Multi-View Ensemble Classification of Brain Connectivity Images for Neurodegeneration Type Discrimination

Brain connectivity analyses using voxels as features are not robust enough for single-patient classification because of the inter-subject anatomical and functional variability. To construct more robust features, voxels can be aggregated into clusters that are maximally coherent across subjects. Moreover, combining multi-modal neuroimaging and multi-view data integration techniques allows generating multiple independent connectivity features for the same patient. Structural and functional connectivity features were extracted from multi-modal MRI images with a clustering technique, and used for the multi-view classification of different phenotypes of neurodegeneration by an ensemble learning method (random forest). Two different multi-view models (intermediate and late data integration) were trained on, and tested for the classification of, individual whole-brain default-mode network (DMN) and fractional anisotropy (FA) maps, from 41 amyotrophic lateral sclerosis (ALS) patients, 37 Parkinson’s disease (PD) patients and 43 healthy control (HC) subjects. Both multi-view data models exhibited ensemble classification accuracies significantly above chance. In ALS patients, multi-view models exhibited the best performances (intermediate: 82.9%, late: 80.5% correct classification) and were more discriminative than each single-view model. In PD patients and controls, multi-view models’ performances were lower (PD: 59.5%, 62.2%; HC: 56.8%, 59.1%) but higher than at least one single-view model. Training the models only on patients, produced more than 85% patients correctly discriminated as ALS or PD type and maximal performances for multi-view models. These results highlight the potentials of mining complementary information from the integration of multiple data views in the classification of connectivity patterns from multi-modal brain images in the study of neurodegenerative diseases.     

Post-transplantation hepatocellular carcinoma recurrence: Patterns and relation between vascularity and differentiation degree

AIM: To evaluate the relationship between hepatocellular carcinoma(HCC) vascularity and grade; to describe patterns and vascular/histopathological variations of post-transplantation recurrence.METHODS: This retrospective study included 165 patients(143 men, 22 women; median age 56.8 years, range 28-70.4 years) transplanted for HCC who had a follow-up period longer than 2 mo. Pre-transplantation dynamic computed tomography or magnetic resonance examinations were retrospectively reviewed, classifying HCC imaging enhancement pattern into hypervascular and hypovascular based on presence of wash-in during arterial phase. All pathologic reports of the explanted livers were reviewed, collecting data about HCC differentiation degree. The association between imaging vascular pattern and pathological grade was estimated using the Fisher exact test. All follow-up clinical and imaging data were reviewed for evidence of recurrence. Recurrence rate was calculated and imaging features of recurrent tumor were collected, classifying early and late recurrences based on timing(< or ≥ 2 years after transplantation) and intrahepatic, extrahepatic and both intrahepatic and extrahepatic recurrences based onlocation. All intrahepatic recurrences were classified as hypervascular or hypovascular and the differentiation degree was collected where available. The presence of variations in imaging enhancement pattern and pathological grade between the primary tumor and the intrahepatic recurrence was evaluated and the association between imaging and histopatholgical variations was estimated by using the χ2 test. RESULTS: Of the 163 patients with imaging evidence of viable tumor, 156(95.7%) had hypervascular and 7(4.3%) hypovascular HCC. Among the 125 patients with evidence of viable tumor in the explanted liver, 19(15.2%) had grade 1, 56(44.8%) grade 2, 40(32%) grade 3 and 4(3.2%) grade 4 HCC, while the differentiation degree was not assessable for 6 patients(4.8%). A significant association was found between imaging vascularity and pathological grade(P = 0.035). Post-transplantation recurrence rate was 14.55%(24/165). All recurrences occurred in patients who had a hypervascular primary tumor. Three patients(12.5%) experienced late recurrence; the location of the first recurrence was extrahepatic in 14 patients(58.3%), intrahepatic in 7 patients(29.2%) and both intrahepatic and extrahepatic in 3 patients(12.5%). Two patients had a variation in imaging characteristics between the primary HCC(hypervascular) and the intrahepatic recurrent HCC(hypovascular), while 1 patient had a variation of histopathological characteristics(from moderate to poor differentiation), however no association was found between imaging and histopathological variations.CONCLUSION: A correlation was found between HCC grade and vascularity; some degree of variability may exist between the primary and the recurrence imaging/histopathological characteristics, apparently not correlated.     

Rituximab therapy for primary glomerulonephritis: Report on two cases

The evidence in the medical literature on the efficacy and safety of rituximab therapy for primary glomerulonephritis is limited and controversial. We describe two male Caucasian patients with rapidly progressive kidney failure due to primary proliferative glomerulonephritis. Both of them received high-dose intravenous corticosteroids and oral cyclophosphamide with limited benefit. The first patient(hepatitis C virus-negative mixed cryoglobulinemia) underwent plasma-exchange with intravenous immunoglobulins; he showed significant benefit on kidney function(he became dialysis independent with serum creatinine going back to 1.6 mg/d L) after one rituximab pulse even if urinary abnormalities were still present. No improvement in renal function or urinary changes occurred in the second patient. Both these individuals developed sepsis over the follow-up, the first patient died two months after rituximab therapy. This report is in keeping with the occurrence of severe infections after rituximab therapy in patients with renal impairment at baseline and concomitant high-dose steroids.     

