Impact of Hematopoietic Stem Cell Transplantation-Associated Thrombotic Microangiopathy on Renal Failures Requiring Dialysis in Adult Patients ≥ 40 Years

Abstract

Introduction: Allogeneic hematopoietic cell transplantation (allo-HSCT)-associated thrombotic microangiopathy (TA-TMA) and pre-transplant renal dysfunction are recognized risk factors for mortality after allo-HSCT. Utilizing the data from the Center for International Blood and Marrow Transplant Research (CIBMTR), we investigated the association between onset of TA-TMA and pre-HSCT renal dysfunction on renal failure requiring dialysis (RFD). Methods: We evaluated TA-TMA as a time-dependent covariate in a multivariate Cox regression model for RFD in Allo-HSCT recipients aged ≥ 40 years between 2008 and 2016. Pre-HSCT patients were divided into two groups, estimated GFR (eGFR) < 60 mL/min/1.73 m² group and eGFR ≥ 60 mL/min/1.73 m². Cumulative hazards of RFD in patients with and without onset of TA-TMA were estimated. Results: TA-TMA was significantly associated with increased risk (6.6-fold compared to No TA-TMA) for RFD, the highest of all the significant risk factors. The estimated cumulative hazard for patients with TA-TMA in the two pre-HSCT renal function groups was significantly elevated when compared to similar patients with no TA-TMA (80% vs. 12% for eGFR < 60 mL/min and 50% vs. 5% for eGFR ≥ 60 mL/min group, respectively) at 12 months post-HSCT. Conclusions: Our results demonstrate that the adjusted HR of renal failure requiring dialysis and cumulative hazard was much higher in patients with onset of TA-TMA, especially among patients with pre-existing renal dysfunction, underscoring the importance of early recognition and risk-adapted management.

1. Introduction

Allogeneic hematopoietic cell transplantation (allo-HSCT) is an established treatment modality with curative potential for malignant hematological diseases. The number of allo-HSCTs performed every year is increasing, and they have been shown to improve outcomes and quality of life [1,2]. Despite the favorable results post allo-HSCT, it has been associated with sequelae such as thrombotic microangiopathy (TMA) [3].

TMA occurs more commonly after allo-HSCT than autologous HSCT, and the incidence rates have been increasing over time and can occur in up to 39% of the patients [4]. Transplant-associated TMA (TA-TMA) could be acute or subacute but typically occurs 6 to 12 months post allo-HSCT [5]. Many factors have been implicated in TA-TMA, but the most significant variations are seen between calcineurin inhibitors, viral infections, and pre-transplant preparative regimens [6]. The primary pathology is a widespread endothelial injury due to endothelial dysfunction, complement dysregulation, infections, and alloreactivity. Manifesting as hemolytic anemia, thrombocytopenia, proteinuria, hemorrhage, and ischemia of the affected organs, it can lead to life-threatening illness [7]. TA-TMA has also been postulated to be incited by graft versus host disease (GVHD), where complex molecular interactions between cytokines and endothelial cells occur in addition to the activation of macrophages by infectious organisms, causing increased interleukin (IL)-12 and increased permeability of blood vessels with further endothelial damage [6].

While TA-TMA can affect any organ, the kidneys are more often associated with dysfunction. Acute kidney injury (AKI) is a well-recognized complication after allo-HSCT, and this also causes an elevated risk of renal failure requiring dialysis (RFD) and mortality [8]. Notably, early AKI, occurring within 100 days of allo-HSCT, is associated with a poor prognosis [9]. RFD has been linked to poor outcomes and mortality primarily due to cardiovascular events. It has been shown in previous studies that patients with an estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m² tend to have poor renal outcomes [10].

