Revue neurologique最新文献

The genetics of multiple sclerosis (MS) has advanced dramatically through the combined impact of high-throughput genotyping, large biobank resources, worldwide collaborations, and powerful computational analyses. Over the past two decades, genome-wide association studies (GWAS) have significantly reshaped our understanding of the genetic architecture of complex multifactorial diseases such as MS. The number of MS susceptibility-associated genomic regions is now more than 200, underscoring its highly polygenic nature. Within the major histocompatibility complex (MHC), the HLA-DRB1*15:01 allele and its extended haplotypes remain the most robust and reproducible risk factors. This historical association reflects a long-standing link between antigen presentation, immune regulation, and central nervous system autoimmunity. Beyond the MHC, common non-HLA loci implicate a wide network of immune pathways involving T-cell and B-cell activation, cytokine signaling, and antigen presentation. Moreover, rare variant analyses and family-based designs, although limited in power, have uncovered additional susceptibility genes, such as PRF1, CYP27B1, and NLRP1, shedding light on distinct mechanisms of immune modulation and metabolic regulation. The step forward will be now to explore diverse genetic ancestry populations, bearing differences in risk allele frequencies; multi-ethnic and family-based designs are needed to disentangle true genetic effects from environmental confounders. In parallel, progress has been made toward MS progression, as variants potentially influencing disability accumulation and neurodegeneration were identified. These findings have deepened our understanding of MS pathophysiology. They now provide a foundation for future integrative models that combine genetics, environment, and multi-omics data to elucidate disease heterogeneity and guide personalized therapeutic strategies.

Recently, diagnostic criteria for demyelinating diseases, including multiple sclerosis, have been updated with a focus on imaging and fluid markers. The upcoming criteria for neuromyelitis optica (NMO) introduce several clinico-radiological updates, and highlight the need for accurate determination of AQP4-IgG status. Furthermore, seronegative NMO is now considered a syndrome distinct from AQP4-IgG disease. The MOG-antibody associated disease (MOGAD) criteria have been validated in different populations, and a clear role for case-by-case expert review has emerged in light of some limitations to clinical application of the criteria. The impact and the limitations of these criteria will be discussed in this mini-review.

Most clinical questions in neurology are inherently causal, yet observational studies have long been interpreted in terms of associations rather than causal effects of well-defined interventions. This limitation has contributed to persistent biases in neurology research, including inappropriate covariate adjustment, temporal misalignment, immortal-time bias, and susceptible depletion. Recent developments in causal inference offer a formal framework to overcome these issues through counterfactual reasoning, causal diagrams, and g-methods. The target trial emulation approach translates these concepts into a practical strategy: investigators first specify the protocol of the randomized trial that would ideally address their question, and then emulate this protocol using observational data. This requires explicit definition of eligibility, treatment strategies, assignment procedures, time zero, follow-up, outcomes, causal contrasts, and handling of intercurrent events, together with prespecified causal assumptions and an appropriate analysis plan. By aligning the design of observational studies with the structure of randomized trials, target trial emulation reduces design-related bias, clarifies the causal question, and provides a transparent foundation for real-world evidence generation. Validation studies, including recent work in multiple sclerosis, have demonstrated good concordance between emulated trials using observational data and randomized controlled trials. Target trial emulation therefore represents an essential complement to randomized evidence, particularly for questions that are difficult or impossible to address through randomization, and has the potential to substantially improve causal inference in neurology.

There are two main evolutionary theories for why modern humans develop multiple sclerosis (MS). The first, antagonistic pleiotropy, argues that genetic variants that predispose people to MS are protective against infections; this explains why these genetic variants persist at high frequency. The second, the Old Friends hypothesis, argues that reduced exposure to ancient, co-evolved micro-organisms like helminths results in immune dysregulation and overreactions to harmless infections and substances; this explains why MS prevalence increases with sanitation. Here, I assess these theories in the light of recent ancient DNA (aDNA) studies. These suggest that populations from the Pontic-Caspian Steppe which migrated across Eurasia in the Bronze Age evolved a strong pro-inflammatory immune response due to increased zoonotic infections resulting from their pastoralist lifestyle. It is possible that this may have occurred in a context of high levels of anti-inflammatory helminth infections, which resulted in a balanced pro- and anti- inflammatory response. In the modern sanitary world, with a lower helminth burden, the immune system 'overshoots' the level of inflammation required. This explains why genetic risk for MS is higher in northern Europe where people have higher genetic ancestry from the ancient Steppe population, and why disease penetrance is increasing.

The 2024 update to the McDonald criteria for multiple sclerosis aims to be more inclusive, covering paediatric and elderly patients, as well as those with atypical symptoms or comorbidities. These criteria emphasise objective evidence of central nervous system involvement through clinical signs and biomarkers from fluid analysis or imaging. Several new biomarkers have been added, although some are not yet widely accessible but are expected to be implemented soon: kappa-free light chains in cerebrospinal fluid, which are more reproducible and cost-effective than oligoclonal bands, along with advanced imaging techniques such as the central vein sign and paramagnetic rim lesions on susceptibility-weighted imaging. Cases with incidental magnetic resonance imaging hyperintensities, identified as radiologically isolated syndrome (RIS), may also be classified as preclinical MS if they show T2 lesions in at least two typical locations and meet other specific biomarker criteria. Careful, step-by-step evaluation is essential to prevent misdiagnosis, especially in paediatric and late-onset MS, and in cases with comorbidities. The inclusion of RIS individuals or those with atypical presentations marks a crucial shift in the MS spectrum, requiring expertise and the use of specific biomarkers.

Aquaporin-4-IgG-positive neuromyelitis optica spectrum disorder (AQP4+NMOSD) and myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) are rare, antibody-mediated inflammatory disorders of the central nervous system (CNS). Although historically grouped under the umbrella of multiple sclerosis, both entities differ markedly in pathophysiology, clinical course, prognosis and therapeutic requirements. The identification of disease-specific autoantibodies has not only refined diagnostic accuracy but has also highlighted fundamental differences in therapeutic strategy, particularly regarding the need for and choice of long-term immunosuppression. While acute management of both conditions relies on similar approaches, the long-term treatment paradigms diverge. Sustained immunosuppression may not be required for all patients with MOGAD, whereas AQP4+NMOSD typically demands continuous therapy, in which agents directed against complement activation, IL-6 signaling, and B-cell-mediated pathways form the cornerstone of relapse prevention. The rarity of these disorders continues to challenge the conduct of large, rigorous therapeutic trials. Nonetheless, innovative targeted therapies have already been developed and transformed the therapeutic landscape in NMOSD while more off-label approaches are being investigating in MOGAD. This review synthesizes current evidence on established and investigational therapies in NMOSD and MOGAD, providing an updated framework to inform clinical practice and guide future research in these complex, antibody-mediated CNS disorders.