Feasibility and outcomes of kidney transplantation in patients with end‐stage kidney disease after haemopoietic stem cell transplantation

Abstract

Background

Advances in haemopoietic stem cell transplantation (HSCT) have improved long‐term survival, but they have also led to late complications, such as nephropathy. However, the safety and feasibility of kidney transplantation (KT) in patients with HSCT‐related end‐stage kidney disease (ESKD) remain unclear.

Aim

This study aimed to evaluate the feasibility and clinical outcomes of KT in patients with ESKD following HSCT.

Methods

This study retrospectively analysed eight patients with ESKD who underwent KT following HSCT for haematologic disorders, including aplastic anaemia (n = 5), acute myeloid leukaemia (n = 2) and chronic myeloid leukaemia (n = 1).

Results

The median time to progression to ESKD was 104 months after HSCT, with a median interval of 138 months between HSCT and KT. The causes of ESKD included diabetic nephropathy (n = 1), hypertensive nephropathy (n = 1), focal segmental glomerulosclerosis (n = 1), thrombotic microangiopathy (n = 1) and unknown aetiology (n = 4), with three diagnoses confirmed through kidney biopsy. Three patients received both HSCT and KT from the same donor, allowing them to completely discontinue immunosuppressive therapy. During a median follow‐up of 55 months (range 2–143 months), there was only one case of acute T‐cell‐mediated rejection, and one patient died from sepsis due to a urinary tract infection. Notably, none of the patients experienced a recurrence of their original haematologic disease.

Conclusion

Despite concerns regarding immunosuppression and the risk of infection in this high‐risk population, both patient and graft outcomes were favourable. These findings suggest that KT is a feasible and effective treatment option for patients with ESKD following HSCT, particularly when managed with personalised strategies.

Introduction

Haemopoietic stem cell transplantation (HSCT) is a curative treatment for various haematologic diseases, including acute and chronic leukaemias, as well as bone marrow (BM) failure syndromes like aplastic anaemia. ¹ Advancements in transplantation techniques, supportive care and immunosuppressive regimens have significantly improved the survival rates of patients undergoing HSCT. As a result, long‐term complications, particularly those affecting non‐haematologic organs, have become increasingly important. Among these complications, chronic kidney disease (CKD) and its progression to end‐stage kidney disease (ESKD) are serious outcomes that can severely impair quality of life and overall survival. ² , ³

The pathogenesis of ESKD after HSCT is multifactorial. It includes direct nephrotoxicity from conditioning regimens, such as total body irradiation and calcineurin inhibitors, as well as thrombotic microangiopathy, opportunistic infections and metabolic complications like hypertension and diabetes mellitus. ³ , ⁴ As survival rates after HSCT improve, an increasing number of patients with HSCT‐related ESKD may be eligible for kidney transplantation (KT). However, the safety and feasibility of KT for this specific population are not well established. There are concerns about the potential risks of immunologic complications, infections, and the recurrence of the underlying haematologic disease. Previous studies largely consisted of case reports or small series, and there is currently no consensus on the optimal approach for managing patients who undergo both HSCT and KT. ³ , ⁵ , ⁶

Notably, when the same donor provides both HSCT and KT, immune tolerance may allow for minimisation or even withdrawal of immunosuppressive therapy. ⁵ , ⁶ , ⁷ When donors differ, conventional immunosuppressive protocols are generally necessary, which raises concerns about infection risk and long‐term safety.

In this study, we retrospectively analysed the clinical outcomes of eight patients who developed ESKD after HSCT and subsequently underwent KT. Our goal was to assess whether KT could be performed safely and effectively in patients with ESKD following HSCT, with a specific focus on graft outcomes, infection risk and control of haematologic disease with post‐KT immunosuppression.

Methods

In this retrospective study, we extracted data from the medical records of recipients who underwent KT following HSCT at Seoul St. Mary's Hospital.