RFD also causes worsening of pre-existing chronic kidney disease (CKD) and increases the risk for end-stage renal disease (ESRD) [11]. Epperla et al. showed that TA-TMA increased renal replacement therapy requirements [12]. In another study, Farhadfar et al. showed that patients with decreased pre-transplant kidney function post allo-HSCT had higher risks of RFD [13]. However, according to a study by Sakaguchi et al., AKI was determined to be a negative indicator for future outcomes after allo-HSCT, but CKD did not show the same association [9]. Since there is a bidirectional relationship between RFD and CKD, we postulated that patients with CKD have an increased rate of worsening renal failure and requirements for renal replacement therapy. We assessed the population ≥ 40 years since the incidence of CKD is lower in people < 40 years old [14].

Taken together, these observations highlight the complex interplay between TA-TMA, AKI, CKD and post-transplant renal outcomes. However, the combined effect of TA-TMA and baseline renal dysfunction on the risk of dialysis requiring renal failure (RFD) has not been systematically studied. To address this gap, we performed a secondary analysis of the Center for International Blood and Marrow Transplant Research (CIBMTR) dataset that was previously examined by Farhadfar et al. [13]. We focused on allo-HST recipients aged ≥ 40 years, among whom CKD is more prevalent, to evaluate the incidence of RFD in patients with TA-TMA, characterize outcomes by renal function group, and determine factors associated with RFD one year post-transplant.

2. Methods

2.1. Data Source

The Center for International Blood and Marrow Transplant Registry (CIBMTR) is a research collaboration of over 360 international centers contributing detailed data on allogeneic and autologous HSCT and cellular therapies. This dataset was collected by CIBMTR, which is supported primarily by the Public Health Service U24CA076518 from the National Cancer Institute; the National Heart, Lung, and Blood Institute; the National Institute of Allergy and Infectious Diseases; 75R60222C00011 from the Health Resources and Services Administration; N00014-23-1-2057 and N00014-24-1-2507 from the Office of Naval Research; NMDP; and the Medical College of Wisconsin. This is an observational study with already-available data and, as such, ethics approval was not required. Our dataset included patients ≥40 years old with hematologic malignancies who underwent a first allo-HSCT between 2008 and 2016 and had eGFR values reported prior to allo-HSCT. Data used in our study were publicly available from the CIBMTR [15]. In this secondary analysis, patients were stratified by pre-allo-HSCT eGFR (<60 mL/min and ≥60 mL/min).

2.2. Inclusion and Exclusion Criteria and Patient Population Characteristics

Inclusion Criteria: Patients ≥ 40 years with hematologic malignancies who underwent first allo-HSCT between January 2008 and December 2016 and had serum creatinine and eGFR values reported prior to HCT were included in the analysis, to evaluate the association between onset of TA-TMA and pre-HSCT renal dysfunction on RFD.

Exclusion Criteria: Patients less than 40 yrs of age were excluded from this cohort due to low incidence of renal dysfunction in patients < 40 years. Patients receiving ex vivo T-cell depletion or CD34+ selected grafts were excluded.

Patient Population characteristics: Pre-allo-HSCT characteristics of all patients were summarized by patient strata, defined by pre-allo-HSCT eGFR values and post-allo-HSCT onset of TA-TMA. Categorical variables were summarized by proportions and continuous variables were summarized by medians and interquartile ranges.

2.3. Statistical Methods

All potential risk factors were evaluated for association with renal failure requiring dialysis in a multivariable Cox regression model with stepwise forward and backward selection approach with a threshold of 0.05 for both entry and stay in the model. In addition, we added confounders and risk factors previously described in other studies [12,13]. TA-TMA was used as a time-dependent covariate, while pre-transplant renal health and all other potential risk factors, viz. sex, age, race, pre-HSCT renal health, donor type, prior autologous transplant, Karnofsky performance score (KPS) [16], donor–recipient cytomegalovirus status, conditioning regimens, use of anti-thymocyte globulin (ATG)/alemtuzumab, graft type, veno-occlusive disease (VOD), idiopathic pneumonia syndrome (IPS), acute GVHD and acute GVHD prophylaxis were included as time-independent (fixed) covariates in the model. Acute GVHD Grade 2–4, IPS and VOD cases were only included as risk factors if they occurred in patients before renal failure requiring dialysis. Karnofsky Performance Status (KPS) scores range from 0 to 100, with higher scores reflecting greater ability to perform daily activities. For analysis, KPS was dichotomized as 90–100 versus <90, distinguishing patients with preserved versus impaired functional capacity. Adjusted hazard ratios (aHR) were evaluated to identify potential risk factors for RFD. Cumulative hazards of RFD in patients with and without onset of TA-TMA were estimated for the two pre-HSCT renal function groups with varying baseline hazard functions. Statistical analyses were conducted using SAS 9.4 (SAS Institute Inc. 2023, Cary, NC, United States) [17].

