations
were tested using LMM after regressing out total lesion volume and total Neurite Density effects.
Figure 6. Cognitive deficits as a function of Neurite Density within Yeo networks and Neurosynth maps. (a)
Cognitive deficits during recovery (postoperative, 3 months and 12 months) as a function of preoperative
Neurite Density within brain networks. For a given participant, the same Neurite Density (preoperative) value
was used in all assessments, resulting in vertical lines between them. (b) Cognitive deficits as a function of
Neurite Density within brain networks during recovery (postoperative, 3 months and 12 months). Cognitive
deficits are normalised to preoperative values and Neurite Density corresponds with the average density within
the White Matter of each network after excluding tumour and lesion regions. None of the other domains or
networks showed any significant association that survived correction for multiple comparisons. Colours illustrate
each individual participant with lines connecting them.
DISCUSSION
In this study, we combined MPRAGE and NODDI diffusion MRI with normative brain network data
from healthy participants and functional metanalysis maps derived from Neurosynth to determine
whether cognitive trajectories are affected by the tumour in cognitive-related circuits. We found that
attention and memory deficits were associated with tumour (for preoperative assessments) and lesion
(for postoperative and follow-up assessments) overlap with Attentional Network, DMN and
attentional-related regions. Conversely, Neurite Density derived from NODDI was compromised not
only at the location of the tumour, but also in the area surrounding the tumour, revealing that focal
tumours can induce long-distance disruption in brain tissue. Here, attention and memory recovery
were associated with higher Neurite Density within Frontoparietal networks (pre-operatively) and also
within the DMN and attention networks (postoperatively and follow-up). Overall, these results suggest
that the effect of a tumour on the brain and its cognitive consequences depends on interactions with
brain networks and cognitive-related regions at both local and global levels.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
By using a tablet-based assessment that does not require a trained neuropsychologist to evaluate
cognitive performance, we facilitated follow-up screening while collecting several other pieces of data
that are difficult to acquire using traditional pen and paper clinical interviews tests, such as accurate
reaction times. Despite a previous paper version of this test having shown sensitivity to detect
longitudinal changes in all domains under other neurological conditions (Demeyere et al., 2015; Kong
et al., 2016), our OCS-Bridge assessments did not find longitudinal language and praxis changes in this
sample, presumably because deficits were too subtle and effect sizes too small at our sample size,
suggesting that OCS-Bridge may be complemented by other evaluations that specifically screen these
domains. For the other four domains attention, memory, perception and number processing we
observed a wide variety of cognitive recovery trajectories after surgery and during subsequent
treatment that included: no cognitive change, postsurgical deterioration and then improvement, as
well as, postsurgical improvement and then deterioration. Although the mechanisms behind cognitive
improvement after major surgery are not understood, previous studies have reported similar rates of
cognitive recovery following tumour surgery (Habets et al., 2014; Talacchi et al., 2011).
Notwithstanding, given that pre- and postoperative assessments were performed in a relatively short
period of time, we cannot discard the possibility of practice and learning effects on the tasks.
Tumour location is one of the most relevant features to be considered when estimating the cognitive
risks of surgical resection. In our cohort, gliomas were mainly located in frontal, temporal and insular
cortices on the left hemisphere. As an inclusion criterion required that only patients who underwent
awake surgical resection were included in the study, the higher prevalence of left-hemisphere
tumours is a consequence of recruitment bias. On the other hand, the higher prevalence of frontal,
temporal and insular tumours is consistent with previous studies showing that low- and high-grade
gliomas are relatively scarce in primary cortices and occipital lobes (De Witt Hamer et al., 2013;
Larjavaara et al., 2007). Although several developmental, cytomyeloarchitectonic, neurochemical,
metabolic, and functional reasons have been proposed, the mechanisms behind this preferential
location of gliomas across the brain is still an ongoing debate (Duffau and Capelle, 2004; Ghumman et
al., 2016). The presence of gliomas in secondary and association cortices that have been traditionally
associated with cognitive processing (Goldman-Rakic, 1988) may be an important factor to understand
cognitive deficits induced by the tumour and its treatment. Notwithstanding, we found no cognitive
recovery differences between patients with frontal, temporal or insular tumours. We detected a trend
of a negative association between cognitive recovery and total tumour volume that was significant
only for number processing. Previous studies have reported that patients with larger tumours have
higher risk of cognitive impairment before treatment (Tucha et al., 2000) and are aggravated by the
clinical requirement of having more extensive surgical resections (Talacchi et al., 2011) and larger
irradiation volumes and doses during radiotherapy (Klein et al., 2002).
Beyond the impact that tumour volume itself has on cognitive recovery, we hypothesised that
cognitive deficits are associated with treatment-induced disruption of brain networks that have been
previously identified as fundamental for cognition. In support of this hypothesis, we found significant
associations between treatment-induced attention and memory deficits and lesion overlap with the
DMN, Attentional Networks and attention-related regions derived from the Neurosynth meta-
analysis. Functional networks were defined using normative data from healthy individuals and we did
not explore functional networks from affected patients. Nevertheless, a decrease of functional
connectivity within DMN in patients with brain tumours has been consistently reported in the
literature. DMN functional connectivity is reduced in glioma patients when compared with controls in
both the hemisphere ipsilateral (Esposito et al., 2012) and contralateral (Maesawa et al., 2015) to the
tumour, which is particularly prominent for tumours located on the left side of the brain (Ghumman
et al., 2016). Harris et al. (2014) reported that DMN integrity was associated with WHO tumour grade,
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
but not with total lesion volume. Other networks such as language (Briganti et al., 2012) and motor
(Otten et al., 2012) also had significantly lower functional connectivity in glioma patients. It has been
hypothesised that network-specific functional disruption may mediate the treatment-induced decline
in some cognitive domains such as attention (Charras et al., 2015) and executive function (D’Agata et
al., 2013). However, more research is needed to better understand the role of these networks on the
cognitive deterioration in brain tumour patients.
