Rolle von Hyperkonnektivität zwischen Tumor und gesundem Gehirn bei Glioblastompatientinnen und -patienten

Glioblastoma is a systemic brain disease. In recent years, it could be shown that glioblastoma cells integrate into normal neural circuits, even forming synapses with neurons. Furthermore, experimental data showed that increased connectivity between glioblastomas and healthy brain was associated with decreased survival. Here, we used resting-state fMRI (rsfMRI) to reproduce these experimental results with an advanced imaging technique in vivo.

30 patients with glioblastoma were prospectively enrolled. All patients underwent rsfMRI and patient data was compared to a reference cohort of 1000 healthy controls to determine abnormality of functional connectivity on an individual level. Next, we used this data to quantify whether the whole tumor region showed more or less connectivity to the rest of the brain when compared to the corresponding brain region in healthy controls. The results were then correlated with survival data. Furthermore, rsfMRI data analysis - capable navigation software was used in a subset of patients (n = 4) to determine whether there were region – specific connectivity differences within the tumor.

24 patients had tumor progression (median PFS 8.2 months) and 19 patients died (median OS 13.4 months) during the observation period. Univariate cox regression models showed that higher functional connectivity between the tumor and the rest of the brain was associated with lower PFS (p<0.001) and lower OS (p<0.0001). This held true in multivariate models which included potential confounders (age, KPS, tumor volume, resection status, MGMT promoter methylation). Furthermore, connectivity analysis of the tumor showed that there were regions within the tumor which were hyperconnected to the rest of the brain while other areas within the tumor showed no connectivity to the healthy brain.

We found that increased connectivity between the tumor and the rest of the brain was associated with decreased PFS and OS. We therefore speculate that rsfMRI might be a useful tool to determine how extensively glioblastomas have integrated into the neuronal circuitry of the brain. RsfMRI could therefore serve as an imaging correlate for the biological relationships described by the emerging field of cancer neuroscience. Additionally, our results suggest that hyperconnectivity between tumor and healthy brain detected in our cohort could be due to specific hyperconnected regions within the tumor, an observation which could open up new avenues of treatment.