Haemopoietic stem cells transplantation procedure

All patients underwent allogeneic HSCT at our institution between 1999 and 2017. In the earlier era of HSCT, BM grafts were predominantly used, while peripheral blood (PB) stem cell grafts became more common in later years. Human leukocyte antigen (HLA) typing for donor selection initially relied on low‐resolution techniques covering the HLA‐A, ‐B, ‐C and ‐DRB1 loci but transitioned to high‐resolution molecular methods. HSCTs were performed using matched sibling donors (MSDs), unrelated donors (UDs) (7/8 or 8/8 matched) or haploidentical donors (HIDs) (4/8 matched). For conditioning regimens for aplastic anaemia, cyclophosphamide (50–60 mg/kg IV for 2 days) and anti‐thymocyte globulin (2.5 mg/kg for 4 days) were the mainstay for MSD transplants, while fludarabine and total body irradiation were selectively added for MUD or HID transplants. ⁸ , ⁹ , ¹⁰ , ¹¹

In HSCT for leukaemia, conditioning regimens included cyclophosphamide (60 mg/kg/day for 2 days) combined with total body irradiation (330 cGy/day for 4 days) or Busulfex (3.2 mg/kg/day for 4 days) combined with cyclophosphamide. ¹² For graft‐versus‐host disease (GVHD) prophylaxis, a calcineurin inhibitor – cyclosporin A for MSD or tacrolimus for other donor types – was routinely administered in combination with a short course of methotrexate. This treatment was then tapered based on GVHD activity or the risk of relapse.

Kidney transplantation procedure

Following HSCT, KT was performed between 2012 and 2025 to treat HSCT‐related ESKD. After KT, induction and maintenance immunosuppression (IS) regimens were administered as previously dscrebed. ¹³ , ¹⁴ Briefly, anti‐CD25 antibody (basiliximab) or anti‐thymocyte globulin (ATG) was used for induction therapy, and triple maintenance immunosuppression, including tacrolimus, mycophenolate mofetil (MMF) and prednisolone, was initiated 2 days before transplantation. For highly sensitised patients, we employed a desensitisation protocol consisting of rituximab, plasmapheresis (PP) and intravenous immunoglobulin. ¹³ , ¹⁴ The degree of PP was determined based on the crossmatch or donor‐specific HLA antibody (DSA) level, and maintenance IS was initiated 7 days prior to KT. ¹³ , ¹⁵

Parameters of clinical outcomes

The primary outcome was patient survival. The secondary outcomes included the allograft survival rate and clinical outcomes such as allograft function, allograft failure, acute rejection and infectious complications. The initiation of dialysis indicated allograft kidney failure. The glomerular filtration rate (GFR) was calculated using the CKD‐EPI 2021 equation, with the baseline estimated GFR determined based on the highest serum creatinine level measured before KT. Surgical complications included bleeding and infarction after the operation, regardless of whether reoperation was required.

Statistical analysis

In descriptive statistics, quantitative variables are summarised using the median and range for continuous data, while categorical variables are reported as actual values and percentages of patients. Survival estimates were calculated using the Kaplan–Meier method. All statistical analyses were conducted with spss (IBM spss Statistics, version 29).

Ethics statement

This study was approved by the Institutional Review Board of The Catholic University of Korea, Seoul, St. Mary's Hospital (IRB No. KC25RASI0821), and was conducted according to the principles expressed in the Declaration of Helsinki. Informed consent was not required because this study involved a retrospective analysis with no more than minimal risk to the subjects.

Results

Baseline characteristics at HSCT and occurrence of ESKD

Table 1 summarises the baseline characteristics of patients who underwent HSCT. Eight patients (four with aplastic anaemia, three with acute myeloid leukaemia and one with chronic myeloid leukaemia) received allogeneic HSCT between 1999 and 2017, with a median age of 23 years (range, 14–52 years) at the time of transplantation. Graft sources included BM (n = 5) and PB stem cells (n = 3). Donors consisted of one partially matched sibling (case 1), three fully matched siblings (cases 4, 6 and 8), three unrelated donors (cases 2, 3 and 7) and one haploidentical donor (case 5). HLA mismatches involved HLA‐DR (case 1), HLA‐A (case 7) and combined HLA‐A/B/DR (case 5). Patients developed ESKD at a median of 104 months (range, 5–260 months) after HSCT. The original causes of ESKD varied: one patient had diabetic nephropathy, one had hypertensive nephropathy, one had thrombotic microangiopathy, one had focal segmental glomerulosclerosis, and two patients had unknown cause. In three cases, the cause of ESKD was clinically diagnosed via percutaneous kidney biopsy.

Table 1.