3. Results

3.1. Baseline Characteristics

A total of 13,176 patients with non-missing pre-allo-HSCT eGFR values and post-allo-HSCT TA-TMA information were included in the analysis. There were 1151 (8.7%) patients in the pre-HSCT eGFR < 60 mL/min group and 12,025 (91.7%) patients in the pre-HSCT eGFR ≥ 60 mL/min group. Pre-HSCT characteristics for the two eGFR groups are summarized in

Table 1. Demographics and HCT characteristics by pre-HCT eGFR group.

	eGFR < 60 (N = 1151)	eGFR ≥ 60 (N = 12,025)	p-Value
Sex: Female	44.5%	40.3%	0.006
Race: Caucasian	90.2%	87.4%	0.010
Age at HCT (median, IQR)	64 (58, 68)	58 (51, 65)	0.001
HCT-CI ≥ 3	56.0	45.1	<0.001
Age ≥ 60 years	66.4%	42.2%	<0.001
Donor age < 30/30–49/≥50	31.4%/28.5%/27.9%	28.7%/31.6%/25.1%	0.008
Donor type (Related/Unrel. well-matched/Other unrel./Cord blood)	35.7%/43.8%/9.0%/10.3%	36.6%/41.7%/9.5%/11.4%	0.577
Chronic GvHD (Yes)	42.2%	45.6%	0.082
ATG and/or alemtuzumab	25.9%	26.7%	<0.001
KPS < 90	46.1%	40.7%	<0.001
Previous autologous transplant	9.7%	5.6%	<0.001
Year of transplant (2012–2016)	66.5%	61.2%	<0.001

Abbreviations: HCT, hematopoietic cell transplantation; HCT-CI, hematopoietic cell transplantation–comorbidity index; IQR, interquartile range; graft versus host disease; ATG, anti-thymocyte globulin; KPS, Karnofsky performance score.

. There were significant differences in the baseline characteristics in the two eGFR patient groups. These variables were included as fixed (time-invariant) covariates in the multivariable Cox regression analysis for RFD.

Of the 13,176 patients who had allo-HSCT for various underlying diseases, 382 (2.9%) patients developed TA-TMA post-HSCT. Incidence was similar in the two eGFR groups (3.4% for <60 mL/min and 2.8% in ≥60 mL/min group, p = 0.300). Pre-HSCT characteristics for these patients by post-HSCT TA-TMA onset are summarized in

Table 2. Demographics and HSCT characteristics of entire cohort, TA-TMA versus no TA-TMA.

	No TA-TMA (N = 12,794)	TA-TMA (N = 382)	p-Value
Sex: Female	40.4%	49.2%	<0.001
Race: Caucasian	87.7%	85.9%	0.055
Age at HCT (median, IQR)	59 (52, 65)	57 (49, 64)	0.001
HCT-CI ≥ 3	46.0%	49.2%	0.118
Renal function group: Pre-HCT eGFR < 60	8.7%	10.2%	0.300
Age ≥ 60 years	44.5%	40.6%	0.001
Donor age < 30/30–49/≥50	28.9%/31.2%/25.5%	29.1%/34.4%/39.1%	0.653
Donor type (Related/Unrel. well-matched/Other unrel./Cord blood)	36.8%/41.8%/9.6%/11.3%	30.4%/44.2%/13.0%/11.0%	0.044
Chronic GvHD	45.1%	53.7%	0.003
ATG and/or alemtuzumab	26.8%	19.9%	0.020
KPS < 90	41.2%	40.6%	0.728
Previous autologous transplant	5.9%	10.0%	0.003
Year of transplant (2012–2016)	61.8%	56.5%	0.037

Abbreviations: HCT, hematopoietic cell transplantation; HCT-CI, hematopoietic cell transplantation–comorbidity index; IQR, interquartile range; graft versus host disease; ATG, anti-thymocyte globulin; KPS, Karnofsky performance score.

