Cervical and thoracic cord atrophy in multiple sclerosis phenotypes: Quantification and correlation with clinical disability.

OBJECTIVE: We sought to characterize spinal cord atrophy along the entire spinal cord in the major multiple sclerosis (MS) phenotypes, and evaluate its correlation with clinical disability. METHODS: Axial T1-weighted images were automatically reformatted at each point along the cord. Spinal cord cross-sectional area (SCCSA) were calculated from C1-T10 vertebral body levels and profile plots were compared across phenotypes. Average values from C2-3, C4-5, and T4-9 regions were compared across phenotypes and correlated with clinical scores, and then categorized as atrophic/normal based on z-scores derived from controls, to compare clinical scores between subgroups. In a subset of relapsing-remitting cases with longitudinal scans these regions were compared to change in clinical scores. RESULTS: The cross-sectional study consisted of 149 adults diagnosed with relapsing-remitting MS (RRMS), 49 with secondary-progressive MS (SPMS), 58 with primary-progressive MS (PPMS) and 48 controls. The longitudinal study included 78 RRMS cases. Compared to controls, all MS groups had smaller average regions except RRMS in T4-9 region. In all MS groups, SCCSA from all regions, particularly the cervical cord, correlated with most clinical measures. In the RRMS cohort, 22% of cases had at least one atrophic region, whereas in progressive MS the rate was almost 70%. Longitudinal analysis showed correlation between clinical disability and cervical cord thinning. CONCLUSIONS: Spinal cord atrophy was prevalent across MS phenotypes, with regional measures from the RRMS cohort and the progressive cohort, including SPMS and PPMS, being correlated with disability. Longitudinal changes in the spinal cord were documented in RRMS cases, making it a potential marker for disease progression. While cervical SCCSA correlated with most disability and progression measures, inclusion of thoracic measurements improved this correlation and allowed for better subgrouping of spinal cord phenotypes. Cord atrophy is an important and easily obtainable imaging marker of clinical and sub-clinical progression in all MS phenotypes, and such measures can play a key role in patient selection for clinical trials.

BK virus-specific T cells for immunotherapy of progressive multifocal leukoencephalopathy: an open-label, single-cohort pilot study.

BACKGROUND: Progressive multifocal leukoencephalopathy, a rare disease of the CNS caused by JC virus and occurring in immunosuppressed people, is typically fatal unless adaptive immunity is restored. JC virus is a member of the human polyomavirus family and is closely related to the BK virus. We hypothesised that use of partly HLA-matched donor-derived BK virus-specific T cells for immunotherapy in progressive multifocal leukoencephalopathy would be feasible and safe. METHODS: We did an open-label, single-cohort pilot study in patients (aged 18 years or older) with clinically definite progressive multifocal leukoencephalopathy and disease progression in the previous month at the National Institutes of Health (NIH) Clinical Center (Bethesda, MD, USA). Overlapping peptide libraries derived from large T antigen and major capsid protein VP1 of BK virus with high sequence homology to JC virus counterparts were used to generate polyomavirus-specific T cells cross-recognising JC virus antigens. Polyomavirus-specific T cells were manufactured from peripheral blood mononuclear cells of first-degree relative donors aged 18 years or older. These cells were administered to patients by intravenous infusion at 1 × 106 polyomavirus-specific T cells per kg, followed by up to two additional infusions at 2 × 106 polyomavirus-specific T cells per kg. The primary endpoints were feasibility (no manufacturing failure based on meeting release criteria, achieving adequate numbers of cell product for clinical use, and showing measurable antiviral activity) and safety in all patients. The safety monitoring period was 28 days after each infusion. Patients were followed up with serial MRI for up to 12 months after the final infusion. This trial is registered at ClinicalTrials.gov, NCT02694783. FINDINGS: Between April 7, 2016, and Oct 19, 2018, 26 patients were screened, of whom 12 were confirmed eligible and received treatment derived from 14 matched donors. All administered polyomavirus-specific T cells met the release criteria and recognised cognate antigens in vitro. 12 patients received at least one infusion, ten received at least two, and seven received a total of three infusions. The median on-study follow-up was 109·5 days (range 23-699). All infusions were tolerated well, and no serious treatment-related adverse events were observed. Seven patients survived progressive multifocal leukoencephalopathy for longer than 1 year after the first infusion, whereas five died of progressive multifocal leukoencephalopathy within 3 months. INTERPRETATION: We showed that generation of polyomavirus-specific T cells from healthy related donors is feasible, and these cells can be safely used as an infusion for adoptive immunotherapy of progressive multifocal leukoencephalopathy. Although not powered to assess efficacy, our data provide additional support for this strategy as a potential life-saving therapy for some patients. FUNDING: Intramural Research Program of the National Institute of Neurological Disorders and Stroke of the NIH.

Progressive multifocal leukoencephalopathy lesion and brain parenchymal segmentation from MRI using serial deep convolutional neural networks.

Progressive multifocal leukoencephalopathy (PML) is a rare opportunistic brain infection caused by the JC virus and associated with substantial morbidity and mortality. Accurate MRI assessment of PML lesion burden and brain parenchymal atrophy is of decisive value in monitoring the disease course and response to therapy. However, there are currently no validated automatic methods for quantification of PML lesion burden or associated parenchymal volume loss. Furthermore, manual brain or lesion delineations can be tedious, require the use of valuable time resources by radiologists or trained experts, and are often subjective. In this work, we introduce JCnet (named after the causative viral agent), an end-to-end, fully automated method for brain parenchymal and lesion segmentation in PML using consecutive 3D patch-based convolutional neural networks. The network architecture consists of multi-view feature pyramid networks with hierarchical residual learning blocks containing embedded batch normalization and nonlinear activation functions. The feature maps across the bottom-up and top-down pathways of the feature pyramids are merged, and an output probability membership generated through convolutional pathways, thus rendering the method fully convolutional. Our results show that this approach outperforms and improves longitudinal consistency compared to conventional, state-of-the-art methods of healthy brain and multiple sclerosis lesion segmentation, utilized here as comparators given the lack of available methods validated for use in PML. The ability to produce robust and accurate automated measures of brain atrophy and lesion segmentation in PML is not only valuable clinically but holds promise toward including standardized quantitative MRI measures in clinical trials of targeted therapies. Code is available at: https://github.com/omarallouz/JCnet.