Comparison of clinical features of nephrotic syndrome after haploidentical and matched donor hematopoietic stem cell transplantation

To the Editor: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is one of the most effective options for hematological diseases. However, allo-HSCT treatment can cause serious complications. Post-transplant kidney damage is an important complication. In this study, we retrospectively analyzed the frequency of nephrotic syndrome (NS) after allo-HSCT and for the first time compared the frequency and clinical characteristics of NS between haploidentical donor (HID) and matched donor (MD) HSCT (including matched sibling donors [MSD] and unrelated donors [URD]). Medical records of 8943 patients with hematological diseases who received either HID or MD HSCT in Peking University Institute of Hematology center from January 2006 to March 2022 and survived at least 100 days after their HSCT were enrolled in this study (HID, n = 6528; MD, n = 2415). NS was used as a retrieval condition to identify NS cases in this database. Matched patients (at a 1:5 ratio) with the same transplant time and transplant type (HID or MD) were selected as control group. The endpoint of the last follow-up was May 31, 2022. The study protocol was approved by the Ethics Committee of Peking University People's Hospital (No. 2020PHB376-01). The clinical activities were consistent with the Declaration of Helsinki. The definitions and statistical methods were presented in detail in the supplementary material, https://links.lww.com/CM9/B660. Forty-one patients (HID, n = 16; MSD, n = 22; URD, n = 3; comprised of 30 males and 11 females) had NS, with an incidence rate of 0.45% (41/8943). Among them, 16 patients received HID HSCT, giving an incidence rate of 0.25% (16/6528), while 25 patients received MD HSCT, giving an incidence rate of 1.04% (25/2415), with a statistically significant difference (χ2 = 24.100, P <0.001). The median time from transplantation to NS onset for HID HSCT recipients was earlier than that of MD HSCT, with a median time of 188 (30–1386) days in HID HSCT group and 529 (80–2334) days in MD HSCT group (U = 164.000, P = 0.041). The 24-h urine protein level at the time of NS diagnosis in HID HSCT group was significantly lower as compared to that of MD HSCT group (6.2 ± 3.3 g vs. 12.4 ± 12.8 g, U = 101.000, P = 0.009). The serum albumin level in HID HSCT group was significantly higher than that of MD HSCT group (25.4 ± 4.0 g/L vs. 20.2 ± 6.1 g/L, U = 93.000, P = 0.004). HID HSCT recipients with NS had a significantly higher proportion of anti-thymocyte globulin (ATG) administration in their conditioning regimen as compared with those receiving MD HSCT (100.0% [16/16] vs. 24.0% [6/25], χ2 = 22.700, P <0.001). Compared with MD HSCT, HID HSCT patients with NS had a longer median time of platelet engraftment (19 [8–70] days vs. 13 [8–73] days, U = 104.000, P = 0.042), a higher incidence of acute graft versus host disease (aGVHD, 75.0% vs. 20.0%, χ2 = 12.200, P <0.001) and CMV infection (75.0% vs. 36.0%, χ2 = 5.900, P = 0.015), as shown in Table 1. The potentially contributing causes of NS for HID and MD groups were different, as shown in Supplementary Table 1, https://links.lww.com/CM9/B660. Table 1 - Comparison of clinical characteristics between HID and MD allo-HSCT with NS. Characteristics HID allo-HSCT (n = 16) MD allo-HSCT (n = 25) Statistical value P-value Age at HSCT (years) 35 (7–59) 40 (9–62) 187.500 0.748* Patient sex (male/female) 12 (75.0)/4 (25.0) 18 (72.0)/7 (28.0) 0.042 0.833† Diagnosis 4.500 0.467† ALL 3 (18.75) 4 (16.0) AML 7 (43.75) 11 (44.0) PCL 1 (6.25) 0 ABL 0 1 (4.0) MPN 2 (12.5) 3 (12.0) MDS 1 (6.25) 5 (20.0) AA 2 (12.5) 1 (4.0) 24-h urine protein at diagnosis (g) 6.2 ± 3.3 12.4 ± 12.8 101.000 0.009* Serum albumin at diagnosis of NS (g/L) 25.4 ± 4.0 20.2 ± 6.1 93.000 0.004* Number of cases who underwent renal biopsy 6 (37.5) 19 (76.0) 6.100 0.014† Conditioning regimen containing ATG 16 (100.0) 6 (24.0) 22.700 <0.001† Time from HSCT to neutrophil engraftment (days) 15 (11–22) 13 (10–19) 132.000 0.138* Time from HSCT to platelet engraftment (days) 19 (8–70) 13 (8–73) 104.000 0.042* aGVHD 12 (75.0) 5 (20.0) 12.200 <0.001† cGVHD 14 (87.5) 24 (96.0) 1.000 0.308† CMV infection 12 (75.0) 9 (36.0) 5.900 0.015† Data are presented as median (range), n (%) or mean ± standard deviation. AA: Aplastic anemia; ABL: Acute bi-phenotypic leukemia; aGVHD: Acute graft versus host disease; ALL: Acute lymphoblastic leukemia; allo-HSCT: Allogeneic hematopoietic stem cell transplantation; AML: acute myeloid leukemia; ANCA: Anti-neutrophil cytoplasmic antibody; ATG: Anti-thymocyte globulin; cGVHD: Chronic graft versus host disease; CMML: Chronic myelomonocytic leukemia; CMV: Cytomegalovirus; HID: Haploidentical donor; HSCT: Hematopoietic stem cell transplantation; MD: Matched donor; MDS: Myelodysplastic syndrome; MPN: Myeloproliferative neoplasm; NS: Nephrotic syndrome; PCL: Plasma cell leukemia.