Figure 4. Reduction in PANDAS IgG binding to cholinergic interneurons correlates with decreases in CY-BOCS score following IVIG treatment.

Sera were collected at baseline (pre-IVIG) and 6–12 weeks after intravenous immunoglobulin (IVIG) treatment (post-IVIG) and tested for CIN binding. (A) Representative images of co-staining of anti-human IgG (green) and anti-ChAT (red) before and after IVIG treatment. Arrowheads indicate antibody binding to CINs. Scale bar represents 40 μm. (B) IVIG treatment results in decreased PANDAS IgG binding to CINs in the original 5 sera (paired t-test: t[4]=6.870, p=0.002); this constitutes a technical replication of our previous analysis of the same samples, using a different assay (47). (C) The change in IgG binding to striatal CINs correlated at trend level with symptom improvement (Pearson’s correlation: r²=0.753, p=0.057). (D) IVIG treatment similarly produced decreased PANDAS IgG binding to CINs in the second cohort (t[5]=5.413, p=0.003). (E) Change in IgG binding to striatal CINs correlated significantly with symptom improvement in the second cohort (r²=0.747, p=0.026). (F) The consistent effect of IVIG treatment across the two cohorts was particularly apparent in pooled data (t[10]=8.191, p<0.0001) (G) In pooled data, the correlation between change in serum IgG binding to CINS and improvement in symptoms was particularly robust (r²=0.762, p=0.0005). Each data point represents mean value obtained from 4–6 mice for one serum. **p<0.01, ***p<0.001; N=5 in B and C, N=6 in D and E, N=11 in F and G, for each group. Dotted lines in correlation analyses indicate 95% confidence intervals. ●, – 5 serum pairs from the first cohort (47); ▲, – 6 serum pairs from the second cohort.