Introduction 1 2 3 5 1 4 6 7 8 10 11 13 14 15 16 17 18 19 13 16 19 20 21 22 25 22 23 24 25 Since higher cognitive functions are presumed to depend on the integrated activity of several specialized brain areas, it is suggested that neurocognitive deficits may have a stronger correlation with diffuse alterations in resting state brain oscillatory activity than focal abnormalities in the brain tumor population. In the present MEG-study, power spectral analysis of oscillatory brain activity was used to compare resting state brain oscillatory activity in both LGG patients and healthy controls. We hypothesize that (1) LGG patients show, in addition to the local abnormalities at the tumor site, global slowing of the resting state brain activity compared to healthy controls which will vary between patients with a tumor in the left or right hemisphere (2) changes in resting state brain oscillatory activity reflect an intermediate level between the impact of tumor and tumor-related treatment on the one hand (‘input’) and the neurocognitive deficits (‘output’) on the other hand and that diffuse slowing is correlated with higher neurocognitive dysfunction. Materials and methods Patients and controls Twenty-three LGG patients were asked to participate in this study. Patients were eligible if: (a) they had a suspected or histologically confirmed LGG; (b) there was no radiological (confirmed by MR or CT scan) and/or clinical tumor progression in the previous 6 months; (c) they did not use medication possibly interfering with neurocognitive function, other than anti-epileptic drugs (AEDs). Patients were recruited from the VU University Medical Center (VUmc) and the Academic Medical Center (AMC), both tertiary referral centres in Amsterdam for brain tumor patients, after the institutional ethical review boards of both participating hospitals approved the study protocol. Relatives of the patients were asked to participate as healthy controls. Healthy controls were eligible if they: (a) did not suffer from any neurological disease; (b) did not use any medication that might influence cognitive function. For patients who could not provide a healthy control participant, VU University Medical Center staff members were included. Magnetoencephalography 26 For this study, 149 of the 151 channels could be used. MEG recordings were converted to ASCII files. From these ASCII files four artefact free epochs of 13 s per subject (4096 samples) were carefully selected by visual analysis by one of the authors (IB). Magnetic field frequencies ranging from 0.5 to 80 Hz were recorded. The MEG data were digitally filtered off-line in the following frequency bands: delta (0.5–4 Hz), theta (4–8 Hz), lower alpha (8–10 Hz), upper alpha (10–13 Hz), beta (13–30 Hz), gamma (30–50 Hz). 1 Fig. 1 Distribution of MEG regions  Neurocognitive assessment 1 z Table 1 Description of neuropsychological test battery 27 This test provides a measure of psychomotor performance that is relatively unaffected by intellectual prowess and is suitable for groups with an age range exceeding 60 years. The number of items written down in 90 s is registered, as in the decrease in performance when graphomotor speed is involved 27 This version of the Rey Auditory Verbal Learning Test calls for various aspects of verbal learning and recall. Measures used for analysis are memory performance on trial 1 as indicator of immediate recall, total recall after five trials, delayed recall and recognition after 20 min as indicators of memory consolidation into long-term memory, and a delta score as a measure of learning capacity 27 This test is a selective attention task aiming at measuring interference susceptibility and consists of three subtasks with increasing task complexity 28 A simple task requiring the generation of words from semantic categories (animals) within a limited time 29 This test, which has two conditions of complexity, predominately measures functions associated with executive function, especially visual scanning and conceptual tracking. The motor component of this task is measured by three dummy conditions in which no cognitive capacity except for graphomotor speed is required Memory comparison test (MCT)  Selective attention, mental concentration, memory and information processing 30 z 31 Statistical analysis U z z U ρ z Results Patient characteristics From the initial patient group, six patients were excluded, four patients due to metal artefacts on the MEG measurements, one due to severe epileptic seizures, and one due to tumor progression at the time of registration. The final analyses were performed on a sample of 17 patients and 17 matched healthy control participants. p p p Eleven of those 16 patients underwent debulking, whereas 3 patients underwent a stereotactic biopsy and another two patients had an open biopsy. Of the 16 patients with a histologically confirmed LGG, the pathological diagnosis was grade II astrocytoma in ten patients, oligodendroglioma grade II in four patients and oligoastrocytoma grade II in two patients. 