Serological Markers of Clinical Improvement in MuSK Myasthenia Gravis

Abstract

Background and objectives: In this retrospective longitudinal study, we aimed at exploring the role of (a) MuSK-immunoglobulin G (IgG) levels, (b) predominant MuSK-IgG subclasses, and (c) antibody affinity as candidate biomarkers of severity and outcomes in MuSK-MG, using and comparing different antibody testing techniques.

Methods: Total MuSK-IgGs were quantified with radioimmunoassay (RIA), ELISA, flow cytometry, and cell-based assay (CBA) serial dilutions using HEK293 cells transfected with MuSK-eGFP. MuSK-IgG subclasses were measured by flow cytometry. SAffCon assay was used for determining MuSK-IgG affinity.

Results: Forty-three serum samples were obtained at different time points from 20 patients with MuSK-MG (median age at onset: 48 years, interquartile range = 27.5-72.5; women, 16/20), with 9 of 20 (45%) treated with rituximab. A strong correlation between MuSK-IgG levels measured by flow cytometry and RIA titers was found (r_s = 0.74, 95% CI 0.41-0.89, p = 0.0003), as well as a moderate correlation between CBA end-point titers and RIA titers (r_s = 0.47, 95% CI 0.01-0.77, p = 0.0414). A significant correlation was found between MuSK-IgG flow cytometry levels and disease severity (r_s = 0.39, 95% CI 0.06-0.64, p = 0.0175; mixed-effects model estimate: 2.296e-06, std. error: 1.024e-06, t = 2.243, p = 0.032). In individual patients, clinical improvement was associated with decrease in MuSK-IgG levels, as measured by either flow cytometry or CBA end-point titration. In all samples, MuSK-IgG4 was the most frequent isotype (mean ± SD: 90.95% ± 13.89). A significant reduction of MuSK-IgG4 and, to a lesser extent, of MuSK-IgG2, was seen in patients with favorable clinical outcomes. A similar trend was confirmed in the subgroup of rituximab-treated patients. In a single patient, MuSK-IgG affinity increased during symptom exacerbation (K_D values: 62 nM vs 0.6 nM) while total MuSK-IgG and IgG4 levels remained stable, suggesting that affinity maturation may be a driver of clinical worsening.

Discussion: Our data support the quantification of MuSK antibodies by flow cytometry. Through a multimodal investigational approach, we showed that total MuSK-IgG levels, MuSK-IgG4 and MuSK-IgG2 levels, and MuSK-IgG affinity may represent promising biomarkers of disease outcomes in MuSK-MG.

Figure 1. MuSK-IgG Levels, End-Point Titers, and Clinical Status

Comparison of MuSK-IgG levels (antigen binding capacity, ABC) in serum samples collected from patients during an acute MG phase and at a second time point, when they either achieved a favorable (MGFA PIS: MM-or-better = YES) unfavorable (MGFA PIS: MM-or-better = NO) outcome. (A) A reduction of MuSK-IgG levels was found in patients who achieved a favorable clinical outcome (p = 0.0295), with no significant changes in those who did not. (B) A decrease in MuSK-IgG CBA end-point titers was associated with the achievement of an MGFA PIS of MM-or-better (p = 0.0156) while no significant changes were observed in patients who were still symptomatic after immunotherapy. (C) Mixed-effects modeling showed a significant positive correlation between MuSK-IgG levels (ABC) and clinical severity classified according to the MGFA clinical classification (r_s = 0.3885, 95% CI 0.06383–0,6387, p = 0.0175; mixed-effects model estimate: 2.296e-06, std. error: 1.024e-06, t = 2.243, p = 0.032) (a simple linear regression line is plotted; dotted lines show 95% CI). (D) Mixed-effects modeling did not show a significant correlation between MuSK-IgG CBA end-point titers and clinical severity classified according to the MGFA clinical classification (r_s = 0.2304, 95% CI-0.1110-0.5232, p = 0.1701; mixed-effects model estimate: 2.474e-04, std. error: 2.077e-04, t = 1.191, p = 0.242) (a simple linear regression line is plotted; dotted lines show 95% CI). MGFA = Myasthenia Gravis Foundation of America; MM = minimal manifestation; PIS = postintervention status.

