Aging, immunosenescense, and very late-onset neuromyelitis optica spectrum disorders CURRENT STATUS:

Objective : To clarify the clinical features of very late-onset neuromyelitis optica spectrum disorders NMOSD (VLO-NMOSD). Methods : According to the age at onset, we classified patients with NMOSD into three subgroups of early-onset NMOSD (EO-NMOSD), late-onset NMOSD (LO-NMOSD), and VLO-NMOSD. We evaluated the clinical characteristics, MRI findings, laboratory data, and immunotherapies among the groups. Results : Overall, eight males and 36 females with a median age at onset of 43.00 years (interquartile range, 29.00–54.75) and median duration of disease of 6.00 months (interquartile range, 3.00–11.75) were included. This included 7 (16%) patients with VLO-NMOSD, 11 (25%) with LO-NMOSD, and 26 (59%) with EO-NMOSD. Patients with EO-NMOSD had significantly longer disease duration than those did with VLO-NMOSD and LO-NMOSD (p=0.015). Optic nerve lesions on MRI and optic neuritis were significantly less frequent in patients with VLO-NMOSD than in those with LO-NMOSD and EO-NMOSD (p=0.013 and p=0.046, respectively), whereas the length of the spinal lesion was significantly longer in those with VLO-NMOSD in comparison with those with LO-NMOSD and EO-NMOSD (p=0.038). Conclusions : Spinal cord lesion in patients with VLO-NMOSD was significantly longer despite short disease duration, and optic neuritis was significantly less frequent in them.


Brain and spinal cord MRI findings
Optic nerve lesions were observed significantly less frequently in patients with VLO-NMOSD (p=0.013, Table 3), whereas the length of the spinal lesion was significantly longer in these patients (p=0.038, Table 3). Figure 1 depicts the typical spinal MRIs in patients with VLO-NMOSD, LO-NMOSD, and EO-NMOSD.

Illustrative cases of VLO-NMOSD
Case 1: An 81-year-old woman experienced numbness in her legs and severe quadriplegia with urinary retention for 5 days. Spinal MRI demonstrated a long cord lesion extending from C2 to T5. Serum test for AQP4 Abs was positive. She received intravenous methylprednisolone (IVMP) at a dose of 1 g/day for 3 days without signs of improvement. Two weeks after the initial IVMP, she experienced rapidly progressive visual loss in both eyes, which culminated in complete blindness within 5 days. She received a second course of IVMP for 5 days with oral tapering of steroids. Consequently, her visual deficit improved slightly. Finally, 10 days after the second cycle of IVMP, she developed excessive bleeding from a rectal ulcer and shock. Currently, she is 83 years old and had been bedridden for approximately 2 years.  days. Spinal MRI demonstrated spinal cord lesions extending from C7 to C4 and T2 to T4. This patient was seronegative for AQP4 Abs but her serum tests for anti-nuclear antibody (ANA), rheumatoid factor (RF), and anti-SSA and SSB Abs were positive. IVMP was administered for 3 days in combination with oral PSL, which resulted in clinical improvement. Her symptoms gradually resolved, and she eventually was able to walk with a cane.