. The incidence of TA-TMA was higher in female patients, patients with chronic GVHD and prior autologous transplant. Also, the incidence of TA-TMA was significantly higher in years 2012–2016 compared to 2008–2011, most likely due to improvements in TA-TMA diagnosis.

Pre-HSCT characteristics of patients who developed TA-TMA, in the two eGFR groups, are summarized in

Table 3. Demographics and HSCT characteristics of TA-TMA patients.

	eGFR < 60 (N = 39)	eGFR ≥ 60 (N = 343)	p-Value
Sex: Female	53.8%	48.7%	0.542
Race: Caucasian	89.7%	85.4%	0.252
Age at HCT (median, IQR)	59 (52, 65)	57 (49, 64)	0.001
HCT-CI ≥ 3	71.8%	46.6%	<0.001
Age ≥ 60	61.5%	38.2%	0.018
Donor age < 30/30–49/≥50	33.3%/30.8%/30.8%	28.6%/34.7%/21.0%	0.391
Donor type (Related/Unrel. well-matched/Other unrel./Cord blood)	41.1%/41.0%/12.8%/5.1%	29.2%/44.6%/13.2%/11.7%	<0.001
Chronic GvHD	41.0%	55.1%	0.095
ATG and/or alemtuzumab	28.2%	21.8%	0.090
KPS < 90	48.7%	39.7%	0.546
Previous autologous transplant	12.8%	9.6%	0.494
Year of transplant 2012–2016	69.2%	55.1%	0.092

Abbreviations: HCT, hematopoietic cell transplantation; HCT-CI, hematopoietic cell transplantation–comorbidity index; IQR, interquartile range; graft versus host disease; ATG, anti-thymocyte globulin; KPS, Karnofsky performance score.

. There were no significant differences in sex, race, donor age, chronic GVHD, KPS, ATG/CAMPATH and year of transplant, between the two eGFR groups, in developing TA-TMA. Patients with eGFR < 60 mL/min were older and had significantly more morbidities (≥3) than patients with eGFR ≥ 60 mL/min.

3.2. Renal Failure Requiring Dialysis

TA-TMA was significantly associated with increased risk (aHR 6.60 95% CI 4.71–9.26, p < 0.001) compared to no TA-TMA for RFD. Renal dysfunction (eGFR < 60 mL/min) prior to allo-HSCT was a significant risk factor (aHR 2.40, 95% CI 1.96–2.93, p < 0.001) for RFD when compared to eGFR ≥ 60 mL/min group (

Table 4. Incidence of TA-TMA, renal failure requiring dialysis and adjusted risk factors for renal failure requiring dialysis post-hematopoietic cell transplantation.

