**3. Results**

#### *3.1. Demographic, Clinical, and Laboratory Findings*

The clinical characteristics of the cohorts from each country at the time of renal biopsy are presented in Table 1. In the Japanese and Korean cohorts, there were no significant differences in age or sex between patients with IgAN and the disease controls. Compared to Japanese patients with IgAN, Korean patients showed higher levels of proteinuria and lower kidney function at the time of renal biopsy.

**Table 1.** Clinical characteristics of patients with IgAN and disease controls at the time of renal biopsy.


IgAN, IgA nephropathy; DC, disease control; M, male; F, female; sCr, serum creatinine; eGFR, estimated glomerular filtration rate; UPCR, urinary protein-to-creatinine ratio.

Other renal diseases included lupus nephritis, anti-neutrophil cytoplasmic antibody (ANCA)-associated glomerulonephritis, membranous nephropathy, focal segmental glomerulosclerosis, minimal change disease, membranoproliferative glomerulonephritis, non-IgA mesangial proliferative glomerulonephritis, tubulointerstitial nephritis, diabetic kidney disease, renal amyloidosis, nephrosclerosis, and thin basement membrane disease (Table 2).




**Table 2.** *Cont.*

sCr, serum creatinine; eGFR, estimated glomerular filtration rate; UPCR, urinary protein-to-creatinine ratio; ANCA, anti-neutrophil cytoplasmic antibody.

#### *3.2. Urinary Levels of Gd-IgA1 in the Japanese Cohort*

In the Japanese cohort, urinary Gd-IgA1 levels were significantly elevated in patients with IgAN compared to those in disease controls (*p* < 0.0001) (Figure 1a). The levels of urinary Gd-IgA1 in each disease control group are shown in Figure 1b. Overall, urinary Gd-IgA1 levels were lower in the disease controls than in IgAN patients.

**Figure 1.** Urinary levels of Gd-IgA1 in Japanese cohort. (**a**) Urinary Gd-IgA1 levels were significantly elevated in patients with IgAN compared to those in disease controls. (**b**) Urinary Gd-IgA1 levels were lower in the disease controls than in IgAN patients. \*\*\* *p* < 0.0001.

#### *3.3. Correlation between Urinary Gd-IgA1 and Laboratory and Pathological Findings in IgAN Patients from the Japanese Cohort*

We assessed the association between urinary Gd-IgA1 levels and the clinical data and pathological parameters in patients with IgAN from the Japanese cohort. The levels of urinary Gd-IgA1 were associated with the levels of serum Gd-IgA1 (*p* < 0.0001) (Figure 2a), but not with proteinuria, in patients with IgAN (Figure 2b). As shown in Figure 2c, urinary Gd-IgA1 levels were positively correlated with the histological grade (*R* = 0.4108, *p* < 0.0001). Meanwhile, there were no significant correlations between the urinary Gd-IgA1 levels and the MEST-C scores of the Oxford classification. Then, we placed a threshold on the urinary Gd-IgA1 levels, i.e., >50 ng/mL, and analyzed the association between urinary Gd-IgA1 levels and MEST-C Oxford classification. As shown in Figure 2d, there were significant correlations between the urinary Gd-IgA1 levels and the T score of the Oxford classification in cases with urinary Gd-IgA1 greater than 50 ng/mL (*p* < 0.01).

**Figure 2.** Correlation between urinary Gd-IgA1 and laboratory and pathological findings in IgAN patients from the Japanese cohort. (**a**) The levels of urinary Gd-IgA1 associated with the levels of serum Gd-IgA1. (**b**) The levels of urinary Gd-IgA1 did not correlate with the levels of proteinuria. (**c**) Urinary Gd-IgA1 levels were positively correlated with the histological grade. (**d**) Urinary Gd-IgA1 levels were positively correlated with the T score of the Oxford classification in cases with urinary Gd-IgA1 greater than 50 ng/mL. \* *p* < 0.01, n.s.: not significant, Abbreviations: M (mesangial hypercellularity); E (endocapillary hypercellularity); S (segmental glomerulosclerosis); T (tubular atrophy/interstitial fibrosis); and C (cellular/fibrocellular crescents).

#### *3.4. Validation in Korean Cohort*

Next, we validated the use of urinary Gd-IgA1 levels for the diagnosis of IgAN using the Korean cohort. In the Korean cohort, the levels of urinary Gd-IgA1 in patients with IgAN were significantly higher compared with those in disease controls (*p* < 0.0001) (Figure 3a). Moreover, the levels of urinary Gd-IgA1 in patients with IgAN were higher than those in any of the disease controls (Figure 3b).

#### *3.5. Difference in Clinical Features at the Time of Renal Biopsy in Patients with IgAN*

Compared with the Japanese cohort, Korean and Taiwanese patients showed greater levels of proteinuria and lower kidney function at the time of renal biopsy (Table 3). In addition, levels of urinary Gd-IgA1 in the Korean and Taiwanese patients with IgAN were higher than those in the Japanese patients (Supplementary Figure S1).

**Figure 3.** Urinary levels of Gd-IgA1 in Korean and Taiwanese cohorts. (**a**) The levels of urinary Gd-IgA1 in Korean patients with IgAN were significantly higher compared with those in disease controls. (**b**) The levels of urinary Gd-IgA1 in Korean patients with IgAN were higher than those in the any other disease controls. \*\*\* *p* < 0.0001.


**Table 3.** Clinical features of patients with IgAN at the time of renal biopsy.

IgAN, IgA nephropathy; M, male; F, female; sCr, serum creatinine; eGFR, estimated glomerular filtration rate; UPCR, urinary protein-to-creatinine ratio.

#### *3.6. Urinary Gd-IgA1 Excretion with Trace Proteinuria*

We assessed the urinary Gd-IgA1 excretion in cases with trace proteinuria (less than 0.3 g/gCr) using the Japanese cohort. Even in cases with trace proteinuria, urinary Gd-IgA1 was detected, and the levels of urinary Gd-IgA1 were higher in patients with IgAN than in disease controls (Figure 4).

**Figure 4.** Urinary Gd-IgA1 excretion in cases with trace amounts of proteinuria (less than 0.3 g/gCr). \*\*\* *p* < 0.0001.
