Diagnostic Value of the Kappa Free Light Chain Index to Distinguish MOGAD, NMOSD, and MS
Apolline Tournier1, Antoine Gavoille1,2,3, Julie Pique1,4
1Service de Neurologie, Sclérose en Plaques, Pathologies de la Myéline et Neuro-inflammation, Hospices Civils de Lyon, Bron, France.
Background And Objectives:
The differential diagnosis between aquaporin-4-immunoglobulin G-positive neuromyelitis optica spectrum disorder (AQP4-NMOSD), myelin oligodendrocyte glycoprotein (MOG) antibody-associated disease (MOGAD), and multiple sclerosis (MS) can be complex. Kappa free light chain index (KFLC-Index) emerged as an effective biomarker for distinguishing patients with MS from patients with other conditions. The main aim of this study was to assess the diagnostic performance of KFLC-Index in differentiating MOGAD, AQP4-NMOSD, and MS and to compare it with CSF-restricted oligoclonal bands (OCB) performance.
Methods:
We conducted a retrospective case-control study involving 18 French centers through our national NOMADMUS database. Patients were eligible if they received MOGAD or AQP4-NMOSD diagnosis and if OCB status and KFLC-Index levels were available or could be measured retrospectively. As a comparator, we included a group of patients with MS from the Lyon center.
Results:
We included 303 patients with 140 MOGAD, 37 AQP4-NMOSD, and 126 MS. The sex ratio F/M was 1/1 in the MOGAD group, 5/1 in the NMOSD group, and 4/1 in the MS group. The median age was 38 years in the MOGAD group, 52 in the NMOSD group, and 32 in the MS group. The median KFLC-Index (interquartile range) was the lowest in the MOGAD group (0.60 [0.33-2.09]) with 71% of undetectable CSF-KFLC, intermediate in the NMOSD group (8.85 [3.04-16.78]), and the highest in the MS group (49.90 [16.31-113.39]). KFLC-Index had very good diagnostic performance in differentiating MOGAD from MS (area under the curve [AUC] 0.93, 95% CI 0.9-0.96) with an optimal threshold of 5.7 (95% CI 4.74-6.85), a sensitivity of 0.89 (0.83-0.94), and a specificity of 0.89 (0.83-0.94). KFLC-Index had good performance in differentiating NMOSD from MS (AUC 0.82, 95% CI 0.74-0.88) and MOGAD from NMOSD (AUC 0.77, 95% CI 0.69-0.86). KFLC-Index did not significantly outperformed OCB in separating MOGAD from MS and NMOSD from MS; however, the combination of the 2 tests was more performant than used separately.
Discussion:
KFLC-Index is a valuable tool in distinguishing MS from AQP4-NMOSD and MOGAD and provides additional information to that given by OCB. We suggest that KFLC-Index as an additional supporting criterion for the differential diagnosis between MOGAD and MS.