Discussion
In the present study, the clinical features, laboratory findings, and MRI findings were investigated in patients with NMOSD, with a particular focus on the age of onset. This study is the first to our knowledge to describe the clinical characteristics of patients with VLO-NMOSD. Two important results of this study were: i) spinal cord lesions in patients with VLO-NMOSD were significantly longer within short disease duration, and ii) optic neuritis was significantly less frequent in VLO-NMOSD.
Recent studies have reported that patients with EO-NMOSD tended to manifest optic neuritis as the first symptom, whereas those with LO-NMOSD tended to develop myelitis first [16,17]. Interestingly, optic neuritis is also less frequently seen in patients with late-onset multiple sclerosis than in those with early-onset (< 50 years) multiple sclerosis [31]. Therefore, there is a possibility that the optic nerves are more vulnerable in the younger population rather than an older one. In contrast, myelitis in patients with VLO-NMOSD was very aggressive according to the findings of short disease duration and intervals between episodes when compared with patients with LO-NMOSD and EO-NMOSD.
Furthermore, we observed that both EDSS score and Nurick scale tended to be higher in patients with VLO-NMOSD than those in patients with LO-NMOSD and EO-NMOSD, although the differences were not significant. We believe that the symptoms and clinical course in patients with VLO-NMOSD were severe due to very long cord lesions, as we have demonstrated with illustrative cases 1 and 2.
Sepulveda et al. focused on the relationship between the serostatus for Abs and disability in patients with NMOSD divided by age groups [32]. They reported that patients with LO-NMOSD (≥50 years in that study) who had AQP4 Abs had worse outcomes over a short period of follow-up than did those with EO-NMOSD who had AQP4 Abs. Our results are in agreement with their results.
Our findings suggest that the clinical characteristics of patients with VLO-NMOSD may be related to immunosenescence, which may be importance when considering the pathogenesis of autoimmune diseases [1][2][3][4][5][6]33]. Our findings suggest that the clinical characteristics of patients with VLO-NMOSD may be related to immunosenescence. We were interested in the results of the presence or absence of other Abs in the serum and immunoglobulin G (IgG) index in CSF as indicators of immunosenescence. In their study, Sepulveda et al. reported that concomitant autoimmune diseases were frequently confirmed in patients with LO-NMOSD [32]. In this study, the seroprevalence of ANA, the average CSF protein, and the average IgG index were higher in patients with VLO-NMOSD than in those with LO-NMOSD, although the differences were not statistically significant. Autoimmune abnormalities in the elderly are accompanied with increased production of autoantibodies such as ANA, RF, and anti-phospholipid Abs [3]. In case 3 described above, we detected ANA, RF, and anti-SSA and SSB Abs in the patient's serum, and confirmed remarkably elevated IgG index. We considered this case as the typical case of VLO-NMOSD with immunosenescence. Although autoimmunity should not be considered an obvious characteristic of the elderly [34,35], a large proportion of the elderly may demonstrate autoantibodies, which may be related to higher exposure to exogenous factors, such as infections and medications. Although the immunopathogenesis appears to be complex, we should verify the alterations in the immune functions due to aging in patients with VLO-NMOSD [6]. 6 The limitations of the present study include its retrospective design and the small sample size.

Conclusions
We reported the clinical characteristics of NMOSD according to age of onset. Patients with VLO-NMOSD tended to manifest severe myelitis with long cord lesions but not optic neuritis. Prospective, interventional, multi-centre studies according to the age group are necessary to confirm the relationships between the clinical features, levels of AQP4 Abs, biomarkers (including other Abs and CSF analysis), and MRI findings versus the outcomes of immunotherapies in patients with VLO-NMOSD.

Patients and study design
We reviewed the clinical records of patients diagnosed with NMOSD who were treated at the All patients underwent review of full medical history and neurological examinations by at least two neurologists. Comprehensive clinical and serological assessments were performed in all patients. We extensively reviewed the following clinical information: age, sex, age of onset of first symptom, disease duration, index episode, attack interval, estimation of the disability during an acute episode and at the last visit (e.g., EDSS [28], Nurick scale score [29]), and comorbidities (e.g., Charlson comorbidity index [30]). Additionally, the following laboratory data were reviewed: qualitative data of AQP4 Abs, other Abs tests, CSF analysis and immunotherapies.

MRI evaluation of the brain and spinal cord
MRI scans at the Kumamoto University Hospital were acquired with 3-T MRI systems (Achieva, Philips Healthcare, The Netherlands; MAGNETOM Trio, MAGNETOM Prisma fit , Siemens, Erlangen, Germany).
These images were separately interpreted by two neurologists (KN and SN) based on the reports by the neuroradiologist. The presence or absence of MRI lesions in the cerebrum, cerebellum, brainstem, optic nerve, and spinal cord in the cervical, thoracic, and lumber sections were evaluated. The length of the spinal cord lesions was expressed in terms of the number of vertebral segments.

Statistics
In this study observational cross-sectional study, the sample size was determined by considering the number of outpatients and inpatients at the Kumamoto University Hospital during the survey period.

Ethics approval and consent to participate
For a retrospective analysis that is Institutional Review Board (IRB) of Kumamoto University-approved (Permit Number: 1918), state that approval from an ethical standards committee to conduct this study was received.

Consent for publication
Not applicable.

Availability of data and materials
Anonymized data will be shared by request from any qualified investigator. All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan (JSPS KAKENHI Grant Numbers 16K09695 and 19H03549).