Automated Brain Tumor Detection and Segmentation for Treatment Response Assessment Using Amino Acid PET.

Evaluation of metabolic tumor volume (MTV) changes using amino acid PET has become an important tool for response assessment in brain tumor patients. MTV is usually determined by manual or semiautomatic delineation, which is laborious and may be prone to intra- and interobserver variability. The goal of our study was to develop a method for automated MTV segmentation and to evaluate its performance for response assessment in patients with gliomas. Methods: In total, 699 amino acid PET scans using the tracer O-(2-[18F]fluoroethyl)-l-tyrosine (18F-FET) from 555 brain tumor patients at initial diagnosis or during follow-up were retrospectively evaluated (mainly glioma patients, 76%). 18F-FET PET MTVs were segmented semiautomatically by experienced readers. An artificial neural network (no new U-Net) was configured on 476 scans from 399 patients, and the network performance was evaluated on a test dataset including 223 scans from 156 patients. Surface and volumetric Dice similarity coefficients (DSCs) were used to evaluate segmentation quality. Finally, the network was applied to a recently published 18F-FET PET study on response assessment in glioblastoma patients treated with adjuvant temozolomide chemotherapy for a fully automated response assessment in comparison to an experienced physician. Results: In the test dataset, 92% of lesions with increased uptake (n = 189) and 85% of lesions with iso- or hypometabolic uptake (n = 33) were correctly identified (F1 score, 92%). Single lesions with a contiguous uptake had the highest DSC, followed by lesions with heterogeneous, noncontiguous uptake and multifocal lesions (surface DSC: 0.96, 0.93, and 0.81 respectively; volume DSC: 0.83, 0.77, and 0.67, respectively). Change in MTV, as detected by the automated segmentation, was a significant determinant of disease-free and overall survival, in agreement with the physician's assessment. Conclusion: Our deep learning-based 18F-FET PET segmentation allows reliable, robust, and fully automated evaluation of MTV in brain tumor patients and demonstrates clinical value for automated response assessment.

Early treatment response evaluation using FET PET compared to MRI in glioblastoma patients at first progression treated with bevacizumab plus lomustine.

BACKGROUND: The goal of this prospective study was to compare the value of both conventional MRI and O-(2-18F-fluoroethyl)-L-tyrosine (FET) PET for response evaluation in glioblastoma patients treated with bevacizumab plus lomustine (BEV/LOM) at first progression. METHODS: After chemoradiation with concomitant and adjuvant temozolomide, 21 IDH wild-type glioblastoma patients at first progression (age range, 33-75 years; MGMT promoter unmethylated, 81%) were treated with BEV/LOM. Contrast-enhanced MRI and FET-PET scans were performed at baseline and after 8-10 weeks. We obtained FET metabolic tumor volumes (MTV) and tumor/brain ratios. Threshold values of FET-PET parameters for treatment response were established by ROC analyses using the post-progression overall survival (OS) ≤/>9 months as the reference. MRI response assessment was based on RANO criteria. The predictive ability of FET-PET thresholds and MRI changes on early response assessment was evaluated subsequently concerning OS using uni- and multivariate survival estimates. RESULTS: Early treatment response as assessed by RANO criteria was not predictive for an OS>9 months (P = 0.203), whereas relative reductions of all FET-PET parameters significantly predicted an OS>9 months (P < 0.05). The absolute MTV at follow-up enabled the most significant OS prediction (sensitivity, 85%; specificity, 88%; P = 0.001). Patients with an absolute MTV below 5 ml at follow-up survived significantly longer (12 vs. 6 months, P < 0.001), whereas early responders defined by RANO criteria lived only insignificantly longer (9 vs. 6 months; P = 0.072). The absolute MTV at follow-up remained significant in the multivariate survival analysis (P = 0.006). CONCLUSIONS: FET-PET appears to be useful for identifying responders to BEV/LOM early after treatment initiation.

Association between Cortical GABA and Loudness Dependence of Auditory Evoked Potentials (LDAEP) in Humans.

Loudness dependence of auditory evoked potentials (LDAEP) is a widely used EEG-based biomarker for central serotonergic activity. Serotonin has been shown to be associated with different psychiatric disorders such as depression and schizophrenia. Despite its clinical significance, the underlying neurochemical mechanism of this promising marker is not fully understood, and further research is needed to improve its validity. Other neurotransmitters might have a significant impact on this measure. Thus, we assessed the inhibitory action through individual GABA/H20 concentrations and GABA/glutamate ratios by means of magnetic resonance spectroscopy at 3T in healthy subjects. The measurements were assessed in the primary auditory cortex to investigate the association with the LDAEP, whose generators are mainly in the primary auditory cortex. For the first time, this study examines the link between GABAergic neurotransmission and LDAEP, and the data preliminary show that GABA may not contribute to the generation of EEG-based LDAEP.

Detection of remote neuronal reactions in the Thalamus and Hippocampus induced by rat glioma using the PET tracer cis-4-[¹8F]fluoro-D-proline.

After cerebral ischemia or trauma, secondary neurodegeneration may occur in brain regions remote from the lesion. Little is known about the capacity of cerebral gliomas to induce secondary neurodegeneration. A previous study showed that cis-4-[(18)F]fluoro-D-proline (D-cis-[(18)F]FPro) detects secondary reactions of thalamic nuclei after cortical infarction with high sensitivity. Here we investigated the potential of D-cis-[(18)F]FPro to detect neuronal reactions in remote brain areas in the F98 rat glioma model using ex vivo autoradiography. Although the tumor tissue of F98 gliomas showed no significant D-cis-[(18)F]FPro uptake, we observed prominent tracer uptake in 7 of 10 animals in the nuclei of the ipsilateral thalamus, which varied with the specific connectivity with the cortical areas affected by the tumor. In addition, strong D-cis-[(18)F]FPro accumulation was noted in the hippocampal area CA1 in two animals with ipsilateral F98 gliomas involving hippocampal subarea CA3 rostral to that area. Furthermore, focal D-cis-[(18)F]FPro uptake was present in the necrotic center of the tumors. Cis-4-[(18)F]fluoro-D-proline uptake was accompanied by microglial activation in the thalamus, in the hippocampus, and in the necrotic center of the tumors. The data suggest that brain tumors induce secondary neuronal reactions in remote brain areas, which may be detected by positron emission tomography (PET) using D-cis-[(18)F]FPro.

Neurofunctional modulation of brain regions by distinct forms of motor cognition and movement features.

Extrastriate, parietal, and frontal brain regions are differentially involved in distinct kinds of body movements and motor cognition. Using functional magnetic resonance imaging, we investigated the neural mechanisms underlying the observation and mental imagery of meaningful face and limb movements with or without objects. The supplementary motor area was differentially recruited by the mental imagery of movements while there were differential responses of the extrastriate body area (EBA) during the observation conditions. Contrary to most previous reports, the EBA responded to face movements, albeit to a lesser degree than to limb movements. The medial wall of the intraparietal sulcus and adjacent intraparietal cortex was selectively recruited by the processing of meaningful upper limb movements, irrespective of whether these were object-related or not. Besides reach and grasp movements, the intraparietal sulcus may thus be involved in limb gesture processing, that is, in an important aspect of human social communication. We conclude that subregions of a frontal-parietal network differentially interact during the cognitive processing of body movements according to the specific motor-related task at hand and the particular movement features involved.