The term neuromyelitis optica spectrum disorder(NMOSD)refers to a central nervous system inflammatory demyelination disease characterized by predominant involvement of the optic nerves and spinal cord.NMOSD has a high annualized relapse rate,and repeated relapses and cumulative damage often lead to permanent blindness or paralysis.Sequential treatment during remission is the primary method of reducing the relapse of NMOSD and preventing the progression of disability.Rituximab(RTX)significantly reduces the relapse of NMOSD by deleting B cells from the peripheral circulation.Many studies have confirmed that continuous depletion of B cells is associated with the maintenance of a clinically relapse-free state,thus requiring repeated use of RTX.Common RTX regimens include high-dose B-cellderived tumor clearance regimens and multiple reduced-dose RTX regimens with repeat dosing guided by B-cell proliferation.Our previous observations in clinical practice and some studies have shown that patients with NMOSD are not prone to relapse at higher levels of total CD19+B cells,so presumably,the requirements for B-cell clearance in NMOSD may not be as high as those required for the treatment of lymphoma.The exact level at which B cells rapidly increase the chances of NMOSD relapse has not been clarified,and the role of optimizing long-term maintenance regimens for RTX is particularly important in preventing NMOSD relapse.However,B-cell clearance therapy can not result in remission in all NMOSD patients.The efficacy of mycophenolate mofetil(MMF)is comparable to RTX in reducing NMOSD relapse but does not require monitoring and maintaining lower B-cell levels as with RTX.Apart from B cells,what other biomarkers are of interest in the monitoring of the efficacy of immunosuppressive drugs and the assessment of disease prognosis?Studies have shown that NMOSD is a neuroimmune disease involving both T and B cells.The differentiation of primary T cells into antigen-specific T cells by the synergistic signaling of co-stimulatory molecules provided by antigen-presenting cells is a prerequisite for aquaporin 4-antibody to enter the central nervous system and exert pathogenic effects.The Inducible costimulator(ICOS)and its ligand(ICOSL)and programmed death 1(PD-1)and its ligand(PD-L1)are important positive and negative costimulatory molecules in the B7-CD28 family,respectively.Membrane-bound and soluble ICOS/ICOSL and PD1/PD-L1 participate in regulating different stages after T-cell activation on peripheral blood lymphocytes and in plasma.Their main functions include the promotion of the germinal center formation and antibody type conversion.In patients with NMOSD,peripheral blood membrane-bound ICOS/ICOSL and PD-1/PDL1 are overexpressed in the acute phase,tend to decline during the remission phase but then rise again on relapse.In other autoimmune diseases,abnormal expression of both membrane-bound and soluble molecules is often present and correlates with disease activity,and can be an important indicator of disease activity monitoring.While the detection of soluble molecules is relatively convenient in clinical applications,the significance of soluble ICOS/ICOSL(sICOS/sICOSL)and PD-1/PD-L1(sPD-1/sPD-L1)in the assessment of the efficacy and prognosis of NMOSD remains to be investigated.In this study,based on the modified reduced-dose RTX(mRTX)regimen previously proposed by our research group,we compared the efficacy and safety of mRTX with other immunosuppressive agents in the prevention of NMOSD relapse using repeated dosing when total CD19+B cells exceeded 3%of lymphocytes,and analyzed in depth the proliferation characteristics of B cells under the mRTX regimen and the significance of dynamic monitoring of B cells in guiding the repeated dosing of mRTX.We also explored the value of sICOS/sICOSL and sPD-1/sPD-L1 in predicting NMOSD relapse,aiming to provide a new basis for disease monitoring and relapse prediction in patients with NMOSD.Part Ⅰ Efficacy and safety of a modified reduced-dose rituximab regimen in patients with NMOSDObjective:To evaluate the efficacy and safety of a modified reduced-dose rituximab(mRTX)regimen compared with azathioprine(AZA)and