Figure 2. MuSK-IgG Subclasses and Disease Severity

Scatter plots showing MuSK-IgG subclass levels (ABC) measured by flow cytometry live CBA and disease severity assessed by MGFA classification at sampling. Mixed-effects modeling revealed no significant correlations between disease severity and the levels of MuSK-IgG1 (r_s = 0.04202, 95% CI-0.295-0.3697, p = 0.8049; mixed-effects model estimate: 6.970e-05, std. error: 8.458e-05, t = 0.824, p = 0.416) (A), MuSK-IgG2 (r_s = 0.2342, 95% CI −0.1070 to 0.5261, p = 0.1629; mixed-effects model estimate: 1.290e-05, std. error: 1.175e-05, t = 1.097, p = 0.280) (B), and MuSK-IgG3 levels (r_s = −0.099, 95% CI −0.4181 to 0.2418, p = 0.5598; mixed-effects model estimate: −7.752e-06, std. error: 1.814e-05, t = −0.427, p = 0.671) (C) and a significant positive correlation with MuSK-IgG4 levels (r_s = 0.3808, 95% CI 0.05487–0.6334, p = 0.0201; mixed-effects model estimate: 2.740e-06, std. error: 1.129e-06, t = 2.426, p = 0.018) (D). A simple linear regression line is plotted; dotted lines show 95% CI. MGFA = Myasthenia Gravis Foundation of America.

Figure 3. MuSK-IgG Subclasses and Clinical Outcomes

(A) MuSK-IgG4 reduction was associated with achievement of a favorable clinical outcome (p = 0.0245). (B) No significant changes in MuSK-IgG4 levels were observed in patients who were still symptomatic after immunotherapy (p = 0.7422). (C) Pairwise comparison of MuSK-IgG2 levels show a significant reduction in patients who achieved an MGFA PIS of MM-or-better (p = 0.0085). (D) No significant changes in MuSK-IgG2 levels were observed in patients with an unfavorable clinical outcome (p = 0.5469). MGFA = Myasthenia Gravis Foundation of America; PIS = postintervention status.

Figure 4. MuSK-IgG Affinity and Clinical Status

(A) Affinity-concentration plot showing SAffCon assay results in longitudinal samples from the same patient, who had an MGFA PIS of MM-or-better at time of initial sampling, with antibody properties of K_D = 62 nM and concentration (antibody-binding sites) = 1,900 nM. The second serum sample was collected during a phase of clinical exacerbation (classified as “acute phase”), with antibodies showing a 100-fold increase in affinity (K_D = 0.6 nM) and a concentration = 140 nM. The K_D is plotted on the x-axis as –log10(M); thus, higher values correspond to lower K_D (and hence tighter binding). Antibody concentration is plotted on the y-axis as –log10(M); thus, lower values correspond to higher antibody concentrations. (B) Total MuSK-IgG levels (ABC), CBA end-point titers, and proportion of specific IgG4 of the same patient and time points showed in panel A, demonstrating the same end-point titers, a mild decrease in MuSK-IgG levels, and predominance of IgG4 at both time points (90.8% and 81.3%, respectively). (C) Affinity-concentration plot showing SAffCon assay results in longitudinal serum specimens of another patient, who was sampled during the acute MG phase at the first time point, with antibody properties of K_D = 2 nM and concentration = 160 nM. At the second time point, when a favorable clinical outcome was achieved (MGFA PIS: MM-or-better) after immunotherapy, the antibody properties were K_D = 1.3 nM and concentration = 24 nM. K_D and antibody concentrations are plotted as described above. (D) Total MuSK-IgG levels (ABC), CBA end-point titers, and proportion of specific IgG4 of the same patient and time points showed in panel C, demonstrating a reduction of both MuSK-IgG levels (ABC) and CBA end-point titers while the predominant MuSK-IgG subclass was IgG4 at both time points (99.6% and 99.1%, respectively). MGFA = Myasthenia Gravis Foundation of America; PIS = postintervention status.