Not surprisingly, we found a consistent Neurite Density decrease within the tumour. Given that
Neurite Density derived from NODDI and Fractional Anisotropy are strongly correlated (Zhang et al.,
2012), our results align with previous evidence found from DTI studies showing that glioblastoma have
reduced Fractional Anisotropy compared with corpus callosum, subcortical white matter (Beppu et
al., 2005) and the rest of the brain (Sinha et al., 2002). Fractional Anisotropy reduction in glioblastomas
has been associated with decreased fibre density (Roberts et al., 2005) and reduced cell density
markers derived from histopathological evaluation (Kinoshita et al., 2008) and cell proliferation
markers (Beppu et al., 2005; Irie et al., 2018). This is also supported by histological studies revealing a
strong positive correlation between FA and tumour cell density in the mouse (Kinoshita et al., 2008)
and rat brains (Jespersen et al., 2010). Consequently, high-grade tumours show increased FA (White
et al., 2011) that has been associated with low overall survival (Qu et al., 2016). However, the
capability of FA as an independent prognostic parameter beside other established factors such as age
and patient functional status is still unclear (Huber et al., 2016). For its part, NODDI has shown higher
sensitivity for glioma grade differentiation than other diffusion sequences (Vellmer et al., 2018). Zhao
et al. (Zhao et al., 2018) recently reported that NODDI in combination with patient age can predict
glioma grade with a sensitivity and specificity of 92% and 89%, respectively. Despite these promising
findings, the potential of NODDI as a predictor of patient cognitive recovery has not yet been explored
in the literature.
We additionally found evidence in support of the hypothesis that tumours have both local and long-
distant effects. Neurite Density was decreased not only within the tumour mask, but also beyond its
boundary, being particularly disrupted in larger tumours. Thus, despite peri-tumoral regions being
identified as non-affected by the experienced neurosurgeon that delineated the mask, and by the
semi-automatic segmentation procedure, these regions had decreased neurite density compared with
the contralateral hemisphere. Long-distance tumour effects have been observed in regard to
functional connectivity (Ghinda et al., 2018) and functional complexity (Hart et al., 2019). Disrupted
white matter integrity has been reported in DTI studies that show decreased Fractional Anisotropy in
peritumoural regions (Holly et al., 2017) and white matter tracts (Miller et al., 2012). Interestingly,
Masjoodi et al. (2019) reported that Fractional Anisotropy, but not NODDI have been found to
distinguish edematous White Matter fibres.
Extratumoural Neurite Density disruption had a differential impact on cognition depending on the
affected brain network. We found that preoperative Neurite Density within the Frontoparietal
network was associated with memory and attention recovery after surgery. Frontoparietal Network
has a central role in cognitive control and network adaptability that is made possible by flexible, highly-
connected regions (i.e. hubs) that shift more rapidly than other networks across a variety of task states
(Cole et al., 2013). LaBar et al. (1999) hypothesized that memory and attention are subserved by
neuronal networks that intersect at several Frontoparietal sites. More recently, attention and working
memory are increasingly seen as overlapping constructs modulated by top-down mechanisms
(Gazzaley and Nobre, 2012). Functional MRI studies have consistently reported activation in
frontoparietal regions during memory and attentional tasks (Borst and Anderson, 2013; Huang et al.,
2013). Hubs and the topological efficiency of the Frontoparietal Network have been associated with
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
alerting and executive control subfunctions of attention (Markett et al., 2014) and memory (Sala-
Llonch et al., 2014). Overall this suggests that increased Neurite Density (pre and post-operatively)
within the Frontoparietal Network is a protective factor in memory and attention recovery that is
mediated by the prominent role of brain regions implicated in these cognitive domains. Moreover,
when follow-up NODDI images were included in the analyses, we found a strong association with the
DMN, Frontoparietal and Attention Networks. The DMN is affected by allocation of attentional and
memory resources to the task-relevant region due to task demands (Koshino et al., 2014; Mayer et al.,
2010). The existing negative correlation between DMN and attentional networks have been
traditionally interpreted as reflective competing functions. However, this anti-correlation exhibits
substantial variability across time and is coordinated with broader dynamics involving the
Frontoparietal Network (Dixon et al., 2017). From the connectomic perspective, cognitive
performance of brain tumour patients has been associated with hub-related structural connectivity
within the DMN and Frontoparietal Network in hemispheres contralateral to the tumour (Douw et al.,
2019). By using connectomic metrics derived from both DTI and fMRI, Liu et al. (2016) achieved a 75%
accuracy when predicting survival of high-grade glioma patients. However, further research is needed
to better understand the mechanisms by which low- and high-grade glioma, and treatment disruption
mediate cognitive decline.
Limitations
In the present work, brain networks were defined using normative data from healthy individuals (Yeo
et al., 2011) and meta-analytic maps (Neurosynth). However, the mechanical and physiological
disturbances induced by the tumour and the surgical intervention presumably have a major impact on
the spatial pattern of brain networks. By using brain network templates and meta-analytic maps in
standard space we neglected the potential spatial shift of brain functioning in our participants. Thus,
our results should be interpreted as the potential of the tumour to disrupt the corresponding healthy
network, not the actual tumour/lesion overlap with each participant’s network.
Some patients had difficulties completing cognitive assessments due to the impact of surgery on their
general condition, consequently missing follow-up data has the risk of attrition bias. Moreover, OCS-
Bridge did not detect longitudinal changes nor inter-subject differences in language or praxis
performance, which raises the question of whether patients had no deficits, or the assessment has
not enough sensitivity to detect them. Therefore, further refinement of cognitive testing techniques
is warranted, with research focused on the sensitivity of different methods to detect subtle subclinical
deficits.
Treatment of the patients was decided solely on clinical criteria. As a consequence, 6 patients had only
a surgical intervention while 11 additionally had different chemo-radiotherapy regimes. Chemo-
radiotherapy has a dose-dependent effect on white matter structural integrity that has been
associated with poor cognitive performance (Chapman et al., 2013; Deprez et al., 2012; Michael
Connor et al., 2017). Unfortunately, the contribution of the different treatment strategies to the
observed brain disruption and cognitive decline cannot be untangled here due to the limited sample
size.
Conclusion
Our findings highlight memory and attention deficits that are associated with tumours overlapping
the DMN, the Attentional Network and Attention-related regions. Using NODDI, we also found that
Neurite Density is decreased beyond tumour boundary and that high values within frontoparietal,
DMN and Attention networks are associated with better memory and attention recovery. Taken
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
together, these results reveal the potential of combining brain network data and novel diffusion
sequences to better understand and predict the impact of brain tumours and surgery on patients’
cognitive recovery.
Acknowledgements
We thank all patients for generous involvement in the study. We also thank to Luca Villa, Rohitashwa
Sinha and Jessica Ingham for their contribution to the study.
Funding
R.R.G was supported by a Guarantors of Brain non-clinical fellowship. The Brain Tumour Charity
funded imaging acquisition.
Competing interests
The authors report no competing interests.
REFERENCES
Ahles, T.A., Root, J.C., Ryan, E.L., 2012. Cancer- and cancer treatment-associated cognitive change:
An update on the state of the science. J. Clin. Oncol. 30, 36753686.
https://doi.org/10.1200/JCO.2012.43.0116
Ajithkumar, T., Price, S., Horan, G., Burke, A., Jefferies, S., 2017. Prevention of radiotherapy-induced
neurocognitive dysfunction in survivors of paediatric brain tumours: the potential role of
modern imaging and radiotherapy techniques. Lancet Oncol. 18, e91e100.
https://doi.org/10.1016/S1470-2045(17)30030-X
Anderson, S.W., Damasio, H., Tranel, D., 1990. Neuropsychological impairments associated with
lesions caused by tumor or stroke. Arch. Neurol. 47, 397405.