Characteristic	Pre-HCT eGFR (mL/min)
	<60 (n = 1151)	≥60 (n = 12,025)	p-Value
Incidence of TA-TMA	3.4%	2.8%	0.300
Incidence of Renal Failure Requiring Dialysis	12.2%	5.8%	<0.001
With No TA-TMA	11.8% (131/1112)	5.4% (627/11,682)	<0.001
With TA-TMA	23.1% (9/39)	21.0% (72/343)
a Significant Risk Factors for Renal Failure Requiring Dialysis Post-Hematopoietic Cell Transplantation
Risk Factors		HR (95% CI)	p-Value
TA-TMA (vs. No TA-TMA)		6.60 (4.71, 9.26)	<0.001
eGFR < 60 m/min (vs. eGFR ≥ 60 mL/min)		2.40 (1.96, 2.93)	<0.001
HCT-CI = 1 (vs. 0)		1.22 (0.93, 1.59)	0.146
HCT-CI = 2 (vs. 0)		1.34 (1.03, 1.75)	0.031
HCT-CI = 3 (vs. 0)		1.55 (1.26, 1.89)	<0.001
Acute GvHD Grade 2 (vs. No acute GvHD or Grade 1)		0.53 (0.43, 0.65)	<0.001
Acute GvHD Grade 3–4 (vs. No acute GvHD or Grade 1)		1.18 (0.98, 1.42)	0.083
KPS < 90 (vs. 90–100)		1.20 (1.03, 1.40)	0.019
Male (vs. Female)		1.23 (1.05, 1.43)	0.010
Cord blood (vs. BM)		1.54 (1.12, 2.13)	0.008
PB (vs. BM)		0.79 (0.62, 1.01)	0.060
Veno-occlusive disease: Yes (vs. No)		3.97 (2.85, 5.52)	<0.001
Idiopathic pneumonia syndrome: Yes (vs. No)		1.40 (1.06, 1.84)	0.018
ATG/alemtuzumab: Yes (vs. No)		1.51 (1.28, 1.79)	<0.001
Conditioning regimen
BuCy (vs. CyTBI)		0.66 (0.48, 0.91)	0.011
BuFlu (vs. CyTBI)		0.48 (0.36, 0.64)	<0.001
FluCyTBI (vs. CyTBI)		0.84 (0.56, 1.27)	0.414
FluMel (vs. CyTBI)		0.80 (0.57, 1.12)	0.190
Others (vs. CyTBI)		0.84 (0.61, 1.17)	0.307
Conditioning regimen intensity
NMA (vs. Myeloablative)		0.60 (0.45, 0.81)	<0.001
RIC (vs. Myeloablative)		0.89 (0.71, 1.10)	0.284
Underlying disease
ALL (vs. AML)		1.08 (0.80, 1.46)	0.605
Lymphoma (vs. AML)		1.46 (1.13, 1.88)	0.004
MDS (vs. AML)		1.25 (1.02, 1.53)	0.029
MPN/CML (vs. AML)		1.17 (0.88, 1.55)	0.275
Other malignant hematologic disease (vs. AML)		1.77 (1.34, 2.33)	<0.001
Year of Transplant 2012–2016 (vs. 2008–2011)		0.68 (0.58, 0.80)	<0.001

^a Cox multivariate regression, stepwise selection with a threshold of 0.05 for both entry and stay in the model. Cy, cyclophosphamide; TBI, total-body irradiation; Bu, busulfan; NMA, Non-myeloablative; Flu, fludarabine.

). The risk of RFD was high with increasing hematopoietic cell transplantation–comorbidity index (HCT-CI) scores, aHR 1.34, 95% CI 1.03–1.75, p = 0.031; aHR 1.55, 95% CI 1.26–1.89, p < 0.001 for 2 and ≥3 co-morbidities, respectively. Transplantation graft source, cord blood (CB) vs. bone marrow (BM) was significantly associated with RFD (aHR 1.54, 95% CI 1.12–2.13, p  =  0.008). While the incidence of VOD, a rare complication post-HSCT, was similar across the two groups (2.1%), the onset of VOD was a significant risk factor for RFD (aHR 3.97, 95% CI 2.85–5.52, p < 0.001). IPS, with an incidence of 7.2% for eGFR < 60 mL/min and 5.9% for eGFR ≥ 60 mL/min, was also a significant risk factor for RFD (aHR 1.40, 95% CI 1.06–1.84, p = 0.018). Patients with lymphoma (aHR 1.46 95% CI 1.13–1.88, p = 0.004), myelodysplastic syndrome (MDS) (aHR 1.25 95% CI 1.02–1.53, p = 0.029) and other malignant disease groups (aHR 1.77, 95% CI 1.34–2.33, p < 0.001) were at elevated risks for RFD, when compared to acute myeloid leukemia (AML). Patients with KPS < 90 had a significantly elevated risk for RFD (aHR 1.20, 95% CI 1.03, 1.40, p = 0.019). ATG and/or alemtuzumab use increased the risk of RFD.

