In this retrospective study, children (≤18 years) with MCD diagnosed by kidney biopsy from 2010 to 2022 were recruited from Jinling hospital. We enrolled cases who met the following criteria: (1) 24-h urine protein to creatinine ratio (UP/CR) ≥0.2 g/g; (2) patients with MCD whose biopsies were performed at any age between 1 and 18 years. Individuals with one of the following conditions were excluded: (1) infectious diseases (e.g., hepatitis B or C, tuberculosis, AIDS); (2) individuals of contemporary autoimmune disease, diabetes mellitus, malignant tumors; (3) drug-induced NS; (4) missing data on IgM deposition in renal pathology; (5) the glomerular number on light microscopy <10.

All clinical and pathological data were obtained retrospectively from medical records. The baseline investigations were taken during the kidney biopsy. Follow-up data, including treatment, steroid response, time to the first remission and relapse, was collected and analyzed at 1, 2, 3, 6, 12 months and longer follow-up. The last follow-up time was estimated from the biopsy to the previous encounter in our center. The diagnosis of hypertension was based on the 2017 AAP Blood Pressure Clinical Practice Guidelines (6). The estimated glomerular filtration rate (eGFR) was calculated using the Schwartz formula for children aged 16 years or less (7), and the CKD-EPI recipe for children aged 17 years or older (8).

(1) IgM-positive was defined as the intensity of the IgM staining by immunofluorescence ≥1+. (2) Grade of tubulointerstitial injury scored as none (<10%), mild (10%–24%), moderate (25%–49%), and severe (≥50%), with the percentages mentioned in parentheses. The change of chronic tubulointerstitial injury (CTI) indicated tubular atrophy/interstitial fibrosis. The manifestation of acute tubulointerstitial injury (ATI) was interstitial edema, inflammatory cellular infiltration, and necrosis in tubular epithelia. (3) Steroid-sensitive was defined as CR after the 4 weeks of full-dose oral corticosteroid (equivalent to a prednisone dose of 60 mg/m²/day or 2 mg/kg/day) and no relapse during corticosteroid therapy. (4) Steroid-resistant was defined as a lack of CR for at least 4 weeks of full-dose oral prednisone. (5) Steroid-dependent was defined as follows: (1) Reach CR after the 4 weeks of full-dose oral corticosteroid; (2) Two consecutive relapses during corticosteroid therapy or within 15 days of discontinuing corticosteroid therapy; (3) Infectious factors should be excluded at the same time; (4) When relapse occurs, patients could reach CR once the corticosteroid dose is increased (either at full dose or beyond the relapse dose).

Partial remission (PR) was a decrease in 24-h UP/Cr to <2 g/g but >0.2 g/g (or >20 and <200 mg/mmol) with a 50% reduction from its peak value, or the test result of the dipstick decreases by at least 1+, but it does not reach negative. CR was a decrease in 24-h UP/Cr ≤0.2 g/g (or 20 mg/mmol or negative or trace dipstick) on three or more consecutive occasions. Remission was the achievement of PR or CR. Relapse was defined as an increase in UP/Cr to ≥2.0 g/g in cases with a 50% increase from its valley value or in urine dipstick to ≥3+after the remission. Frequent relapse was defined as ≥2 relapses per 6 months or ≥4 relapses per 12 months.

This cohort study's primary and secondary endpoints were the first remission and the first relapse, respectively. When patients achieved the first remission after biopsy, we considered this the first remission. And the time interval from treatment to the first day of remission would be recorded. Time to the first relapse was the time interval from the first remission to the occurrence of the first relapse.

The SPSS 26.0, Graph Pad Prism 8.0, and Adobe Illustrator 2020 were used for statistical analysis in this cohort study. The continuous variables were compared between groups using the Mann-Whitney U test, represented by the median (lower quartile, upper quartile), while categorical variables using the chi-square test or Fisher exact test were expressed as frequency (percentage). All data were tested two-sided, with P < 0.05 regarded as statistically significant.

As shown in Figure 1, from 2010 to 2022, 283 children were diagnosed with MCD based on renal biopsy in our single center, including 136 (48.1%) in the IgM-positive group and 147 (51.9%) in the IgM-negative group. Baseline demographic characteristics and clinical features at renal biopsies of these children are listed in Table 1. Compared the two groups, there were no statistically significant differences in age of onset, gender, BMI, hypertension, duration, serum IgG, serum IgM, serum C3, proteinuria levels, hematuria cases, Acute Kidney Injury (AKI), total glomeruli, ATI and CTI (All P > 0.05). No severe ATI and moderate to severe CTI were observed in the renal pathology of all MCD individuals.

The median age at the renal biopsies of patients was 12.6 years old for the IgM-positive group and 15.2 years for the IgM-negative group (P < 0.001). In immunofluorescence, 16 (11.8%) patients also had positive C3 deposition in the IgM-positive group, while only 7 (4.8%) were in the IgM-negative group (P = 0.048). The IgM-positive group had a lower level of serum albumin (median, IQR: 22, 17.7–27.9 g/L vs. 24, 19.7–31.5 g/L, P = 0.02) and a higher level of eGFR (median, IQR: 171.9, 140.0–213.8 ml/min/1.73 m² vs. 148.9, 134.1–191.0 ml/min/1.73 m², P = 0.002) than the IgM-negative group.