Beppu, T., Inoue, T., Shibata, Y., Yamada, N., Kurose, A., Ogasawara, K., Ogawa, A., Kabasawa, H.,
2005. Fractional anisotropy value by diffusion tensor magnetic resonance imaging as a
predictor of cell density and proliferation activity of glioblastomas. Surg. Neurol. 63, 5661.
https://doi.org/10.1016/j.surneu.2004.02.034
Borst, J.P., Anderson, J.R., 2013. Using model-based functional MRI to locate working memory
updates and declarative memory retrievals in the fronto-parietal network. Proc. Natl. Acad. Sci.
U. S. A. 110, 16281633. https://doi.org/10.1073/pnas.1221572110
Bressler, S.L., Menon, V., 2010. Large-scale brain networks in cognition: emerging methods and
principles. Trends Cogn. Sci. https://doi.org/10.1016/j.tics.2010.04.004
Briganti, C., Sestieri, C., Mattei, P.A., Esposito, R., Galzio, R.J., Tartaro, A., Romani, G.L., Caulo, M.,
2012. Reorganization of functional connectivity of the language network in patients with brain
gliomas. Am. J. Neuroradiol. 33, 19831990. https://doi.org/10.3174/ajnr.A3064
Caverzasi, E., Papinutto, N., Castellano, A., Zhu, A.H., Scifo, P., Riva, M., Bello, L., Falini, A., Bharatha,
A., Henry, R.G., 2016. Neurite Orientation Dispersion and Density Imaging Color Maps to
Characterize Brain Diffusion in Neurologic Disorders. J. Neuroimaging 26, 494498.
https://doi.org/10.1111/jon.12359
Chapman, C.H., Nazem-Zadeh, M., Lee, O.E., Schipper, M.J., Tsien, C.I., Lawrence, T.S., Cao, Y., 2013.
Regional Variation in Brain White Matter Diffusion Index Changes following
Chemoradiotherapy: A Prospective Study Using Tract-Based Spatial Statistics. PLoS One 8.
https://doi.org/10.1371/journal.pone.0057768
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Charras, P., Herbet, G., Deverdun, J., de Champfleur, N.M., Duffau, H., Bartolomeo, P., Bonnetblanc,
F., 2015. Functional reorganization of the attentional networks in low-grade glioma patients: A
longitudinal study. Cortex 63, 2741. https://doi.org/10.1016/j.cortex.2014.08.010
Cole, M.W., Reynolds, J.R., Power, J.D., Repovs, G., Anticevic, A., Braver, T.S., 2013. Multi-task
connectivity reveals flexible hubs for adaptive task control. Nat. Neurosci. 16, 13481355.
https://doi.org/10.1038/nn.3470
Colgan, N., Siow, B., OCallaghan, J.M., Harrison, I.F., Wells, J.A., Holmes, H.E., Ismail, O., Richardson,
S., Alexander, D.C., Collins, E.C., Fisher, E.M., Johnson, R., Schwarz, A.J., Ahmed, Z., ONeill, M.J.,
Murray, T.K., Zhang, H., Lythgoe, M.F., 2016. Application of neurite orientation dispersion and
density imaging (NODDI) to a tau pathology model of Alzheimers disease. Neuroimage 125,
739744. https://doi.org/10.1016/j.neuroimage.2015.10.043
DAgata, F., Costa, T., Caroppo, P., Baudino, B., Cauda, F., Manfredi, M., Geminiani, G., Mortara, P.,
Pinessi, L., Castellano, G., Bisi, G., 2013. Multivariate analysis of brain metabolism reveals
chemotherapy effects on prefrontal cerebellar system when related to dorsal attention
network. EJNMMI Res. 3, 19. https://doi.org/10.1186/2191-219X-3-22
De Witt Hamer, P.C., Hendriks, E.J., Mandonnet, E., Barkhof, F., Zwinderman, A.H., Duffau, H., 2013.
Resection Probability Maps for Quality Assessment of Glioma Surgery without Brain Location
Bias. PLoS One 8. https://doi.org/10.1371/journal.pone.0073353
Demeyere, N., Riddoch, M.J., Slavkova, E.D., Bickerton, W.L., Humphreys, G.W., 2015. The Oxford
Cognitive Screen (OCS): Validation of a stroke-specific short cognitive screening tool. Psychol.
Assess. 27, 883894. https://doi.org/10.1037/pas0000082
Deprez, S., Amant, F., Smeets, A., Peeters, R., Leemans, A., Van Hecke, W., Verhoeven, J.S.,
Christiaens, M.R., Vandenberghe, J., Vandenbulcke, M., Sunaert, S., 2012. Longitudinal
assessment of chemotherapy-induced structural changes in cerebral white matter and its
correlation with impaired cognitive functioning. J. Clin. Oncol. 30, 274281.
https://doi.org/10.1200/JCO.2011.36.8571
Dixon, M.L., Andrews-Hanna, J.R., Spreng, R.N., Irving, Z.C., Mills, C., Girn, M., Christoff, K., 2017.
Interactions between the default network and dorsal attention network vary across default
subsystems, time, and cognitive states. Neuroimage 147, 632649.
https://doi.org/10.1016/j.neuroimage.2016.12.073
Douw, L., Klein, M., Fagel, S.S., van den Heuvel, J., Taphoorn, M.J., Aaronson, N.K., Postma, T.J.,
Vandertop, W.P., Mooij, J.J., Boerman, R.H., Beute, G.N., Sluimer, J.D., Slotman, B.J., Reijneveld,
J.C., Heimans, J.J., 2009. Cognitive and radiological effects of radiotherapy in patients with low-
grade glioma: long-term follow-up. Lancet Neurol. 8, 810818. https://doi.org/10.1016/S1474-
4422(09)70204-2
Douw, L., Miller, J.J., Steenwijk, M.D., Stufflebeam, S.M., Gerstner, E.R., 2019. Altered structural hub
connectivity and its clinical relevance in glioma. https://doi.org/10.1101/610618
Duffau, H., Capelle, L., 2004. Preferential brain locations of low-grade gliomas: Comparison with
glioblastomas and review of hypothesis. Cancer 100, 26222626.
https://doi.org/10.1002/cncr.20297
Esposito, R., Mattei, P.A., Briganti, C., Romani, G.L., Tartaro, A., Caulo, M., 2012. Modifications of
default-mode network connectivity in patients with cerebral glioma. PLoS One 7.
https://doi.org/10.1371/journal.pone.0040231
Fan, G.G., Deng, Q.L., Wu, Z.H., Guo, Q.Y., 2006. Usefulness of diffusion/perfusion-weighted MRI in
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
patients with non-enhancing supratentorial brain gliomas: A valuable tool to predict tumour
grading? Br. J. Radiol. 79, 652658. https://doi.org/10.1259/bjr/25349497
Ferroli, P., Broggi, M., Schiavolin, S., Acerbi, F., Bettamio, V., Caldiroli, D., Cusin, A., La Corte, E.,
Leonardi, M., Raggi, A., Schiariti, M., Visintini, S., Franzini, A., Broggi, G., 2015. Predicting
functional impairment in brain tumor surgery: The Big Five and the Milan Complexity Scale.