When compared to the conditioning regimen with cyclophosphamide and total-body irradiation (TBI), the risk of RFD was lower among patients who received busulfan cyclophosphamide (Bu Cy) (aHR 0.66 95% CI 0.48–0.91, p = 0.011) and Bu-fludarabine (Flu) (aHR 0.48, 95%CI 0.36–0.64, p < 0.001). When comparing conditioning regimen intensity, non-myeloablative (NMA) (aHR 0.60, 95% CI 0.45–0.81, p < 0.001) had less risk of RFD compared to myeloablative regimens. There was lower risk for RFD with allo-HSCT after 2012 (aHR 0.68 95% CI, 0.58–0.80, p < 0.001) when compared to transplants prior to 2012.

The estimated cumulative hazards for other significant risk factors at 6 months and 12 months post-HSCT are presented in

Table 5. Marginal effect of significant risk factors on cumulative risk of renal failure requiring dialysis at 6 months post-hematopoietic stem cell transplantation.

Cumulative Hazard (95% Confidence Interval)
Patient Profile	eGFR < 60 mL/min		eGFR ≥ 60 mL/min
	No TA-TMA	TA-TMA	No TA-TMA	TA-TMA
Reference	0.09	0.61	0.04	0.38
Acute GvHD
Grade 2	0.05	0.32	0.02	0.20
Grade 3–4	0.11	0.71	0.05	0.45
Conditioning regimen
BuCy	0.06	0.40	0.03	0.25
BuFlu	0.04	0.29	0.02	0.18
FluMel	0.07	0.48	0.03	0.30
FluCyTBI	0.08	0.51	0.03	0.32
Other	0.08	0.51	0.03	0.32
Conditioning Intensity
RIC	0.08	0.54	0.03	0.34
NMA	0.06	0.36	0.02	0.23
Underlying disease
ALL	0.10	0.66	0.04	0.41
MDS	0.11	0.76	0.05	0.48
MPN/CML	0.11	0.71	0.04	0.45
Lymphoma	0.13	0.88	0.06	0.56
Other MDH	0.16	1.07	0.07	0.67
Graft Source
PB	0.07	0.48	0.03	0.30
CB	0.14	0.94	0.06	0.59
VOD	0.36	2.41	0.15	1.51
IPS	0.13	0.85	0.05	0.53
# Comorbidities = 1	0.11	0.74	0.05	0.47
# Comorbidities = 2	0.12	0.81	0.05	0.35
# Comorbidities ≥ 3	0.14	0.94	0.06	0.59
Female	0.11	0.74	0.05	0.38
No ATG or alemtuzumab	0.14	0.92	0.06	0.31
KPS < 90	0.11	0.73	0.05	0.56
Year 2008–2011	0.07	0.49	0.03	0.38

KPS, Karnofsky performance score; IPS, idiopathic pneumonia syndrome; VOD, veno-occlusive disease; MDH, malignant hematologic disease; ATG, anti-thymocyte globulin.

and

Table 6. Marginal effect of significant risk factors on cumulative risk of renal failure requiring dialysis at 12 months post-hematopoietic stem cell transplantation.