Disconnectome of the migraine brain: a “connectopathy” model

BackgroundIn the past decades a plethora of studies has been conducted to explore resting-state functional connectivity (RS-FC) of the brain networks in migraine with conflicting results probably due to the variability and susceptibility of signal fluctuations across the course of RS-FC scan. On the other hand, the structural substrates enabling the functional communications among the brain connectome, characterized by higher stability and reproducibility, have not been widely investigated in migraine by means of graph analysis approach. We hypothesize a rearrangement of the brain connectome with an increase of both strength and density of connections between cortical areas specifically involved in pain perception, processing and modulation in migraine patients. Moreover, such connectome rearrangement, inducing an imbalance between the competing parameters of network efficiency and segregation, may underpin a mismatch between energy resources and demand representing the neuronal correlate of the energetically dysfunctional migraine brain.MethodsWe investigated, using diffusion-weighted MRI imaging tractography-based graph analysis, the graph-topological indices of the brain “connectome”, a set of grey matter regions (nodes) structurally connected by white matter paths (edges) in 94 patients with migraine without aura compared to 91 healthy controls.ResultsWe observed in migraine patients compared to healthy controls: i) higher local and global network efficiency (p < 0.001) and ii) higher local and global clustering coefficient (p < 0.001). Moreover, we found changes in the hubs topology in migraine patients with: i) posterior cingulate cortex and inferior parietal lobule (encompassing the so-called neurolimbic-pain network) assuming the hub role and ii) fronto-orbital cortex, involved in emotional aspects, and visual areas, involved in migraine pathophysiology, losing the hub role. Finally, we found higher connection (edges) probability between cortical nodes involved in pain perception and modulation as well as in cognitive and affective attribution of pain experiences, in migraine patients when compared to healthy controls (p < 0.001). No correlations were found between imaging and clinical parameters of disease severity.ConclusionThe imbalance between the need of investing resources to promote network efficiency and the need of minimizing the metabolic cost of wiring probably represents the mechanism underlying migraine patients’ susceptibility to triggers. Such changes in connectome topography suggest an intriguing pathophysiological model of migraine as brain “connectopathy”.Supplementary InformationThe online version contains supplementary material available at 10.1186/s10194-021-01315-6.     

Synthesis and Biological Evaluation of 2-Amino-3-(4-chlorobenzoyl)-4-[(4-arylpiperazin-1-yl)methyl]-5-substituted-thiophenes. Effect of the 5-Modification on Allosteric Enhancer Activity at the A(1) Adenosine Receptor

We have recently reported a detailed structure-activity relationship study around a wide series of 2-amino-3-(4-chlorobenzoyl)-4-[(4-arylpiperazin-1-yl)methyl]thiophene derivatives as potent allosteric enhancers of the A(1) adenosine receptor. In the current study, we have continued to explore the potential of these molecules by synthesizing of a novel series of analogues that share a common 2-amino-3-(4-chlorobenzoyl)thiophene nucleus. Modifications were focused on varying the nature and the position of electron-withdrawing or electron-releasing groups on the phenyl of an arylpiperazine moiety attached at the 4-position of the thiophene ring by a methylene chain, combined with the presence of small alkyl groups (methyl or ethyl), bromine, or aryl moieties at the thiophene C-5 position. In this series of compounds, substitution at the 5-position had a fundamental effect on activity, with the 5-aryl group contributing additively to the allosteric enhancer activity. The thiophene C-5 aryl derivatives 4ad, 4ak, and 4al were the most active compounds in binding and functional experiments.     

Antimetastatic effect of sulfamate carbonic anhydrase IX inhibitors in breast carcinoma xenografts

A panel of compounds belonging to the underexposed sulfamate class of carbonic anhydrase (CA, EC 4.2.1.1) inhibitors was generated that displayed high specificity at nanomolar levels for the tumor-associated CA IX/XII isoforms. Three of the specific CA IX/XII inhibitors showed a positive response in in vitro assays for tumor cell migration and spreading. One of them, 4-(3'-(3″,5″-dimethylphenyl)ureido)phenyl sulfamate (S4), was taken forward into the orthotopic MDA-MB-231 (breast carcinoma) model in mice. Treatment with a 10 mg/kg maintenance dosage of S4 given daily on a "5 days on, 2 days off" regimen reduced metastatic tumor burden in the lung while not affecting primary tumor growth or mouse condition. CA inhibitors of the sulfamate class specifically targeting the tumor-associated isoforms are potential candidates in antimetastatic therapy.