Myelin oligodendrocyte glycoprotein (MOG) antibody–associated disease (MOGAD) is an inflammatory demyelinating CNS condition characterized by the presence of MOG autoantibodies. We sought to investigate whether human MOG autoantibodies are capable of mediating damage to MOG-expressing cells through multiple mechanisms. We developed high-throughput assays to measure complement activity (CA), complement-dependent cytotoxicity (CDC), antibody-dependent cellular phagocytosis (ADCP), and antibody-dependent cellular cytotoxicity (ADCC) of live MOG-expressing cells. MOGAD patient sera effectively mediate all of these effector functions. Our collective analyses reveal that (a) cytotoxicity is not incumbent on MOG autoantibody quantity alone; (b) engagement of effector functions by MOGAD patient serum is bimodal, with some sera exhibiting cytotoxic capacity while others did not; (c) the magnitude of CDC and ADCP is elevated closer to relapse, while MOG-IgG binding is not; and (d) all IgG subclasses can damage MOG-expressing cells. Histopathology from a representative MOGAD case revealed congruence between lesion histology and serum CDC and ADCP, and we identified NK cells, mediators of ADCC, in the cerebrospinal fluid of relapsing patients with MOGAD. Thus, MOGAD-derived autoantibodies are cytotoxic to MOG-expressing cells through multiple mechanisms, and assays quantifying CDC and ADCP may prove to be effective tools for predicting risk of future relapses.

Central pontine myelinolysis (CPM) is a demyelinating disorder of the central basis pontis that is often associated with osmotic stress. The aquaporin water channels (AQPs) have been pathogenically implicated because serum osmolarity changes redistribute water and osmolytes among various central nervous system compartments. We characterized the immunoreactivity of aquaporin-1 and aquaporin-4 (AQP1 and AQP4) and associated neuropathology in microscopic transverse sections from archival autopsied pontine tissue from 6 patients with pathologically confirmed CPM. Loss of both AQP1 and AQP4 was evident within demyelinating lesions in four of the six cases, despite the presence of glial fibrillary acidic protein (GFAP)-positive astrocytes. Lesional astrocytes were small, and exhibited fewer and shorter processes than perilesional astrocytes. In two of the six cases, astrocytes within demyelinating lesions exhibited increased AQP1 and AQP4 immunoreactivities, and gemistocytes and mitotic astrocytes were numerous. Blinded review of medical records revealed that all four cases lacking lesional AQP1 and AQP4 immunoreactivities were male, whereas the two cases with enhanced lesional AQP1 and AQP4 immunoreactivities were female. This report is the first to establish astrocytic AQP loss in a subset of human CPM cases and suggests AQP1 and AQP4 may be involved in the pathogenesis of CPM. Further studies are required to determine whether the loss of AQP1 and AQP4 is restricted to male CPM patients, or rather may be a feature associated with specific underlying precipitants of CPM that may be more common among men. Non-rodent experimental models are needed to better clarify the complex and dynamic mechanisms involved in the regulation of AQPs in CPM, in order to determine whether it occurs secondary to the destructive disease process, or represents a compensatory mechanism protecting the astrocyte against apoptosis.

Deep networks have achieved great success in image rescaling (IR) task that seeks to learn the optimal downscaled representations, i.e., low-resolution (LR) images, to reconstruct the original high-resolution (HR) images. Compared with super-resolution methods that consider a fixed downscaling scheme, e.g., bicubic, IR often achieves significantly better reconstruction performance thanks to the learned downscaled representations. This highlights the importance of a good downscaled representation. Existing IR methods mainly learn the downscaled representation by jointly optimizing the downscaling and upscaling models. Unlike them, we seek to improve the downscaled representation through a different and more direct way – directly optimizing the downscaled image itself instead of the down-/upscaling models. Consequently, we propose a Hierarchical Collaborative Downscaling (HCD) method that performs gradient descent w.r.t. the reconstruction loss in both HR and LR domains to improve the downscaled representations, so as to boost IR performance. Extensive experiments show that our HCD significantly improves the reconstruction performance both quantitatively and qualitatively. Particularly, we improve over popular IR methods by >0.57 dB PSNR on Set5. Moreover, we also highlight the flexibility of our HCD since it can generalize well across diverse image rescaling models. The code is available at https://github.com/xubingna/HCD.

The Mayo Clinic institutional review board approved this study. In the past 30 years, Mayo Clinic’s Neuroimmunology Laboratory identified 650 PCA-1-IgG-seropositive patients3; 8 (1.2%) were men (online supplementary table 1a), with PCA-1-IgG specificity confirmed by western blot with purified human cerebellar degeneration-related protein 2 (CDR2, the PCA-1 antigen; Euroimmun, Lubeck, Germany). Clinical information was obtained for seven by case record review and physician telephone interviews. A PubMed literature search (1990–2015) identified 10 previously published PCA-1-IgG-positive male cases (online supplementary table 1b).

Kelch-like protein-11 (KLHL11)-IgG is associated with rhombencephalitis and seminoma. It has not previously been described as a neurologic immune checkpoint inhibitor (ICI)-related adverse event (nirAE) or in association with esophageal adenocarcinoma. We describe a 61-year-old man with metastatic esophageal adenocarcinoma treated with folinic acid, fluorouracil, oxaliplatin (FOLFOX), and nivolumab, who subsequently developed diplopia, vertigo, and progressive gait ataxia after 8 weeks of treatment. Owing to a concern for ICI-associated myasthenia gravis, nivolumab was held and he was treated with prednisone and pyridostigmine. EMG showed no neuromuscular junction dysfunction, and acetylcholine-receptor antibodies were negative. Brain MRI was unrevealing. Murine brain tissue immunofluorescence assay revealed KLHL11-IgG in both serum and CSF, confirmed by cell-based assay. Tumor histopathology demonstrated poorly differentiated, highly proliferative adenocarcinoma with increased mitotic figures and cytoplasmic KLHL11 immunoreactivity. He was initiated on 6 months of cyclophosphamide in addition to FOLFOX for post-ICI-associated KLHL11-IgG rhombencephalitis. We report KLHL11-IgG rhombencephalitis associated with poorly differentiated esophageal cancer as a novel nirAE. Tumor staining revealed KLHL11 immunoreactivity, supporting a cancer-antigen-driven ICI-associated paraneoplastic syndrome. Recognition of novel nirAEs can expedite treatment and potentially prevent progressive neurologic disability.

ObjectivesMyelin oligodendrocyte glycoprotein antibody–associated disease (MOGAD) is an autoimmune demyelinating disease rarely associated with malignancy. We report the clinical, MRI, immunopathology, and treatment response in a person with MOGAD and ...