Neurosurg. Focus 39. https://doi.org/10.3171/2015.9.FOCUS15339
Gazzaley, A., Nobre, A.C., 2012. Top-down modulation: Bridging selective attention and working
memory. Trends Cogn. Sci. 16, 129135. https://doi.org/10.1016/j.tics.2011.11.014
Ghinda, D.C., Wu, J.S., Duncan, N.W., Northoff, G., 2018. How much is enoughCan resting state
fMRI provide a demarcation for neurosurgical resection in glioma? Neurosci. Biobehav. Rev. 84,
245261. https://doi.org/10.1016/j.neubiorev.2017.11.019
Ghumman, S., Fortin, D., Noel-Lamy, M., Cunnane, S.C., Whittingstall, K., 2016. Exploratory study of
the effect of brain tumors on the default mode network. J. Neurooncol. 128, 437444.
https://doi.org/10.1007/s11060-016-2129-6
Goldman-Rakic, P., 1988. Topography Of Cognition: Parallel Distributed Networks In Primate
Association Cortex. Annu. Rev. Neurosci. 11, 137156.
https://doi.org/10.1146/annurev.neuro.11.1.137
Habets, E.J.J., Kloet, A., Walchenbach, R., Vecht, C.J., Klein, M., Taphoorn, M.J.B., 2014. Tumour and
surgery effects on cognitive functioning in high-grade glioma patients. Acta Neurochir. (Wien).
156, 14511459. https://doi.org/10.1007/s00701-014-2115-8
Harris, R.J., Bookheimer, S.Y., Cloughesy, T.F., Kim, H.J., Pope, W.B., Lai, A., Nghiemphu, P.L., Liau,
L.M., Ellingson, B.M., 2014. Altered functional connectivity of the default mode network in
diffuse gliomas measured with pseudo-resting state fMRI. J. Neurooncol. 116, 373379.
https://doi.org/10.1007/s11060-013-1304-2
Hart, M.G., Romero-Garcia, R., Price, S.J., Suckling, J., 2019. Global Effects of Focal Brain Tumors on
Functional Complexity and Network Robustness: A Prospective Cohort Study. Neurosurgery 84,
12011213. https://doi.org/10.1093/neuros/nyy378
Holly, K.S., Barker, B.J., Murcia, D., Bennett, R., Kalakoti, P., Ledbetter, C., Gonzalez-Toledo, E.,
Nanda, A., Sun, H., 2017. High-grade Gliomas Exhibit Higher Peritumoral Fractional Anisotropy
and Lower Mean Diffusivity than Intracranial Metastases. Front. Surg. 4, 110.
https://doi.org/10.3389/fsurg.2017.00018
Huang, Q., Zhang, R., Hu, X., Ding, S., Qian, J., Lei, T., Cao, X., Tao, L., Qian, Z., Liu, H., 2014. Disturbed
small-world networks and neurocognitive function in frontal lobe low-grade glioma patients.
PLoS One 9. https://doi.org/10.1371/journal.pone.0094095
Huang, S., Seidman, L.J., Rossi, S., Ahveninen, J., 2013. Distinct cortical networks activated by
auditory attention and working memory load. Neuroimage 83, 10981108.
https://doi.org/10.1016/j.neuroimage.2013.07.074
Huber, T., Bette, S., Wiestler, B., Gempt, J., Gerhardt, J., Delbridge, C., Barz, M., Meyer, B., Zimmer,
C., Kirschke, J.S., 2016. Fractional Anisotropy Correlates with Overall Survival in Glioblastoma.
World Neurosurg. 95, 525534.e1. https://doi.org/10.1016/j.wneu.2016.08.055
Irie, R., Kamagata, K., Kerever, A., Ueda, R., Yokosawa, S., Otake, Y., Ochi, H., Yoshizawa, H., Hayashi,
A., Tagawa, K., Okazawa, H., Takahashi, K., Sato, K., Hori, M., Arikawa-Hirasawa, E., Aoki, S.,
2018. The relationship between neurite density measured with confocal microscopy in a
cleared mouse brain and metrics obtained from diffusion tensor and diffusion kurtosis imaging.
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Magn. Reson. Med. Sci. 17, 138144. https://doi.org/10.2463/mrms.mp.2017-0031
Jakola, A.S., Myrmel, K.S., Kloster, R., Torp, S.H., Lindal, S., Unsgård, G., Solheim, O., 2012.
Comparison of a strategy favoring early surgical resection vs a strategy favoring watchful
waiting in low-grade gliomas. JAMA 308, 18818. https://doi.org/10.1001/jama.2012.12807
Jespersen, S.N., Bjarkam, C.R., Nyengaard, J.R., Chakravarty, M.M., Hansen, B., Vosegaard, T.,
Østergaard, L., Yablonskiy, D., Nielsen, N.C., Vestergaard-Poulsen, P., 2010. Neurite density
from magnetic resonance diffusion measurements at ultrahigh field: Comparison with light
microscopy and electron microscopy. Neuroimage 49, 205216.
https://doi.org/10.1016/j.neuroimage.2009.08.053
Just, N., 2014. Improving tumour heterogeneity MRI assessment with histograms. Br. J. Cancer 111,
22052213. https://doi.org/10.1038/bjc.2014.512
Kamagata, K., Hatano, T., Okuzumi, A., Motoi, Y., Abe, O., Shimoji, K., Kamiya, K., Suzuki, M., Hori,
M., Kumamaru, K.K., Hattori, N., Aoki, S., 2016. Neurite orientation dispersion and density
imaging in the substantia nigra in idiopathic Parkinson disease. Eur. Radiol. 26, 25672577.
https://doi.org/10.1007/s00330-015-4066-8
Kevin S. LaBar, Darren R. Gitelman, Todd B. Parrish, and M.-M.M., 1999. Neuroanatomic Overlap of
Working Memory and Spatial Attention Networks.pdf 704, NeuroImage 10, 695704.