*Mann-Whitney U test; †Fisher's exact tests. A total of 205 randomly matched patients with the same transplant time were selected as the control group (HID, n = 80; MSD, n = 109; URD, n = 16). There were no significant differences between NS and control groups in median age, patient sex, donor sex, disease diagnosis, transplantation type, and follow-up time (all P >0.05), as shown in Supplementary Table 2, http://links.lww.com/CM9/B660. Univariate analysis was performed for all patients in NS and control groups. We found that cGVHD (in particular, moderate to severe cGVHD) was associated with NS onset [Supplementary Table 3, https://links.lww.com/CM9/B660]. Patients with NS were treated in accordance with the treatment principles. For HID HSCT, 13 patients were treated with glucocorticoids and immunosuppressants, out of which six patients were achieving complete response (CR) and seven patients were achieving partial response (PR); one patient was only treated with glucocorticoid and remained not response (NR); one patient received only rituximab and obtained PR; one patient with antineutrophil cytoplasmic antibody (ANCA) associated small vasculitis renal impairment with IgA nephropathy received glucocorticoid in combination with cyclophosphamide, plasma exchange, and high-dose intravenous immunoglobulin, and then obtained PR. For MD HSCT group, 18 patients received glucocorticoid and immunosuppressive therapy, CR was observed in 10 patients, PR was observed in six patients, and NR was observed in two patients; five patients received only glucocorticoids, CR was observed in two patients, PR was observed in two patients, and NR was observed in one patient; one case of hepatitis B associated membranous nephropathy received entecavir and obtained CR; one patient with relapsed acute lymphoblastic leukemia had NR after chemotherapy. The CR rate of NS was 37.5% (6/16) and 52.0% (13/25) (χ2 = 0.800, P = 0.364), and the overall response rate (ORR) (CR + PR) was 93.8% (15/16) and 84.0% (21/25) (χ2 = 0.900, P = 0.352) in HID and MD HSCT groups, respectively. The ORR (CR + PR) was higher in patients receiving combination therapy compared to those receiving single-drug therapy (93.8% [30/32] vs. 66.7% [6/9], χ2 = 4.800, P = 0.043). The 5-year overall survival was 87.5% and 80.6% in HID and MD HSCT groups, respectively (χ2 = 0.065, P = 0.790). We further analyzed the outcome of 27 (including 11 cases of HID and 16 cases of MD) acute leukemia patients. The 5-year relapse-free survival was 80.8% and 63.2% in HID and MD HSCT groups (χ2 = 0.468, P = 0.594). The 5-year cumulative incidence of relapse in HID and MD groups was 18.1% and 36.0% (χ2 = 0.557, P = 0.455), while the 5-year cumulative incidence of NRM in HID and MD groups was 9.1% and 16.0% (χ2 = 0.097, P = 0.756) [Supplementary Figure 1, http://links.lww.com/CM9/B660]. In addition, the number of patients who developed chronic kidney disease (CKD) was two (12.5%) in HID group and three (12.0%) in MD group (χ2 = 0.0002, P = 0.305), with 2 patients (one in HID group and one in MD group) developing kidney failure requiring dialysis. Literature reported a variable incidence of NS after allo-HSCT[1–3]. However, no data are available regarding the incidence of NS after HID HSCT. The present study reported the overall frequency of NS in HID recipients after allo-HSCT. Though this is a retrospective study, the largest number of cases enrolled in this study makes it more reliable. We found that NS was less common in HID HSCT and the clinical manifestations of HID HSCT patients appeared to be milder. Mo et al[4] reported that conditioning regimens without ATG was one of the risk factors for non-malignant late effects post HSCT. Večerić-Haler et al[5] demonstrated that ATG followed by mesenchymal stem cell transplantation significantly improved injured renal function in mice. Since HID HSCT recipients had a higher proportion of ATG in their conditioning regimen, we speculated that ATG might protect the kidney and that the difference in the frequency and clinical manifestations of NS between HID and MD HSCT patients might be related to the different conditioning regimens. Previous literature believed that cGVHD was the main cause of NS.[1] However, we found that aGVHD and thrombotic microangiopathy (TMA) were also the causes of NS for HID HSCT. Since aGVHD and TMA mainly occurred in the early stage after transplantation, this also explained the fact that the onset time of NS for HID HSCT recipients was relatively early. But the mechanism of how aGVHD and TMA induce early NS remains unclear at present. Treatment with immunosuppression is reported to have variable success. Steroids and calcineurin inhibitors are the most commonly used drugs. There are other immunosuppressant drugs, such as cyclophosphamide, mycophenolate mofetil, chlorambucil, and rituximab, which have also been used. We demonstrated that combination therapy based on glucocorticoids and immunosuppressive or cytotoxic agents was more effective than single-drug therapy, providing a basis for the choice of treatment regimen for post-transplant NS in the future. There were limitations in this study. Only nearly 60.0% of NS cases underwent renal biopsy after NS diagnosis; as for the remaining 40.0% of patients, a diagnosis of NS was rendered possible exclusively by relying on clinical diagnostic criteria, and resultantly, in these latter cases, determination of the exact pathological type was not possible. In conclusion, the frequency of NS was lower and onset time was earlier after HID HSCT than those of MD HSCT. The etiology of NS after HID HSCT included not only cGVHD but also aGVHD and TMA. Post-transplant NS has a better outcome with combination therapy, and the overall prognosis is not influenced by the type of transplant. Funding This work was supported by grants from the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (No. 81621001) and Peking University People's Hospital Research and Development Funds (No. RDY2020-01). Conflicts of interest None.