KaSP_AE

A pre-posted experimental design page for the collaborative project between KaSP and OANG

Karolinska Schizophrenia Project-Oxford Autoimmune Neurology Group collaboration project page

Currently, the Karolinska Schitzophrenia Project (KaSP) consortium at Karolinska Institutet and the Oxford Autoimmune Neurology Group (OANG, University of Oxford) is starting a collaboration project aiming at clarifying whether there are first-episode psychosis patients that have potentially pathogenic cell-surface reactive autoantibodies in their CSF, consistent with a diagnosis of autoimmune encephalitis. The collaboration utilizes the biobank collected within the KaSP, that consists of serum, CSF and PBMC samples from around 120 individuals, with a patient/healthy control ratio of 2:1 (see KaSP website).

Experimental outline

All primary screening will be blindly performed on live, antigen-transfected HEK 293T cells. After all experimental procedures have been conducted, unblinding will occur.

Experimental order

1. Screening for autoantibodies against the NMDA-receptor subunit 1 in serum and CSF. This will be performed with constructs of two different NR1 isoforms (1 and 7). To also be able to identify potential autoantibodies only binding the fully assembled NMDA receptor, the NR2B subunit will be co-expressed in both cases.

2. Screening for autoantibodies against CASPR2, LGI1, Dopamine 2 receptor, Glycine receptor, GABAA receptor and GABAB receptor in serum with single antigen expression in HEK293T cells.

3. In individuals where autoantibodies against one or multiple autoantigen(s) were detected in serum during the primary screening described under 2, secondary screening for the detected autoantigens in CSF.

4. End-point titration of autoantibodies in serum and potentially CSF, in all individuals where such have been found.

5. Potentially: a. screening for unknown autoantigens on fixed rat brain sections for all serum / CSF samples. b. if individuals with autoantibodies are detected, CyTOF analysis of their immunocellular phenotype might be possible from an earlier generated dataset, or by running new experiments on stored PBMC samples. An inevitable constraint is that CyTOF data is only available for a sub-group of the patients, which is also true for available PBMC samples. If utilized, the CyTOF data will be requested by sending the codes for antibody positive patients to Karolinska. An amendment to this site will be made by a representative for the KaSP Consortium. The antibody status will only be unblinded in the final step, when all data are already tabulated. A similar strategy will be taken to decide if it is meaningful to include further PBMC analyses in the study: the codes for antibody positive patients will be sent to Karolinska Institutet and the decision will be made there, with the codes and the knowledge of the availability of material. This decision will however not only depend on the availability of samples, but also on the number of identified antibody-positive patients and potential legal changes occurring as a result of Brexit, which might limit the possibility of shipping the samples in question between the institutions.

6. The clinical phenotypes of any antibody-positive patient(s) will be compared to the clinical phenotype of the other patients in the cohort, and comparisons will also be made to previously assembled data for NMDA receptor encephalitis patients, published in: Al-Diwani et al, Lancet Psychiatry, 2019.

UPDATE 2020-03-11 The screening phase is now concluded and it has been decided to go on with an analysis of an associated CyTOF dataset. Due to previous analyses of the CyTOF dataset, it will not be possible to keep the blinding during the CyTOF analysis phase so the labels are now being broken.

UPDATE 2020-09-11 The CyTOF analyses needed to be exluded from the study due to several limitation in the acquisition of the data. The study is currently being made ready for submission.