Baseline characteristics of patients

Case no.	Sex	Haematologic disease	Type of HSCT	HSCT donor	Age at HSCT (years)	HLA MN in HSCT		GVHD	Age at ESKD (years)	Aetiology ESKD
						A, B	DR
1	F	AA	BMT	Sister	15	0	1	Chronic	16	TMA
2	M	AML	BMT	Unrelated	37	0	0	Acute	48	Unknown
3	M	AML	BMT	Unrelated	14	0	0	Chronic	22	Unknown
4	M	AA	BMT	Sister	33	0	0	None	55	FSGS
5	F	AA	PBSCT	Mother	26	2	1	Chronic	27	Unknown
6	F	AA	PBSCT	Sister	52	0	0	None	55	DM
7	F	AA	PBSCT	Unrelated	20	1	0	None	29	HTN
8	M	CML	BMT	Brother	20	0	0	None	37	Unknown

AA, aplastic anaemia; AML, acute myeloid leukaemia; BMT, bone marrow transplantation; CML, chronic myeloid leukaemia; DM, diabetes mellitus; ESKD, end‐stage kidney disease; F, female; FSGS, focal segmental glomerulosclerosis; GVHD, graft versus host disease; HLA, human leukocyte antigen; HSCT, haematopoietic stem cell transplantation; HTN, hypertension; M, male; MN, mismatch number; PBSCT, peripheral blood stem cell transplantation; TMA, thrombotic microangiopathy.

KT characteristics and peri‐transplant immunosuppression

The patients underwent KT at a median of 32 months (range, 0–107 months) following the onset of ESKD. The median interval between HSCT and KT was 138 months (range: 40–260 months). At the time of KT, all patients were in complete haematologic remission, and GVHD was inactive. All but one patient (case 3) received their transplant at our institution; case 3 had KT at an unidentified hospital in China, so details regarding the donor, pre‐transplant evaluation and surgical procedure are unavailable. Prior to KT, case 4 was on low‐dose oral prednisolone for focal segmental glomerulosclerosis, and cases 6 and 8 were taking oral hypoglycaemic agents for diabetes mellitus. The remaining patients did not require any additional pre‐KT management.

Clinical characteristics related to KT, and the initial immunosuppressive regimens are summarised in Table 2. Cases 1, 5 and 8 received KT and HSCT from the same donor and were temporarily treated with tacrolimus, mycophenolate sodium and either methylprednisolone or deflazacort, which were discontinued after a median of 28 months (range: 9–46 months). Following the discontinuation of immunosuppressive therapy, the patients were monitored under surveillance without evidence of rejection for a median of 25 months (range: 8–87 months). The remaining five patients were treated with the institution's standard regimen.

Table 2.

Immunological status of KT patients

Case no.	Age at KT (years)	ABO compatibility	Same donor for HSCT and KT	KT donor	HLA mismatch in KT		CDC crossmatch	Pre‐formed DSA	Immunosuppression
					A, B	DR			Pre‐KT	Induction	Initial maintenance IS	IS discontinue
1	20	Compatible	Yes	Sister	0	0	–	–	None	BAX	Tac + MMF + Pd	Yes
2	48	Incompatible	No	Spouse	4	2	–	–	None	BAX, RTX	Tac + MMF + Pd	No
3	23	Compatible	No	Unrelated	‐	‐	–	–	None	None	Tac + MMF + Pd	No
4	55	Compatible	No	Spouse	3	1	–	–	Pd	BAX	Tac + MMF + Pd	No
5	30	Compatible	Yes	Mother	0	0	–	–	None	BAX	Pd	Yes
6	64	Compatible	No	Unrelated	0	0	–	–	None	ATG	Tac + MMF + Pd	No
7	34	Compatible	No	Unrelated	0	0	–	–	None	ATG	Tac + MMF + Pd	No
8	40	Compatible	Yes	Brother	0	0	–	–	None	BAX	Tac + MMF + Pd	Yes

ATG, anti‐thymocyte globulin; BAX, basiliximab; CDC, complement dependent cytotoxicity; DSA, donor‐specific antibody; HLA, human leukocyte antigen; IS, immunosuppression; KT, kidney transplantation; MMF, mycophenolate mofetil; Pd, prednisolone; RTX, rituximab; Tac, tacrolimus.