Cumulative Hazard
Patient Profile	eGFR < 60 mL/min		eGFR ≥ 60 mL/min
	No TA-TMA	TA-TMA	No TA-TMA	TA-TMA
Reference	0.12	0.80	0.05	0.50
Acute GvHD
Grade 2	0.06	0.42	0.03	0.27
Grade 3–4	0.14	0.94	0.06	0.59
Conditioning regimen
BuCy	0.08	0.53	0.03	0.33
BuFlu	0.06	0.38	0.02	0.24
FluMel	0.10	0.64	0.04	0.40
FluCyTBI	0.10	0.67	0.04	0.42
Other	0.10	0.68	0.04	0.43
Conditioning Intensity
RIC	0.11	0.71	0.04	0.45
NMA	0.07	0.48	0.03	0.30
Underlying disease
ALL	0.13	0.87	0.05	0.55
MDS	0.15	1.00	0.06	0.63
MPN/CML	0.14	0.94	0.06	0.59
Lymphoma	0.18	1.17	0.07	0.74
Other MDH	0.21	1.41	0.09	0.89
Graft Source
PB	0.10	0.63	0.04	0.40
CB	0.19	1.24	0.08	0.78
VOD	0.48	3.18	0.20	2.00
IPS	0.17	1.12	0.07	0.70
# Comorbidities = 1	0.15	0.98	0.06	0.61
# Comorbidities = 2	0.16	1.07	0.07	0.46
# Comorbidities ≥ 3	0.19	1.24	0.08	0.78
Female	0.15	0.98	0.06	0.50
No ATG or alemtuzumab	0.18	1.21	0.08	0.41
KPS < 90	0.15	0.96	0.06	0.74
Year 2008–2011	0.10	0.65	0.04	0.50

KPS, Karnofsky performance score; IPS, idiopathic pneumonia syndrome; VOD, veno-occlusive disease; MDH, malignant hematologic disease; ATG, anti-thymocyte globulin.

, respectively. The estimated cumulative hazards for patients with TA-TMA in the two pre-HSCT renal function groups were significantly elevated when compared to similar patients with no TA-TMA (61% vs. 9% for eGFR < 60 mL/min and 38% vs. 4% for eGFR ≥ 60 mL/min group, respectively) at 6 months post-HSCT. The estimated cumulative hazards for patients with TA-TMA in the two pre-HSCT renal function groups were significantly elevated when compared to similar patients with no TA-TMA (80% vs. 12% for eGFR < 60 mL/min and 50% vs. 5% for eGFR ≥ 60 mL/min group, respectively) at 12 months post-HSCT.

4. Discussion

This large, registry-based retrospective cohort study evaluated the relationship between pre-transplant renal function and TA-TMA with RFD. We identified 13,176 cases of TA-TMA pre-HSCT from the CIBMTR database between 2008 and 2016 that were stratified according to eGFR < 60 mL/min/1.73 m² and eGFR ≥ 60 mL/min/1.73 m². Of these cases, 382 (2.9%) patients developed TA-TMA post-HSCT. Notably, the incidence of TA-TMA was not significantly different between the renal function groups. The presence of renal dysfunction (eGFR < 60 mL/min) prior to allo-HSCT was a significant risk factor for RFD. Our analysis further demonstrated that TA-TMA was a significant risk factor for RFD. In fact, patients with TA-TMA experienced a 6.6-fold increased risk of RFD compared to patients without TA-TMA, which was higher than other significant risk factors.

This study adds to the previous work elucidating the impact of allo-HSCT and TA-TMA on kidney function. Previous studies have shown that patients with pre-transplant decreased kidney function had higher risks of RFD in allo-HSCT patients [13]. Other studies have also shown that patients who developed TA-TMA were at an increased risk for renal replacement therapy [12]. Our results further support these findings, showing that pre-transplant renal function did not significantly affect the incidence of TA-TMA but significantly increased the risk of RFD. We also further identified risk factors for RFD among patients with TA-TMA, including HCT-CI, transplantation graft source, onset of VOD, IPS, and ATG and/or alemtuzumab use. The underlying malignancy was also a significant risk factor for developing RFD as patients with lymphoma, MDS and other malignant disease groups were at elevated risks for RFD, when compared to AML.

