Neuroimmunology Rajeev Krishnadas MD PhD FRCPsych Consultant Psychiatrist ESTEEM Glasgow ESTEEM research theme Finding neural substrates of psychosis MRC Identifying Neural Signatures of AuditoryPredictive Processing in Schizophrenia A MultiModal Imaging Approach ID: 1042584
Download Presentation The PPT/PDF document "ESTEEM – NHS GGC Neuroimaging and" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. ESTEEM – NHS GGCNeuroimaging and NeuroimmunologyRajeev KrishnadasMD PhD FRCPsychConsultant PsychiatristESTEEM – Glasgow
2. ESTEEM – research themeFinding neural substrates of psychosisMRC – Identifying Neural Signatures of Auditory-Predictive Processing in Schizophrenia: A Multi-Modal Imaging ApproachNeuroimmunology – PPiP2Predicting outcomes at an individual level FENO / Meta FENOEMPATH – Sam Leighton
3. What will you getResearch Opportunity ARCPCognitive testingPANSS rating / clinical phenotypingOpportunities for publicationSpecial interest sessions in ESTEEM?Interest in academia??
4.
5. Neurons CODE INFORMATIONSpikeFun: Spiking Neural Network Visualization - 260K neurons, 8M synapsesReceptorsNeuronSpiking / Action potentialNetworks
6. Information is coded as spiking patterns The brain decodes these patterns of spikes to perceive …The brain uses models of the world around to predict “causes” for the spiking patternsE.g. possible pattern of spiking that represent “dogs” and “sheep” or “houses” or “faces” Neurons code INFORMATION (MODELS of world)James Stone. Principles of Neural Information Theory: Computational Neuroscience and Metabolic EfficiencyJakob Hohwy The predictive mind
7. Priors (existing model) influence what we perceive or believe
8. Internal model - prior(faces are convex)Incoming data – Ambiguous - uncertain(Mask rotating)Ignore the data – that is not right (fake news)“priors” determine what we perceive or believePosterior -(concave is perceived as convex)
9. Priors influence what we perceive or believe
10. Priors influence what we perceive or believe
11. Priors influence what we perceive or believe
12. Identifying Neural Signatures of Auditory-Predictive Processing in Schizophrenia: A Multi-Modal Imaging ApproachMRC fundedExisting Model of the worldIncoming dataWhat is perceived
13. Brain – hierarchical Bayesian modelPrecision determines the influence of one piece of information over the otherThis is represented by the “synaptic gain” (the post synaptic activity determined by the pre-synaptic activity)Synaptic gain is determined by NMDARSynaptic gain is modulated by DA D1R increases gain; D2R reduces gainLoss of synaptic gain = Loss of precision at that level.Increase in synaptic gain = Increase of precision at that levelPrecision of both “prior” and “likelihood”…Biological Psychiatry, Volume 84, Issue 9, 1 November 2018, Pages 634-643
14. Feed forward and Feedback can be measured with MEG and 7 T fMRIFeedforward (FF) signaling of PE involve - gamma (30-80 Hz) and theta (3-7 Hz) frequencies. FF-inputs arrive predominantly in layers 3/4Feedback (FB) processes that reflect top-down mediated predictions - activity in alpha (8-12 Hz) and beta (13-30Hz) activity. FB preferentially terminate in superficial outer layers and infragranular deep layers
15. Loss of precision in high level prior (cortical) D1/NMDAR Delusion formation (JTC bias)Increased precision in data/evidence (likelihood) D2 Delusion formation (JTC bias)Compensatory increase in precision of updated prior Delusion sustained (set shifting problem)Increased precision in likelihood hallucinations/misperceptionsIncreased precision in prior internal models take over hallucinationsSUMMARY – POSSIBLE MECHANISMS
16. Hypothesis Propagation of FF-PE modulates BOLD-activation patterns in layer ¾ that correlate with increased theta/gamma-band connectivity. In contrast, FB mediated predictions are implemented by BOLD-activation in infragranular layers and top down alpha/beta-band activity. PE measured by the MMN-potential and spectral power amplitudes.Early-stage ScZ will be characterized by reduced MMN-amplitudes as well as reduced FF and FB signalling in MEG and 7T fMRI-data.30 participants with Schizophrenia / 30 controls.
17.