Kinoshita, M., Hashimoto, N., Goto, T., Kagawa, N., Kishima, H., Izumoto, S., Tanaka, H., Fujita, N.,
Yoshimine, T., 2008. Fractional anisotropy and tumor cell density of the tumor core show
positive correlation in diffusion tensor magnetic resonance imaging of malignant brain tumors.
Neuroimage 43, 2935. https://doi.org/10.1016/j.neuroimage.2008.06.041
Klein, M., Heimans, J.J., Aaronson, N.K., Van Der Ploeg, H.M., Grit, J., Muller, M., Postma, T.J., Mooij,
J.J., Boerman, R.H., Beute, G.N., Ossenkoppele, G.J., Van Imhoff, G.W., Dekker, A.W., Jolles, J.,
Slotman, B.J., Struikmans, H., Taphoorn, M.J.B., 2002. Effect of radiotherapy and other
treatment-related factors on mid-term to long-term cognitive sequelae in low-grade gliomas: A
comparative study. Lancet 360, 13611368. https://doi.org/10.1016/S0140-6736(02)11398-5
Klein, M., Postma, T.J., Taphoorn, M.J.B., Aaronson, N.K., Vandertop, W.P., Muller, M., Van Der
Ploeg, H.M., Heimans, J.J., 2003. The prognostic value of cognitive functioning in the survival of
patients with high-grade glioma. Neurology 61, 17961798.
https://doi.org/10.1212/01.WNL.0000098892.33018.4C
Kong, A.P.H., Lam, P.H.P., Ho, D.W.L., Lau, J.K., Humphreys, G.W., Riddoch, J., Weekes, B., 2016. The
Hong Kong version of the Oxford Cognitive Screen (HK-OCS): validation study for Cantonese-
speaking chronic stroke survivors. Aging, Neuropsychol. Cogn. 23, 530548.
https://doi.org/10.1080/13825585.2015.1127321
Koshino, H., Minamoto, T., Yaoi, K., Osaka, M., Osaka, N., 2014. Coactivation of the default mode
network regions and working memory network regions during task preparation. Sci. Rep. 4, 34
39. https://doi.org/10.1038/srep05954
Larjavaara, S., Mäntylä, R., Salminen, T., Haapasalo, H., Raitanen, J., Jääskeläinen, J., Auvinen, A.,
2007. Incidence of gliomas by anatomic location. Neuro. Oncol. 9, 319325.
https://doi.org/10.1215/15228517-2007-016
Li, C., Wang, S., Yan, J.-L., Piper, R.J., Liu, H., Torheim, T., Kim, H., Zou, J., Boonzaier, N.R., Sinha, R.,
Matys, T., Markowetz, F., Price, S.J., 2018. Intratumoral Heterogeneity of Glioblastoma
Infiltration Revealed by Joint Histogram Analysis of Diffusion Tensor Imaging. Neurosurgery 0,
111. https://doi.org/10.1093/neuros/nyy388
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Liu, L., Zhang, H., Rekik, I., Chen, X., Wang, Q., Shen, D., 2016. Outcome prediction for patient with
high-grade gliomas from brain functional and structural networks, in: Lecture Notes in
Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes
in Bioinformatics). pp. 2634. https://doi.org/10.1007/978-3-319-46723-8_4
Maesawa, S., Bagarinao, E., Fujii, M., Futamura, M., Motomura, K., Watanabe, H., Mori, D., Sobue,
G., Wakabayashi, T., 2015. Evaluation of resting state networks in patients with gliomas:
Connectivity changes in the unaffected side and its relation to cognitive function. PLoS One 10,
113. https://doi.org/10.1371/journal.pone.0118072
Markett, S., Reuter, M., Montag, C., Voigt, G., Lachmann, B., Rudorf, S., Elger, C.E., Weber, B., 2014.
Assessing the function of the fronto-parietal attention network: Insights from resting-state
fMRI and the attentional network test. Hum. Brain Mapp. 35, 17001709.
https://doi.org/10.1002/hbm.22285
Masjoodi, S., Hashemi, H., Oghabian, M.A., Sharifi, G., 2018. Differentiation of edematous, tumoral
and normal areas of brain using diffusion tensor and neurite orientation dispersion and density
imaging. J. Biomed. Phys. Eng. 8, 251260. https://doi.org/10.31661/jbpe.v0i0.874
Mayer, J.S., Roebroeck, A., Maurer, K., Linden, D.E.J., 2010. Specialization in the default mode: Task-
induced brain deactivations dissociate between visual working memory and attention. Hum.
Brain Mapp. 31, 126139. https://doi.org/10.1002/hbm.20850
Meyers, C.A., Smith, J.A., Bezjak, A., Mehta, M.P., Liebmann, J., Illidge, T., Kunkler, I., Caudrelier,
J.M., Eisenberg, P.D., Meerwaldt, J., Siemers, R., Carrie, C., Gaspar, L.E., Curran, W., Phan, S.C.,
Miller, R.A., Renschler, M.F., 2004. Neurocognitive function and progression in patients with
brain metastases treated with whole-brain radiation and motexafin gadolinium: Results of a
randomized phase III trial. J. Clin. Oncol. 22, 157165.
https://doi.org/10.1200/JCO.2004.05.128
Michael Connor, Karunamuni, R., McDonald, C., Seibert, T., White, N., Vitali Moiseenko, Bartsch, H.,
Farid, N., Kuperman, J., Krishnan, A., Dale, A., Hattangadi-Gluth, J.A., 2017. Regional
susceptibility to dose-dependent white matter damage after brain radiotherapy. Radiother.
Oncol. 123, 209217. https://doi.org/10.1016/j.radonc.2016.10.003.Dose-Dependent
Miller, P., Coope, D., Thompson, G., Jackson, A., Herholz, K., 2012. Quantitative evaluation of white
matter tract DTI parameter changes in gliomas using nonlinear registration. Neuroimage 60,
23092315. https://doi.org/10.1016/j.neuroimage.2012.02.033
Mohsen, L.A., Shi, V., Jena, R., Gillard, J.H., Price, S.J., 2013. Diffusion tensor invasive phenotypes can
predict progression-free survival in glioblastomas. Br. J. Neurosurg. 27, 436441.
https://doi.org/10.3109/02688697.2013.771136
Otten, M.L., Mikell, C.B., Youngerman, B.E., Liston, C., Sisti, M.B., Bruce, J.N., Small, S.A., McKhann,
G.M., 2012. Motor deficits correlate with resting state motor network connectivity in patients
with brain tumours. Brain 135, 10171026. https://doi.org/10.1093/brain/aws041
Peng, S.L., Chen, C.F., Liu, H.L., Lui, C.C., Huang, Y.J., Lee, T.H., Chang, C.C., Wang, F.N., 2013. Analysis
of parametric histogram from dynamic contrast-enhanced MRI: Application in evaluating brain
tumor response to radiotherapy. NMR Biomed. 26, 443450.