TMA describes a pathological process in which widespread endothelial injury causes platelet aggregation and thrombus formation in small blood vessels [7]. Given this pathological process, it stands to reason that the etiology of renal insult following TA-TMA is ischemic in nature. Epperla et al. found that TA-TMA was significantly associated with RFD. They also identified several risk factors for the incidence of TA-TMA. Factors that were associated with decreased risk of TA-TMA included male sex (relative to females), use of ATG or alemtuzumab (compared to no ATG/alemtuzumab) and reduced intensity/non-myeloablative conditioning (vs. busulfan-based myeloablative conditioning). Also, among black people, a diagnosis of acute lymphoblastic leukemia (ALL) or aplastic anemia (relative to AML), poor renal function (GFR < 60 relative to GFR ≥ 60), CNI with sirolimus as GVHD prophylaxis (relative to CNI + others), and prior auto-HCT (relative to no prior auto-HCT) was associated with increased risk of TA-TMA. Compared to HLA-identical sibling donors, all other donor types had an increased risk for TA-TMA development. After adjusting for all these risk factors, patients who developed acute GVHD post-transplant had a significantly higher risk of TA-TMA relative to those with Grade I or no acute GVHD [12]. In our study, we found similar results with incidence of TA-TMA also higher in female patients, patients with chronic GVHD and prior autologous transplant.

To date, our study is the largest in evaluating the relationship between pre-HSCT renal dysfunction and the effects of TA-TMA on RFD. Previous large retrospective studies utilizing the CIBMTR patient population to research TA-TMA discussed risk factors impacting the incidence of the condition and outcomes; however, our study is novel in its stratification by pre-transplant renal function. Furthermore, this is the first study to report the incidence and risk factors for TA-TMA in patients with eGFR < 60 mL/min.

Limitations: First, the CIBMTR database utilizes provider-reported TA-TMA, which contrasts with laboratory-detected TA-TMA reported in other studies. Due to this, it is estimated that as many as 64% of laboratory-detected TA-TMA cases may have not been reported/recorded in the registry and not included in the study [12]. It stands to reason that these missed cases would skew towards a less severe disease, contributing to a selection bias and overestimation of TA-TMA’s impact in our study. Also, another drawback common to all retrospective database analyses is that the accuracy of TA-TMA diagnosis depends on its reporting by the treating providers. Second, the use of calcineurin inhibitors for GVHD prophylaxis may have contributed to worsening of renal function. Third, it was not possible to determine the duration of acute kidney injury after allo-HSCT or the severity of renal failure. Fourth, our study is observational and, hence, inherent imbalances are unavoidable. In particular, the number of patients with eGFR < 60 mL/min was substantially lower than those with preserved renal function, reflecting the underlying distribution of our cohort. This smaller subgroup size may limit the precision of risk estimates and should be considered when interpreting the findings. Lastly, there may have been non-uniformity in the estimation of eGFR between institutions. It is possible that some institutions used the Modification of Diet in Renal Disease (MDRD) and others could have used the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI). But, nonetheless, our study has many strengths; the analysis was based on a large database, and we adjusted for various confounders.

5. Conclusions

Our study highlights that TA-TMA needs to be diagnosed early and addressed to manage the risk of RFD. TA-TMA significantly increases the risks of RFD in both patients with normal and decreased pre-transplant kidney function; however, there were no significant differences in TA-TMA incidence between these groups. This study identifies important risk factors for RFD in post-allo-HSCT patients. Future avenues of research include updating and evaluating more recent data from the CIBMTR database, as our study only utilized data up to 2016.

Future research should work to establish uniform consensus diagnostic criteria, preferably including both laboratory biomarkers and clinical diagnosis, to increase the generalizability of studies. Our results offer a benchmark for future studies in allo-HSCT recipients with baseline renal dysfunction and offer a means to assess risk for RFD in TA-TMA patients. Researchers should use a shared frailty survival model to account for unobserved heterogeneity when estimating the impact of hematopoietic stem cell transplantation-associated thrombotic microangiopathy (TA-TMA) on renal failure requiring dialysis among patients ≥ 40 years, characterized by pre-transplant kidney health. A frailty term (e.g., gamma or log-normal) will capture latent risk at the patient or center level that is not explained by measured covariates (e.g., VOD, GVHD grade, graft source, comorbidity burden).