18. Prevalence of pathogenic antibodies in first episode psychosis (PPiP1)Psychosis group (n=228):First episode psychotic illness35 NHS Mental Health Trusts (England)Age:14-35 < 6 weeks on medicationat least one moderate or more severe psychosis symptom ( PANSS)Healthy controls (n=105):General population (Cambridge)No history of mental illness Age, gender and ethnicity matchedLennox et al., 2017, The Lancet Psychiatry
19. PPiP120/228 (8.8%) positive to any of the neuronal cell surface antibodies in comparison to 4/105 (4%) in the group of healthy participants (non-significant difference; unadjusted odds ratio 2.4, 95% CI 0.8–7.3)Patients with psychosis were more likely to have NMDAR antibodies than healthy controls (p=0.0204)No difference in clinical symptoms between patients with and without antibodies.Lennox et al., 2017, The Lancet Psychiatry
20. Screening study (PPiP2) 2015-2022ObjectiveTo identify patients with acute psychosis symptoms and presence of anti-neuronal membranes antibodies of interest in their serum. Methods: DesignObservational study in over 40 NHS Mental Health Trusts in England and ScotlandCollecting a blood sample and testing this for antibodies (Live assays NMDA, LGI1, GABA-A, CASPR2 and other)
21. Methods: ParticipantsInclusion Criteria: Age 16-60If first episode - acute psychosis symptom(s) lasting for at least the past two weeks but no longer than two years. If relapse –same as above, but there must have been 6 months of remission prior to the current episodeExclusion criteria:Any other neurological disorders (multiple sclerosis, epilepsy, cerebrovascular disease, traumatic brain injury, systemic lupus erythematosus, CNS vasculitis)PregnancyAcute psychosis symptoms (relapse or first episode of psychosis)
22. PPiP2 MethodsWho:Local researchers CliniciansWhere:All services – emphasis on acute services (early intervention services, inpatient services, crisis intervention services)Confirming of eligibility (clinical judgement):Severity of symptoms correspond to a definition of moderate or more severe (rating 4 or more) of at least one selected PANSS item at time of blood sample. No PANSS training or assessment requiredPANSS selected symptoms (Andreasen et al, 2005)1. PANSS Positive Scale items: Delusions (P1), Conceptual Disorganisation (P2)Hallucinatory Behaviour (P3)2. PANSS Negative Scale items: Blunted affect (N1),Passive/Apathetic Social Withdrawal (N4), Lack of Spontaneity & Flow of Conversation (N6)3. PANSS General Scale items: Mannerisms & posturing (G5), Unusual Thought Content (G9)
23. Confirming of eligibilityConfirming of period of remission (mild or absent symptoms)If current episode is relapse, there must be a period of at least six months of remission prior to this.Who?Clinical team, care coordinators etc. with knowledge of the patient’s history and through retrospective clinical record review.How?Estimated severity of ALL 8 PANSS selected symptoms correspond to definition of ratings 3 (mild) or less for at least 6 months prior to current episode of psychosisPrevious episode≥ 6 monthsCurrent episode≥ 2 weeks & < 2 years
24. Blood TestLive assays – NMDAR, LGI1, GABAA, CASPR2 (in the PPiP2 lab in Oxford)Test results received within 10 daysLocal researcher and PI informed by emailLocal researcher informs care team (consultant or care coordinator).Negative result – patient informed by a clinician but no further research action requiredPositive result – patient informed within 24-48 hours by a clinician and referred to neurologist at SINAPPS2 site
25. SINAPPS 2A randomised, phase II, double-blinded, placebo controlled trial of intravenous immunoglobulins and rituximab in patients with antibody-associated psychosis.2017-2022Lennox at al., 2019, BMC Trials
26. ObjectiveTo test the efficacy and safety of immunotherapy (IVIG and rituximab) treatment versus placebo in patients with antibody-associated psychosis.SINAPPS2Primary outcomeTime to start of symptomatic recovery defined as symptomatic remission sustained for at least six months.
27. SINAPPS2 TreatmentAll patients:Anti-psychotic treatment to continue, as prescribed by treating psychiatrist. Immunotherapy (over approximately 6-8 weeks):IV Immunoglobulin / placeboRituximab/placeboFollow up for 12-18 monthsLennox et al., 2019, BMC Trials
28. Cambridge – Alasdair ColesOxford – Belinda LennoxNottingham – Akram HosseiniLondon UCLH – Eileen Joyce and Mike ZandiExeter – Tim HarrowerLondon KCH – Ester CoutinhoSalford Royal – Nazir SharafSheffield RHH – Priya ShanmugarajahLiverpool – Saif HudaGlasgow – John GoodfellowSINAPPS2: treating centres
29. SummaryPlacebo (n=35)Active Treatment(n=35)SINAPPS2 ScreeningFollow-up 12 – 18 months Primary outcome measure:Time to symptomatic recovery (symptomatic remission sustained for at least 6 months)SINAPPS2 n = 70PPiP2 screening study n = > 6,400PPiP2 screening study n = > 6,400
30. Thank yourajeev.krishnadas@glasgow.ac.uk