2 Table 2 Tumor lateralization and localization Left hemisphere Right hemisphere Tumor location No. of patients Tumor location No. of patients Left frontal 4 Right frontal 2 Left parietal 3 Right frontoparietal 3 Left temporal 3 Right insular region 1 Left parieto-occipital  1   Total 11 Total 6 In the patient group, all but one patient used AED mono- or poly-therapy. Six of the 16 patients on AED were free of seizures, while the other 10 patients were still having seizures. Differences between patient and healthy control group Neurocognitive functioning p p p p 2 Fig. 2 z Note:  p p  Spectral analysis Global spectral analysis.  U p Spectral analysis of frequency bands within each MEG region.  U p U p U p 3 3 Table 3 Significant differences in relative power between patients and controls per frequency band and accompanying statistics  Patients Controls p  M SD M SD Theta Left parietal 0.131 0.053 0.092 0.026 0.014 Gamma Left central 0.064 0.027 0.095 0.041 0.024 Right central 0.066 0.033 0.102 0.047 0.024 Left frontal 0.048 0.027 0.088 0.043 0.005 Right frontal 0.050 0.030 0.083 0.037 0.005 Significant higher relative power is depicted in bold and in italics Fig. 3 The significant differences in relative power between the patient group and the healthy controls within the different frequency bands. Green area: significant higher relative power in the patient group compared to the healthy controls. Red area: significant lower relative power in the patient group compared to the healthy controls  Influence of tumor lateralization on relative power.  U p p p p p p p p p p U p p p p p p 4 4 Table 4 Significant differences in relative power between patients with a tumor in the left or right hemisphere and controls per frequency band and accompanying statistics  Patients Controls p  M SD M SD Left hemisphere Theta Left central 0.141 0.039 0.095 0.018 0.001 Right central 0.133 0.041 0.096 0.022 0.014 Left parietal 0.149 0.057 0.092 0.026 0.004 Right parietal 0.141 0.065 0.089 0.029 0.029 Left temporal 0.107 0.049 0.072 0.019 0.022 Right temporal 0.100 0.043 0.071 0.023 0.027 Gamma Left central 0.063 0.026 0.095 0.041 0.023 Right central 0.069 0.032 0.102 0.047 0.048 Left frontal 0.048 0.027 0.088 0.043 0.011 Right frontal 0.053 0.031 0.083 0.037 0.019 Right hemisphere Delta Right parietal 0.424 0.144 0.303 0.137 0.050 Gamma Left frontal 0.050 0.028 0.088 0.043 0.005 Right frontal 0.045 0.031 0.083 0.037 0.025 Right temporal 0.042 0.022 0.070 0.045 0.042 Left occipital 0.040 0.029 0.077 0.032 0.025 Right occipital 0.048 0.036 0.099 0.037 0.014 Significant higher relative power is depicted in bold and in italics Fig. 4 Green area: Red area:  32 Associations between patient’s neurocognitive functioning and relative power in distinct MEG regions p p p p p Discussion The primary goal of the present study was to evaluate whether LGG patients show (in addition to the well known MEG slowing around the tumor) diffuse slowing in resting state brain activity. The secondary goal of the study was to investigate whether this slowing is correlated with neurocognitive dysfunction. By means of MEG-registrations we have demonstrated that LGG patients have slowing of the resting state brain activity when compared to healthy controls. The decrease in relative power was mainly found in the gamma frequency band in the bilateral frontocentral MEG regions. Regarding the low frequency bands, an increase in relative power was found in the theta frequency band in the left parietal region. Correlations of neurocognitive functioning with the relative power in the patient population showed clear associations in the lower alpha and theta band, increased slowing correlating with poorer performance. 19 20 16 20 22 22 23 21 24 33 6 34 38 16 13 19 20 Evidently, this study has its limitations. First of all, the patient group is rather small as mentioned before. Secondly, although our patient population consisted of only LGG patients, they did not all receive the same treatment. In our study we hypothesized changes in oscillatory brain activity to be the intermediate between the impact of the tumor and its treatment on the one hand and the neurocognitive deficits as the output on the other hand. This study showed that irrespective of the different treatment options, changes in oscillatory brain activity can be found in brain tumor patients which is associated with neurocognitive function. To get informed on the influence of the different tumor treatments on brain activity and its relation with neurocognitive function is very interesting. A longitudinal study of these effects (e.g. surgery, radiotherapy and chemotherapy) on brain activity is currently under way. The goal of that study is to explore correlations between changes in neurocognitive function and changes in functional brain dynamics during the disease course. Our study is the first to correlate the relative power with neurocognitive functioning in brain tumor patients. The observed correlations were quite strong and showed that an increased activity in the theta and lower alpha band is correlated with impaired executive functioning, information processing and working memory. This is a first step in unraveling the underlying mechanisms of neurocognitive dysfunction in brain tumor patients. MEG power analysis gives us an interesting tool to assess functional alterations in the patient’s brain in the course of disease and to evaluate its relationship with neurocognitive functioning.