https://doi.org/10.1002/nbm.2882
Qu, J., Qin, L., Cheng, S., Leung, K., Li, X., Li, H., Dai, J., Jiang, T., Akgoz, A., Seethamraju, R., Wang, Q.,
Rahman, R., Li, S., Ai, L., Jiang, T., Young, G.S., 2016. Residual low ADC and high FA at the
resection margin correlate with poor chemoradiation response and overall survival in high-
grade glioma patients. Eur. J. Radiol. 85, 657664. https://doi.org/10.1016/j.ejrad.2015.12.026
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Roberts, T.P.L., Liu, F., Kassner, A., Mori, S., Guha, A., 2005. Fiber density index correlates with
reduced fractional anisotropy in white matter of patients with glioblastoma. Am. J.
Neuroradiol. 26, 21832186.
Sagberg, L.M., Drewes, C., Jakola, A.S., Solheim, O., 2017. Accuracy of operating neurosurgeons
prediction of functional levels after intracranial tumor surgery. J. Neurosurg. 126, 11731180.
https://doi.org/10.3171/2016.3.JNS152927
Sala-Llonch, R., Junqué, C., Arenaza-Urquijo, E.M., Vidal-Piñeiro, D., Valls-Pedret, C., Palacios, E.M.,
Domènech, S., Salvà, A., Bargalló, N., Bartrés-Faz, D., 2014. Changes in whole-brain functional
networks and memory performance in aging. Neurobiol. Aging 35, 21932202.
https://doi.org/10.1016/j.neurobiolaging.2014.04.007
Scott, J.N., Brasher, P.M.A., Sevick, R.J., Rewcastle, N.B., Forsyth, P.A., 2002. How often are
nonenhancing supratentorial gliomas malignant? A population study. Neurology 59, 947949.
https://doi.org/10.1212/WNL.59.6.947
Sinha, S., Bastin, M.E., Whittle, I.R., Wardlaw, J.M., 2002. Diffusion tensor MR imaging of high-grade
cerebral gliomas. Am. J. Neuroradiol. 23, 520527.
Talacchi, A., Santini, B., Savazzi, S., Gerosa, M., 2011. Cognitive effects of tumour and surgical
treatment in glioma patients. J. Neurooncol. 103, 541549. https://doi.org/10.1007/s11060-
010-0417-0
Taphoorn, M.J.B., Klein, M., 2004. Review Cognitive deficits in adult patients with brain tumours.
Lancet 31, 159168.
Tucha, O., Smely, C., Preier, M., Lange, K.W., 2000. Cogntive deficits before treatment among
patients with brain tumors. Neurosurgery 47, 324333. https://doi.org/10.1097/00006123-
200008000-00011
van Dellen, E., Douw, L., Hillebrand, A., Ris-Hilgersom, I., Schoonheim, M., Baayen, J., de Witt Hamer,
P., Velis, D., Klein, M., Heimans, J., Stam, C., 2012. MEG Network Differences between Low- and
High-Grade Glioma Related to Epilepsy and Cognition. PLoS One.
Vellmer, S., Tonoyan, A.S., Suter, D., Pronin, I.N., Maximov, I.I., 2018. Validation of DWI pre-
processing procedures for reliable differentiation between human brain gliomas. Z. Med. Phys.
28, 1424. https://doi.org/10.1016/j.zemedi.2017.04.005
Wang, N., Zhang, J., Cofer, G., Qi, Y., Anderson, R.J., White, L.E., Allan Johnson, G., 2019. Neurite
orientation dispersion and density imaging of mouse brain microstructure. Brain Struct. Funct.
224, 17971813. https://doi.org/10.1007/s00429-019-01877-x
Wefel, J.S., Noll, K.R., Rao, G., Cahill, D.P., 2016. Neurocognitive function varies by IDH1 genetic
mutation status in patients with malignant glioma prior to surgical resection. Neuro. Oncol. 18,
16561663. https://doi.org/10.1093/neuonc/now165
Wefel, J.S., Schagen, S.B., 2012. Chemotherapy-related cognitive dysfunction. Curr. Neurol. Neurosci.
Rep. https://doi.org/10.1007/s11910-012-0264-9
White, M.L., Zhang, Y., Yu, F., Jaffar Kazmi, S.A., 2011. Diffusion tensor MR imaging of cerebral
gliomas: Evaluating fractional anisotropy characteristics. Am. J. Neuroradiol. 32, 374381.
https://doi.org/10.3174/ajnr.A2267
Xu, H., Ding, S., Hu, X., Yang, K., Xiao, C., Zou, Y., Chen, Y., Tao, L., Liu, H., Qian, Z., 2013. Reduced
efficiency of functional brain network underlying intellectual decline in patients with low-grade
glioma. Neurosci. Lett. 543, 2731. https://doi.org/10.1016/j.neulet.2013.02.062
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
Brain network disruption predicts memory and attention deficits after surgical resection of glioma
Romero-Garcia et al (2019)
Yeo, B.T., Krienen, F.M., Sepulcre, J., Sabuncu, M.R., Lashkari, D., Hollinshead, M., Roffman, J.L.,
Smoller, J.W., Zöllei, L., Polimeni, J.R., Fisch, B., Liu, H., Buckner, R.L., 2011. The organization of
the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106,
11251165. https://doi.org/10.1152/jn.00338.2011
Yordanova, Y.N., Moritz-Gasser, S., Duffau, H., 2011. Awake surgery for WHO Grade II gliomas within
noneloquent areas in the left dominant hemisphere: toward a supratotal resection. Clinical
article. J. Neurosurg. 115, 2329. https://doi.org/10.3171/2011.3.JNS101333
Zhang, H., Schneider, T., Wheeler-Kingshott, C.A., Alexander, D.C., 2012. NODDI: Practical in vivo
neurite orientation dispersion and density imaging of the human brain. Neuroimage 61, 1000
1016. https://doi.org/10.1016/j.neuroimage.2012.03.072
Zhao, J., Li, J. bin, Wang, J. yan, Wang, Y. liang, Liu, D. wei, Li, X. bei, Song, Y. kun, Tian, Y. su, Yan, X.,
Li, Z. hao, He, S. fu, Huang, X. long, Jiang, L., Yang, Z. yun, Chu, J. ping, 2018. Quantitative
analysis of neurite orientation dispersion and density imaging in grading gliomas and detecting
IDH-1 gene mutation status. NeuroImage Clin. 19, 174181.
