TY - JOUR
T1 - Novel index, neutrophil percentage (%) is a useful marker for disease activity in MOG antibody-associated disease
AU - Baek, Song Ik
AU - Ro, Suho
AU - Chung, Yeon Hak
AU - Ju, Hyunjin
AU - Kwon, Soonwook
AU - Park, Kyung Ah
AU - Min, Ju Hong
N1 - Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/8
Y1 - 2023/8
N2 - Background: Myelin oligodendrocyte glycoprotein antibody–associated disease (MOGAD) is a CNS autoimmune disease affecting the brain, spinal cord, and optic nerve. The neutrophil-to-lymphocyte ratio (NLR) is related to autoimmune disease activity. However, the clinical implication of index ratios such as the NLR is unclear in patients with MOGAD. Objectives: We investigated the relationship between index ratios such as the NLR and disease activity and disability to discover the index that best correlates with an attack in MOGAD. Methods: Using a CNS demyelinating disease cohort, we reviewed 39 patients with MOGAD (age 37.4 ± 12.0 years; F:M = 20:19) who had 390 blood samples available for cell count analysis. We calculated the NLR, eosinophil-to-lymphocyte-ratio (ELR), platelet-to-lymphocyte-ratio (PLR), monocyte-to-lymphocyte ratio (MLR), basophil-to-lymphocyte ratio (BLR), and neutrophil percentage (N%) [neutrophil count (/mm3) / WBC (/mm3) x 100 (%)]. We investigated the associations between each index ratio and disease activity and disability using the receiver operating characteristic (ROC) curve, machine learning program (kNN algorithm), and generalized estimating equations (GEE) analysis. Results: In patients with MOGAD, the NLR, PLR, and N% were higher and ELR was lower during an attack than in remission (all p<0.001). The areas under the ROC curve for the NLR, ELR, PLR, and N% were 0.68, 0.69, 0.61, and 0.68, respectively, with the highest sensitivity of 76.0% in the ELR and the highest specificity of 76.3% in the N%. The classification accuracy scores of the kNN machine learning algorithm were 71% for the NLR, 62% for the ELR, 63% for the PLR, and 72% for the N%. In the GEE analysis of attack samples, both the NLR and treatment-naive had positive associations with the Expanded Disability Status Scale (EDSS) score (β=0.137, p = 0.008 and β=1.142, p = 0.003, respectively), and the PLR was negatively associated with the EDSS score (β=-0.004, p = 0.022). Discussion: Our study suggests that the novel index, neutrophil% is the simplest and the most useful marker to differentiate between attack and remission and shows comparable reliability with NLR in MOGAD. Moreover, the NLR and PLR could be used as supportive biomarkers for disease disability during an attack in patients with MOGAD.
AB - Background: Myelin oligodendrocyte glycoprotein antibody–associated disease (MOGAD) is a CNS autoimmune disease affecting the brain, spinal cord, and optic nerve. The neutrophil-to-lymphocyte ratio (NLR) is related to autoimmune disease activity. However, the clinical implication of index ratios such as the NLR is unclear in patients with MOGAD. Objectives: We investigated the relationship between index ratios such as the NLR and disease activity and disability to discover the index that best correlates with an attack in MOGAD. Methods: Using a CNS demyelinating disease cohort, we reviewed 39 patients with MOGAD (age 37.4 ± 12.0 years; F:M = 20:19) who had 390 blood samples available for cell count analysis. We calculated the NLR, eosinophil-to-lymphocyte-ratio (ELR), platelet-to-lymphocyte-ratio (PLR), monocyte-to-lymphocyte ratio (MLR), basophil-to-lymphocyte ratio (BLR), and neutrophil percentage (N%) [neutrophil count (/mm3) / WBC (/mm3) x 100 (%)]. We investigated the associations between each index ratio and disease activity and disability using the receiver operating characteristic (ROC) curve, machine learning program (kNN algorithm), and generalized estimating equations (GEE) analysis. Results: In patients with MOGAD, the NLR, PLR, and N% were higher and ELR was lower during an attack than in remission (all p<0.001). The areas under the ROC curve for the NLR, ELR, PLR, and N% were 0.68, 0.69, 0.61, and 0.68, respectively, with the highest sensitivity of 76.0% in the ELR and the highest specificity of 76.3% in the N%. The classification accuracy scores of the kNN machine learning algorithm were 71% for the NLR, 62% for the ELR, 63% for the PLR, and 72% for the N%. In the GEE analysis of attack samples, both the NLR and treatment-naive had positive associations with the Expanded Disability Status Scale (EDSS) score (β=0.137, p = 0.008 and β=1.142, p = 0.003, respectively), and the PLR was negatively associated with the EDSS score (β=-0.004, p = 0.022). Discussion: Our study suggests that the novel index, neutrophil% is the simplest and the most useful marker to differentiate between attack and remission and shows comparable reliability with NLR in MOGAD. Moreover, the NLR and PLR could be used as supportive biomarkers for disease disability during an attack in patients with MOGAD.
KW - Disease activity
KW - EDSS
KW - MOG antibody-associated disease (MOGAD)
KW - Neutrophil percentage (N%)
KW - Neutrophil-to-lymphocyte ratio (NLR)
UR - https://www.scopus.com/pages/publications/85161476796
U2 - 10.1016/j.msard.2023.104796
DO - 10.1016/j.msard.2023.104796
M3 - Article
C2 - 37320937
AN - SCOPUS:85161476796
SN - 2211-0348
VL - 76
JO - Multiple Sclerosis and Related Disorders
JF - Multiple Sclerosis and Related Disorders
M1 - 104796
ER -