Cases 1, 2, 4, 5 and 8 were treated with antibody induction using basiliximab (20 mg on days 0 and 4), followed by triple immunosuppression consisting of tacrolimus, mycophenolate sodium and methylprednisolone. In case 1, where the flow cytometric crossmatch was positive, the recipient underwent three sessions of plasma exchange. In case 2, involving an ABO‐incompatible transplant, we added rituximab (375 mg/body) and conducted four sessions of plasma exchange.

Patient survival and graft functions

The median follow‐up duration after KT was 55 months (range: 2–143 months). The Kaplan–Meier estimate of median overall survival was 139 months (Fig. 1). During follow‐up, one patient (case 3) died of sepsis secondary to a urinary tract infection. The patient, case 3, had been advised to perform clean intermittent catheterization due to recurrent voiding dysfunction. Upon admission, the patient presented with fever and altered mental status, and blood cultures confirmed Klebsiella pneumoniae bacteraemia. Despite receiving intensive care for 1 month, the patient eventually died.

Figure 1.

Kaplan–Meier curve of patient survival after kidney transplantation in patients with a prior history of haemopoietic stem cell transplantation. Patient survival was estimated from the time of kidney transplantation to death or last follow‐up. Tick marks indicate censored observation.

Overall, graft functions remained stable, and none of the patients progressed to ESKD. The median serum creatinine level at the last hospital visit was 0.81 mg/dL. Acute rejection episodes occurred only in case 2, who experienced acute T‐cell‐mediated rejection of Banff grade IA at 28 months post‐transplant. This rejection was treated with steroid pulse therapy using deflazacort (36 mg #2). Although the serum creatinine level rose from 1.25 to 1.79 mg/dL during the rejection episode, graft function was promptly restored following the anti‐rejection therapy. No rejection episodes were reported in the other seven patients.

KT complications and haematologic outcome

Infectious complications were observed in two patients (cases 3 and 5), both of whom experienced urinary tract infection and transient BK viraemia. Additionally, case 5 developed cytomegalovirus (CMV) viraemia and herpes zoster 6 months after transplantation, both of which were treated with antiviral therapy. Tacrolimus toxicity was identified in case 3 at the 6‐month mark, confirmed by graft biopsy. No recurrence of haematologic disease was noted during the subsequent period of immunosuppression following KT.

Discussion

This study evaluated the outcomes of KT in patients with ESKD who had previously undergone HSCT for haematologic disorders. While HSCT is an established curative treatment for various haematologic malignancies and BM failure syndromes, long‐term survivors are increasingly affected by non‐haematologic complications, including kidney damage. ¹⁶ In this study, we found that KT after HSCT is both safe and effective; therefore, it can be considered for patients with ESKD following HSCT. The first consideration of this study is the cause of ESKD following HSCT and the timing of renal dysfunction after the procedure. CKD and its progression to ESKD are particularly important sequelae, with reported incidence rates ranging from 13% to 66%, depending on patient populations and diagnostic criteria. ⁴ , ¹⁷ The pathogenesis of CKD development after HSCT is multifactorial, involving nephrotoxic drug exposure, radiation injury, thrombotic microangiopathy and renal injury related to GVHD. ¹⁷ , ¹⁸ At the time of HSCT, none of the patients had been diagnosed with a definitive kidney disease; therefore, their ESKD was most likely attributable to HSCT in this study. While a direct causal link between HSCT and ESKD could not be definitively established for most patients, it was evident in a subset – such as case 1 – where the kidney biopsy revealed thrombotic microangiopathy. This condition is a well‐recognised post‐HSCT complication with multifactorial aetiologies, including calcineurin inhibitor toxicity, conditioning‐related injury, infections and GVHD‐associated endothelial damage. ³ , ⁴

In this study, the interval between HSCT and ESKD varied widely, with a median of approximately 9 years. This highlights the need for long‐term renal surveillance in HSCT survivors, especially among those exposed to calcineurin inhibitors, radiation, hypertension, diabetes or thrombotic microangiopathy. Early identification of renal injury may create opportunities for intervention before progression to ESKD. ¹⁹ , ²⁰ , ²¹ Multidisciplinary collaboration among nephrologists, haematologists and transplant surgeons is essential when evaluating HSCT survivors for KT. This collaboration ensures a comprehensive assessment of remission status, immune recovery and infection history.