MMF in patients with NMOSD.Methods:NMOSD patients treated with AZA,MMF or mRTX in our neurology department were enrolled from 1 January 2015 to 30 June 2021.The primary endpoint event was the initial relapse after immunosuppressant therapy,and the annualized relapse rate(ARR),expanded disability status scale(EDSS)score,and treatment-related adverse events were compared between groups.The differences between three immunosuppressive agents in reducing the risk of recurrence of NMOSD were assessed by the Cox proportional hazard model.Results:In this retrospective cohort study,91 patients with NMOSD were included,15 men and 76 women,with a median age of 42(28,53)years.Among them,24 patients were treated with AZA,30 with MMF,and 37 with mRTX.The relapse-free rate was 33.3%,83.3%and 83.8%in the AZA,MMF and mRTX groups respectively.The ARR was significantly lower in all three groups,and the EDSS score was also significantly lower in the mRTX and MMF groups after treatment,while the EDSS score was slightly higher in the AZA group.mRTX and MMF were both superior to AZA in reducing the EDSS score.In the univariate Cox proportional risk model,both mRTX(HR=0.245,95%CI:0.096-0.629,P=0.003)and MMF(HR=0.265,95%CI:0.096-0.727,P=0.010)significantly reduced the risk of relapse in patients with NMOSD compared to AZA.In contrast,relapse within 1 year before immunosuppressant therapy(HR=3.308,95%CI:1.140-9.605,P=0.028)or ARR before immunosuppressant therapy(HR=1.313,95%CI:1.031-1.672,P=0.027)both increased the risk of NMOSD relapse.Multivariate Cox proportional hazard model analysis showed that both mRTX and MMF significantly reduced the risk of NMOSD relapse even after adjustment for other factors,while relapse within 1 year before immunosuppressant therapy still increased the risk of NMOSD relapse.Additionally,mRTX-treated patients were less likely to use steroids concurrently and had fewer overall adverse events than those treated with AZA and MMF.Conclusion:Compared with AZA,mRTX and MMF significantly reduced the risk of NMOSD relapse.Even after treatment with immunosuppressive agents,relapse within 1 year before immunosuppressant therapy still increased the risk of future relapse in NMOSD patients.mRTX-treated patients presented less concomitant steroid use and fewer adverse events.Part Ⅱ Analysis on the results of B-cell dynamic monitoring in NMOSD treated with a modified reduced-dose rituximab regimen:a case series studyObjective:To explore the proliferation characteristics of B cells under the mRTX regimen and the significance of B cell monitoring in guiding the individualized dosing of mRTX.Methods:NMOSD patients who were treated with the mRTX regimen for at least 6 months at our neurology department were collected from 1 January 2015 to 31 December 2021.The expression of total CD19+B cells and CD27+memory B cells in peripheral blood was measured by flow cytometry,and the proliferation of B cells was monitored dynamically.mRTX treatment was repeated when total CD 19+B cells exceeded 3%of lymphocytes,and relapse events and adverse effects were followed dynamically during dosing.The ARR and EDSS scores were compared before and after mRTX treatment,as well as the differences in B cells between the relapse and nonrelapse groups.Results:During the study period,34 patients with NMOSD were eventually included,4 males and 30 females,with a mean age of onset of 36.3±15.8 years.The mean follow-up time during mRTX treatment was 31.8 months.The EDSS score decreased from 2.5(1.5,3.0)to 1.3(1.0,2.0)and the ARR decreased from 1.0(0,2.0)to 0(0,0)after treatment,both with statistically significant differences(P<0.001).Relapses occurred in 6 patients(8 in total)with 26 doses of mRTX.The median total CD19+B cells were 3.25(2.7,3.7)%and the median CD27+memory B cells were 0.3(0.2,0.3)%before the initial relapse,with a mean dosing interval of 8.2 months(range:5.3-17.6 months).Other 28 patients(82.4%)remained relapse-free during mRTX treatment,with 84 doses of mRTX,and a median total CD19+B cells of 4.00(3.14,5.32)%and a median CD27+memory B cells of 0.26(0.17,0.40)%prior to 56 repeat doses.The mean dosing interval was 9.2 months(range:5.2-20.3 months).Both total CD19+B