https://doi.org/10.1016/j.nicl.2018.04.011
.CC-BY-NC-ND 4.0 International licenseIt is made available under a
is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.was not peer-reviewed) (whichThe copyright holder for this preprint . http://dx.doi.org/10.1101/19008581doi: medRxiv preprint first posted online Oct. 8, 2019 ;
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Advanced biophysical models like neurite orientation dispersion and density imaging (NODDI) have been developed to estimate the microstructural complexity of voxels enriched in dendrites and axons for both in vivo and ex vivo studies. NODDI metrics derived from high spatial and angular resolution diffusion MRI using the fixed mouse brain as a reference template have not yet been reported due in part to the extremely long scan time required. In this study, we modified the three-dimensional diffusion-weighted spin-echo pulse sequence for multi-shell and undersampling acquisition to reduce the scan time. This allowed us to acquire several exhaustive datasets that would otherwise not be attainable. NODDI metrics were derived from a complex 8-shell diffusion (1000–8000 s/mm²) dataset with 384 diffusion gradient-encoding directions at 50 µm isotropic resolution. These provided a foundation for exploration of tradeoffs among acquisition parameters. A three-shell acquisition strategy covering low, medium, and high b values with at least angular resolution of 64 is essential for ex vivo NODDI experiments. The good agreement between neurite density index (NDI) and the orientation dispersion index (ODI) with the subsequent histochemical analysis of myelin and neuronal density highlights that NODDI could provide new insight into the microstructure of the brain. Furthermore, we found that NDI is sensitive to microstructural variations in the corpus callosum using a well-established demyelination cuprizone model. The study lays the ground work for developing protocols for routine use of high-resolution NODDI method in characterizing brain microstructure in mouse models.
Article
Full-text available
Background: Presurigical planning for glioma tumor resection and radiotherapy treatment require proper delineation of tumoral and peritumoral areas of brain. Diffusion tensor imaging (DTI) is the most common mathematical model applied for diffusion weighted MRI data. Neurite orientation dispersion and density imaging (NODDI) is another mathematical model for DWI data modeling. Objective: We studied whether extracted parameters of DTI, and NODDI models can be used to differentiate between edematous, tumoral, and normal areas in brain white matter (WM). Material and methods: 12 patients with peritumoral edema underwent 3T multi-shell diffusion imaging with b-values of 1000 and 2000 smm-2 in 30 and 64 gradient directions, respectively. We fitted DTI and NODDI to data in manually drawn regions of interest and used their derived parameters to characterize edematous, tumoral and normal brain areas. Results: We found that DTI parameters fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD) and radial diffusivity (RD) all significantly differentiated edematous from contralateral normal brain WM (p<0.005). However, only FA was found to distinguish between edematous WM fibers and tumor invaded fibers (p = 0.001). Among NODDI parameters, the intracellular volume fraction (ficvf) had the best distinguishing power with (p = 0.001) compared with the isotropic volume fraction (fiso), the orientation dispersion index (odi), and the concentration parameter of Watson distribution (κ), while comparing fibers inside normal, tumoral, and edematous areas. Conclusion: The combination of two diffusion based methods, i.e. DTI and NODDI parameters can distinguish and characterize WM fibers involved in edematus, tumoral, and normal brain areas with reasonable confidence. Further studies will be required to improve the detectability of WM fibers inside the solid tumor if they hypothetically exist in tumoral parenchyma.
Article
Full-text available
Background: Neurosurgical management of brain tumors has entered a paradigm of supramarginal resections that demands thorough understanding of peritumoral functional effects. Historically, the effects of tumors have been believed to be local, and long-range effects have not been considered. Objective: To test the hypothesis that tumors affect the brain globally, producing long-range gradients in cortical function. Methods: Resting-state functional magnetic resonance imaging (fMRI) data were acquired from 11 participants with glioblastoma and split into discovery and validation datasets in a single-center prospective cohort study. Fractal complexity was computed with a wavelet-based estimator of the Hurst exponent. Distance-related effects of the tumors were tested with a tumor mask-dilation technique and parcellation of the underlying Hurst maps. Results: Fractal complexity demonstrates a penumbra of suppression in the peritumoral region. At a global level, as distance from the tumor increases, this initial suppression is balanced by a subsequent overactivity before finally normalizing. These effects were best fit by a quadratic model and were consistent across different network construction pipelines. The Hurst exponent was correlated with graph theory measures of centrality including network robustness, but graph theory measures did not demonstrate distance-dependent effects. Conclusion: This work provides evidence supporting the theory that focal brain tumors produce long-range gradients in function. Consequently, the effects of focal lesions need to be interpreted in terms of the global changes on functional complexity and network architecture rather than purely in terms of functional localization. Determining whether peritumoral changes represent potential plasticity may facilitate extended resection of tumors without functional cost.
Article
Full-text available
Background and purpose Neurite orientation dispersion and density imaging (NODDI) is a new diffusion MRI technique that has rarely been applied for glioma grading. The purpose of this study was to quantitatively evaluate the diagnostic efficiency of NODDI in tumour parenchyma (TP) and peritumoural area (PT) for grading gliomas and detecting isocitrate dehydrogenase-1 (IDH-1) mutation status. Methods Forty-two patients (male: 23, female: 19, mean age: 44.5 y) were recruited and underwent whole brain NODDI examination. Intracellular volume fraction (icvf) and orientation dispersion index (ODI) maps were derived. Three ROIs were manually placed on TP and PT regions for each case. The corresponding average values of icvf and ODI were calculated, and their diagnostic efficiency was assessed. Results Tumours with high icvfTP (≥0.306) and low icvfPT (≤0.331) were more likely to be high-grade gliomas (HGGs), while lesions with low icvfTP (<0.306) and high icvfPT (>0.331) were prone to be low-grade gliomas (LGGs) (P < 0.001). A multivariate logistic regression model including patient age and icvf values in TP and PT regions most accurately predicted glioma grade (AUC = 0.92, P < 0.001), with a sensitivity and specificity of 92% and 89%, respectively. However, no significant differences were found in NODDI metrics for differentiating IDH-1 mutation status. Conclusions The quantitative NODDI metrics in the TP and PT regions are highly valuable for glioma grading. A multivariate logistic regression model using the patient age and the icvf values in TP and PT regions showed very high predictive power. However, the utility of NODDI metrics for detecting IDH-1 mutation status has not been fully explored, as a larger sample size may be necessary to uncover benefits.