A summary of treatment and outcomes is shown in Table 2. The median follow-up duration is 30.5 months for the IgM-positive group and 49 months for the IgM-negative group. Steroid therapy was administered to 283 patients, 42 of whom were treated with angiotensin-converting enzyme inhibitors (ACEI). 210 patients received immunosuppressive agents other than steroid therapy, which included 107 (78.7%) cases for the IgM-positive group and 103 (70.1%) for the IgM-negative group. There were no statistically significant differences in the data collected above. But the IgM-positive group had a significantly higher proportion of patients who received rituximab (22.6% vs. 11.6%, P = 0.025) than the IgM-negative group. More details of immunosuppressive therapy for the steroid-sensitive nephrotic syndrome (SSNS), steroid-dependent nephrotic syndrome (SDNS) and steroid-resistant nephrotic syndrome (SRNS) patients are shown in Supplementary Tables S1–S4.

The IgM-positive group was less responsive to steroid therapy than the IgM-negative group (P = 0.001). For instance, the IgM-positive group has fewer steroid-sensitive patients (23.5% vs. 40.8%), more steroid-dependent (74% vs. 51%) and steroid-resistant patients (18.4% vs. 8.2%) than the IgM-negative group. 283 (100%) patients reached the first remission during a median follow-up time of 30 days. The median time to the first remission was significantly different between the two groups (median, IQR; IgM-negative group = 27, 15–37 days vs. IgM-positive group = 33, 19–50 days, P = 0.004). 171 (60.8%) patients relapsed at least once during a median follow-up time of 200 days. Although the frequencies of frequent relapses (20.6% vs. 19.7%) and infrequent relapses (44.9% vs. 36.1%) were more common in the IgM-positive group than in the IgM-negative group, there was no statistically significant difference in relapse between the two groups (P = 0.182).

In total, 260 cases were followed up for more than 6 months. And we included these patients to detect the predictors of the first remission and relapse.

The Kaplan-Meier survival analysis showed that the IgM-positive group had a lower cumulative rate of the first remission than the IgM-negative group (Log-rank, P < 0.001) (Figure 2A). The 1-, 2-, 3- and 6-month remission rates were 57.14% and 43.42%, 94.56% and 85.86%, 96.60% and 92.56%, 100.00% and 98.51% for the IgM-negative group and the IgM-positive group, respectively. The result of the univariate Cox regression analysis is shown in Figure 2B. IgM deposition (hazard ratio [HR] = 0.639, 95% confidence interval [CI] = 0.503–0.810, P < 0.001), serum albumin (HR = 1.034, 95% CI = 1.019–1.050, P < 0.001), steroid-resistant (Referance = steroid-sensitive; HR = 0.411, 95% CI = 0.274–0.616, P < 0.001) were associated with the first remission. The result of the multivariate Cox regression analysis is listed in Figure 2C. IgM deposition (HR = 0.625, 95% CI = 0.501–0.848, P = 0.001), serum albumin (HR = 1.031, 95% CI = 1.014–1.049, P < 0.001), steroid-dependent (Referance = steroid-sensitive; HR = 0.749, 95% CI = 0.565–0.994, P = 0.045) and steroid-resistant (Referance = steroid-sensitive; HR = 0.453, 95% CI = 0.297–0.690, P < 0.001) were independently association with the first remission.

However, there is no statistical difference in relapses between the positive and negative groups (IgM-negative group = 23.4% vs. IgM-positive group = 32.8%, P > 0.05) (Figure 2D). Moreover, as shown in the Figures 2E, F, the univariate (HR = 1.493, 95% CI = 0.939–2.374, P = 0.090) and multivariate analysis (HR = 1.345, 95% CI = 0.814–2.224, P = 0.247) showed that IgM deposition was not associated with the first relapse. Serum albumin (HR = 0.956, 95% CI = 0.927–0.987, P = 0.005) and steroid-dependent (Referance = steroid-sensitive; HR = 3.458, 95% CI = 1.806–6.623, P < 0.001) were independently associated with the first relapse in the univariate Cox analysis. In the multivariate Cox analysis, serum albumin (HR = 0.961, 95% CI = 0.928–0.996, P = 0.028) and steroid-dependent (Referance = steroid-sensitive; HR = 3.498, 95% CI = 1.083–6.787, P < 0.001) were still independently associated with the first relapse.

We suggested that a significant portion of patients relapsed after a follow-up of 1 year. Therefore, extended follow-up may be necessary to demonstrate the significance of IgM deposition in relapse. Finally, 237 patients with a follow-up of more than 1 year were included to detect the predictors of the first relapse. The results are shown in Figure 3. Figure 3A shows that the IgM-positive group had a significantly higher proportion of patients relapse (47.2% vs. 34.4%, P = 0.047). Figure 3B shows that the IgM-positive group has a significantly higher cumulative rate of the first relapse (Log-rank, P = 0.034). The 3-month, 6-month, 1-year and 2-year remission rates were 12.98% and 18.87%, 21.37% and 30.19%, 34.35% and 41.17%, 34.35% and 47.17% for the IgM-negative group and the IgM-positive group, respectively. The univariate Cox regression analysis showed IgM deposition (HR = 1.534, 95% CI = 1.025–2.296, P = 0.037) and serum albumin (HR = 0.965, 95% CI = 0.940–0.991, P = 0.009) were associated with the first relapse. Figure 3D shows that IgM deposition (HR = 1.549, 95% CI = 1.010–2.374, P = 0.045) and serum albumin (HR = 0.965, 95% CI = 0.940–0.991, P = 0.009) are independently associated with the first relapse.