cells and CD27+memory B cells continued to proliferate over time after mRTX administration,with significantly faster proliferation rates in the later stages,but there was no correlation between the proliferation rates of these two types of B-cells(r=0.202,P=0.136).In the 28 relapse-free patients,the mean time to reach 1%was 210 days,the mean time to reach 3%was 240 days,and the mean interval from 1%to 3%was 65 days for total CD19+B cells.25 relapse-free patients had no significant difference in maximum,minimum and mean B cell values compared to 6 relapse patients(all P>0.05).By the end of follow-up,4 patients remained on a combination of small doses of steroids,and adverse events occurred in four patients,one of which was a significant adverse event.Conclusion:The high rate of early B-cell repopulation under the mRTX regimen requires closer dynamic B-cell monitoring to guide repeat mRTX dosing.Repeated mRTX with total CD 19+B cells above 3%significantly reduces ARR,prolongs the interval between repeated doses and is better tolerated overall.Part Ⅲ The expression of sICOSL in the peripheral blood and its significance in efficacy surveillance for NMOSD patientsObjective:To explore the expression of sICOSL in the peripheral blood and its significance in efficacy surveillance for NMOSD patients.Methods:Patients with NMOSD attending our neurology department were enrolled from January 1,2015 to December 31,2021.The levels of sICOS/sICOSL and sPD-1/sPD-L1 in the peripheral blood were measured by enzyme-linked immunosorbent assay.Patients were divided into the relapse and stable groups according to whether they relapsed or not,and the differences in clinical data and soluble costimulatory molecules between two groups were compared.The efficacy of peripheral blood soluble costimulatory molecule levels in predicting relapse status in NMOSD patients was evaluated using the receiver operating characteristic(ROC)curve.The relationship between soluble costimulatory molecules levels in the peripheral blood and NMOSD relapse was analyzed by univariate and multivariate logistic regression models.Results:A total of 79 patients with NMOSD were included,14 males and 65 females,with a mean onset age of 42.0± 15.1 years.Relapses occurred in 27 patients and 52 patients were stable.The proportion of patients using RTX in the stable group was significantly higher than that in the relapse group(P=0.001).Patients in the relapse group had significantly higher levels of sICOS(P=0.027),sICOSL(P<0.001),sPD-1(P<0.001)and sPD-L1(P<0.001)than those in the stable group.Multivariate logistic regression model showed that peripheral blood sICOSL(OR=1.051;95%CI:1.020-1.083;P=0.001),sPD-L1(OR=1.005;95%CI=1.001-1.009;P=0.007)were risk factors for predicting NMOSD relapse,while RTX(OR=0.105;95%CI=0.015-0.764;P=0.026)could reduce the relapse of NMOSD.There was a positive correlation between sICOSL and sPD-1(r=0.711,P<0.001),sPD-L1(r=0.506,P<0.001).A positive correlation was also indicated between sICOSL(r=0.331,P=0.003),sPD-1(r=0.309,P=0.006),and sPD-L1(r=0.308,P=0.006)and EDSS scores.The area under the ROC curve for sICOSL predicting NMOSD relapse status was 0.888,95%CI:0.797-0.948,P<0.001,with a cut-off value of 84.56 pg/ml,sensitivity of 92.59%and specificity of 86.54%,respectively,and the area under the ROC curve for sPD-L1 predicting NMOSD relapse status was 0.830,95%CI:0.729-0.905,P<0.001,with a cut-off value of 264.65 pg/ml,sensitivity of 85.19%and specificity of 78.85%.Subgroup analysis showed that sICOSL decreased significantly more than sPD-L1(P<0.001)in the stable phase compared to the acute attack phase.Patients in the RTX group had a higher level of sICOSL decrease during the stable phase than that in the non-RTX group,although the difference was not statistically significant(P=0.174).Conclusion:sICOSL and sPD-1/sPD-L1 in the peripheral blood evolve dynamically,and sICOSL and sPD-L1 can potentially be used to predict relapse in NMOSD patients.RTX significantly reduces the risk of NMOSD relapse,and sICOSL decreases at higher levels in the RTX group than other immunosuppressive agents.sICOSL has potential clinical value in assessing the efficacy of RTX for NMOSD. |