Article
Full-text available
Purpose: Diffusional kurtosis imaging (DKI) enables sensitive measurement of tissue microstructure by quantifying the non-Gaussian diffusion of water. Although DKI is widely applied in many situations, histological correlation with DKI analysis is lacking. The purpose of this study was to determine the relationship between DKI metrics and neurite density measured using confocal microscopy of a cleared mouse brain. Methods: One thy-1 yellow fluorescent protein 16 mouse was deeply anesthetized and perfusion fixation was performed. The brain was carefully dissected out and whole-brain MRI was performed using a 7T animal MRI system. DKI and diffusion tensor imaging (DTI) data were obtained. After the MRI scan, brain sections were prepared and then cleared using aminoalcohols (CUBIC). Confocal microscopy was performed using a two-photon confocal microscope with a laser. Forty-eight ROIs were set on the caudate putamen, seven ROIs on the anterior commissure, and seven ROIs on the ventral hippocampal commissure on the confocal microscopic image and a corresponding MR image. In each ROI, histological neurite density and the metrics of DKI and DTI were calculated. The correlations between diffusion metrics and neurite density were analyzed using Pearson correlation coefficient analysis. Results: Mean kurtosis (MK) (P = 5.2 × 10-9, r = 0.73) and radial kurtosis (P = 2.3 × 10-9, r = 0.74) strongly correlated with neurite density in the caudate putamen. The correlation between fractional anisotropy (FA) and neurite density was moderate (P = 0.0030, r = 0.42). In the anterior commissure and the ventral hippocampal commissure, neurite density and FA are very strongly correlated (P = 1.3 × 10-5, r = 0.90). MK in these areas were very high value and showed no significant correlation (P = 0.48). Conclusion: DKI accurately reflected neurite density in the area with crossing fibers, potentially allowing evaluation of complex microstructures.
Article
Full-text available
Differentiating high-grade gliomas and intracranial metastases through non-invasive imaging has been challenging. Here, we retrospectively compared both intratumoral and peritumoral fractional anisotropy (FA), mean diffusivity (MD), and fluid-attenuated inversion recovery (FLAIR) measurements between high-grade gliomas and metastases. Two methods were utilized to select peritumoral region of interest (ROI). The first method utilized the manual placement of four ROIs adjacent to the lesion. The second method utilized a semiautomated and proprietary MATLAB script to generate an ROI encompassing the entire tumor. The average peritumoral FA, MD, and FLAIR values were determined within the ROIs for both methods. Forty patients with high-grade gliomas and 44 with metastases were enrolled in this study. Thirty-five patients with high-grade glioma and 30 patients with metastases had FLAIR images. There was no significant difference in age, gender, or race between the two patient groups. The high-grade gliomas had a significantly higher tumor-to-brain area ratio compared to the metastases. There were no differences in average intratumoral FA, MD, and FLAIR values between the two groups. Both the manual sample method and the semiautomated peritumoral ring method resulted in significantly higher peritumoral FA and significantly lower peritumoral MD in high-grade gliomas compared to metastases (p < 0.05). No significant difference was found in FLAIR values between the two groups peritumorally. Receiver operating curve analysis revealed FA to be a more sensitive and specific metric to differentiate high-grade gliomas and metastases than MD. The differences in the peritumoral FA and MD values between high-grade gliomas and metastases seemed due to the infiltration of glioma to the surrounding brain parenchyma.
Article
Background: Glioblastoma is a heterogeneous disease characterized by its infiltrative growth, rendering complete resection impossible. Diffusion tensor imaging (DTI) shows potential in detecting tumor infiltration by reflecting microstructure disruption. Objective: To explore the heterogeneity of glioblastoma infiltration using joint histogram analysis of DTI, to investigate the incremental prognostic value of infiltrative patterns over clinical factors, and to identify specific subregions for targeted therapy. Methods: A total of 115 primary glioblastoma patients were prospectively recruited for surgery and preoperative magnetic resonance imaging. The joint histograms of decomposed anisotropic and isotropic components of DTI were constructed in both contrast-enhancing and nonenhancing tumor regions. Patient survival was analyzed with joint histogram features and relevant clinical factors. The incremental prognostic values of histogram features were assessed using receiver operating characteristic curve analysis. The correlation between the proportion of diffusion patterns and tumor progression rate was tested using Pearson correlation. Results: We found that joint histogram features were associated with patient survival and improved survival model performance. Specifically, the proportion of nonenhancing tumor subregion with decreased isotropic diffusion and increased anisotropic diffusion was correlated with tumor progression rate (P = .010, r = 0.35), affected progression-free survival (hazard ratio = 1.08, P < .001), and overall survival (hazard ratio = 1.36, P < .001) in multivariate models. Conclusion: Joint histogram features of DTI showed incremental prognostic values over clinical factors for glioblastoma patients. The nonenhancing tumor subregion with decreased isotropic diffusion and increased anisotropic diffusion may indicate a more infiltrative habitat and potential treatment target.
Article
This study represents a systematic review of the insights provided by resting state functional MRI (rs-fMRI) use in the glioma population. Following PRISMA guidelines, 45 studies were included in the review and were classified in glioma-related neuronal changes (n=28) and eloquent area localization (n=17). Despite the heterogeneous nature of the studies, there is considerable evidence of diffuse functional reorganization occurring in the setting of gliomas with local and interhemispheric functional connectivity alterations involving different functional networks. The studies showed evidence of decreased long distance functional connectivity and increased global local efficiency occurring in the setting of gliomas. The tumour grade seems to correlate with distinct functional connectivity changes. Overall, there is a potential clinical utility of rs-fMRI for identifying the functional brain network disruptions occurring in the setting of gliomas. Further studies utilizing standardized analytical methods are required to elucidate the mechanism through which gliomas induce global changes in brain connectivity.
Article
Diffusion magnetic resonance imaging (dMRI) is a powerful tool in clinical applications, in particular, in oncology screening. dMRI demonstrated its benefit and efficiency in the localisation and detection of different types of human brain tumours. Clinical dMRI data suffer from multiple artefacts such as motion and eddy-current distortions, contamination by noise, outliers etc. In order to increase the image quality of the derived diffusion scalar metrics and the accuracy of the subsequent data analysis, various pre-processing approaches are actively developed and used. In the present work we assess the effect of different pre-processing procedures such as a noise correction, different smoothing algorithms and spatial interpolation of raw diffusion data, with respect to the accuracy of brain glioma differentiation. As a set of sensitive biomarkers of the glioma malignancy grades we chose the derived scalar metrics from diffusion and kurtosis tensor imaging as well as the neurite orientation dispersion and density imaging (NODDI) biophysical model. Our results show that the application of noise correction, anisotropic diffusion filtering, and cubic-order spline interpolation resulted in the highest sensitivity and specificity for glioma malignancy grading. Thus, these pre-processing steps are recommended for the statistical analysis in brain tumour studies.