The next concern is the safety and effectiveness of KT in patients who previously underwent HSCT due to a haematologic disorder. Once ESKD develops, renal replacement therapy becomes inevitable. While dialysis offers limited survival and quality of life, KT represents the optimal treatment option. ²² Nevertheless, KT in post‐HSCT patients has been underutilised due to concerns regarding infection, haematologic relapse and challenges in managing IS. Our findings, however, demonstrate that KT after HSCT is both feasible and effective, with favourable patient and graft survival. During a median follow‐up of over 4 years, we observed only one episode of acute T‐cell‐mediated rejection and one graft loss–unrelated death. Importantly, none of the patients with haematologic malignancies experienced disease relapse after KT, alleviating concerns that post‐transplant immunosuppression might trigger recurrence. These results are consistent with previous small series that reported low rejection rates and acceptable survival in this setting. ⁵ , ⁶ , ²³ Careful patient selection, along with ensuring stable haematologic remission and controlled GVHD, is critical for achieving favourable outcomes.

A particularly novel and clinically meaningful observation was the achievement of stable graft function without the need for maintenance IS in three patients who underwent HSCT and KT from the same donor. This phenomenon suggests that haemopoietic chimerism may provide durable donor‐specific tolerance, a concept previously demonstrated in experimental models and select clinical protocols. ⁵ , ⁶ The potential to safely minimise or discontinue IS is highly relevant because long‐term immunosuppressive therapy poses significant risks, including infection, malignancy, metabolic complications and nephrotoxicity. ²⁴ , ²⁵

The durability of tolerance following drug withdrawal remains uncertain, highlighting the need for close monitoring through protocol biopsies and immunologic assays in these patients. In our study, the three patients who received both HSCT and KT from the same donor exhibited unique donor‐specific tolerance. However, the immunologic context of the other patients requires careful attention, as it fundamentally differs from that of conventional KT. In these instances, the recipient's immune system is not self‐derived; rather, it reflects the HLA background of the HSCT donor. As a result, the immunologic interaction occurs between the HSCT donor's immune system and the kidney from the KT donor, introducing an added layer of alloimmune disparity beyond the typical donor–recipient HLA mismatch. Despite this increased risk of alloimmune conflict, our cohort experienced limited rejection events, and graft outcomes were generally favourable. These results imply that, although the immunologic discordance between the HSCT donor‐derived immune system and the KT graft theoretically complicates alloimmune responses, stable graft outcomes were achieved with standard IS in most patients.

Infections were the most significant complication in our cohort, with one patient dying from sepsis following recurrent urinary tract infections. We also encountered opportunistic viral infections, including CMV, BK virus and varicella‐zoster virus, which are consistent with the increased infectious vulnerability of HSCT recipients. ²⁶ , ²⁷ , ²⁸ The overall infection burden was not significantly higher than what is typically seen in conventional KT populations. This indicates that individualised IS regimens, along with vigilant surveillance and prophylactic strategies, can help reduce this risk. Additionally, the ability to reduce or withdraw IS in cases where donor identity is shared may further lower infection‐related morbidity and mortality.

However, several limitations should be acknowledged. The retrospective, single‐centre design and small sample size restrict the generalisability of the findings. Furthermore, patient heterogeneity regarding haematologic disease, type of HSCT and causes of ESKD introduces potential confounding factors. Moreover, detailed immunologic analyses, such as donor‐specific antibody monitoring or chimerism quantification, were not consistently available, which could have provided deeper insights into the mechanisms of tolerance induction. Lastly, this cohort likely represents a select subgroup of HSCT survivors eligible for KT, as many other recipients may not survive long enough or may become unsuitable for KT because of relapse, GVHD or other comorbidities. Because the study included only KT recipients, we could not assess the overall incidence of post‐HSCT CKD or the proportion of patients who ultimately developed ESKD severe enough to warrant KT.

Conclusion

In conclusion, KT after HSCT can be performed safely, resulting in favourable patient and graft survival rates, a manageable infection profile and no increased risk of haematologic relapse. Notably, the potential for donor‐specific tolerance in patients receiving both HSCT and KT from the same donor is significant, as it paves the way to reducing the need for immunosuppression. These findings advocate for the inclusion of KT into standard renal replacement therapy options for appropriately selected HSCT survivors who develop ESKD. Future multicentre collaborations and registry‐based studies are essential to validate these results, establish standardised protocols for immunosuppression and infection prevention and elucidate the mechanisms of tolerance in this unique population.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.