Abstract
The negative impact of Iron overload (IO) on outcomes of allogeneic hematopoietic cell transplantation (HCT) is well-recognized, but its impact on umbilical cord blood (UCB) transplant outcome is unknown. We retrospectively analyzed outcomes of 150 patients who received UCB-HCT at our institution, stratified by pre-HCT serum ferritin level (SF) of 2000 ng/ml. Two-year overall survival rate among patients with SF >2000 and ≤2000 ng/ml was 26.1% (95%CI: 10.6%-44.7%) and 52.1% (95%CI: 40.1%-62.8%), respectively; HR=2.26 (95%CI: 1.28-4.00, P=0.005). Two-year non-relapse mortality rate was higher among patients with SF >2000 ng/ml (56.5%, 95%CI: 33.3%-74.4%) compared to SF ≤2000 ng/ml (30.1%, 95%CI: 20.0%-40.9%); HR=2.18, (95%CI: 1.10-4.31, P=0.025). Neutrophil engraftment at 42 days was 78.3% (95%CI: 53.5%-90.8%) in patients with SF >2000 ng/ml, versus 91.8% (95%CI: 82.1%-96.4%) in patients with SF ≤2000 ng/ml; HR=0.58 (95%CI: 0.35-0.96, P=0.034). A significant difference in platelet engraftment at 3 months was also observed: 52.2% (95%CI: 29.4%-70.8%) for SF>2000 ng/ml versus 80.8% (95%CI: 69.5%-88.3%) for SF ≤2000 ng/ml; HR=0.48 (95%CI: 0.23-0.98, P=0.044). In conclusion, IO defined by SF of 2000 ng/ml is a strong adverse prognostic factor for UCB-HCT and should be a considered when UCB is chosen as the graft source for patients without a fully matched donor.
Keywords: Iron overload, Serum ferritin, Hematopoietic Cell Transplantation, Umbilical Cord Blood, Non-relapse mortality, Engraftment
INTRODUCTION
Transfusional iron overload is a common complication in recipients of allogeneic hematopoietic cell transplantation (HCT), and has consistently been shown to be associated with adverse HCT outcomes.1–3 A recent meta-analysis has confirmed the adverse impact of iron overload on overall survival (OS), non-relapse mortality (NRM), and blood stream infections after allogeneic HCT.4 In most of the studies examining the impact of iron overload on HCT outcomes, pre-HCT serum ferritin (SF) levels have been used as a measure of body iron burden. Since there is a good correlation between pre-HCT SF levels and liver iron concentration (LIC), SF level is the most practical marker to study to measure the body iron burden, despite the caveat that SF could be elevated in inflammatory states.5 This is particularly true in transfused patients with hematologic disorders, where the pre-test probability of iron overload is high.5 Studies in which liver iron concentration was measured directly have confirmed the adverse impact of iron overload on allogeneic HCT outcomes and further validated SF as a prognostic parameter of iron overload in HCT recipients.3,6
Umbilical cord blood (UCB) transplantation is associated with higher NRM rates compared to other HCT graft sources, mainly due to the delayed engraftment and immune reconstitution, which consequently leads to a higher rate of infectious complications. Iron is an essential nutrient for microorganisms,7 and the availability of free iron at the time of prolonged neutropenia could contribute to higher rates of bacterial and fungal infections after UCB-HCT with a subsequent increase in the NRM.7 Multiple preclinical studies have recently reported the adverse impact of iron overload on hematopoiesis and bone marrow microenvironment by inducing reactive oxygen species.8–11 Such deleterious effects of iron overload on the bone marrow microenvironment, could adversely impact hematopoietic cell engraftment, particularly in UCB-HCT with limited number of infused hematopoietic stem cells.
Previous studies examining the association between iron overload and allogeneic HCT outcomes have predominantly included patients undergoing HCT from graft sources other than UCB. In this study, we examined the impact of iron overload on HCT outcomes in a large cohort of patients, who underwent UCB-HCT at our institution. We further examined the impact of iron overload on specific HCT complications including delayed or failed engraftment as well as bacterial and fungal infections.
SUBJECTS AND METHODS
Study Population
In this study, we included patients who underwent their first UCB-HCT at City of Hope National Medical Center between March 2006 and September 2015 for a malignant or non-malignant hematological disease. Myeloablative conditioning regimen (MAC) was defined as a single dose of fractionated total body irradiation (fTBI) > 500 cGy or total of fractionated doses > 800 cGy, busulfan (> 9 mg/kg), or melphalan (> 150 mg/m2) as previously described.12 Patients received either a single or double UCB graft, based on their age and body weight. All cord blood units were matched at four, five, or six loci to recipients at the antigen level of HLA -A and -B and at the allele level for HLA-DRB1. Infused cord blood units were required to have contained (combined) >3 × 107 total nucleated cells per kilogram of the recipient’s body weight. Disease risk index (DRI) was defined based on DRI score.13 HCT comorbidity index was defined as Sorror et al, scoring system.14 The study was approved by the Institutional Review Board.
Determination of Iron Parameters
Patients whose SF level was measured within 30 days before HCT were included in the analysis. SF levels were measured by standard methodology in our clinical laboratory, using Access Ferritin kit (Beckman Coulter, Brea, CA) following the manufacturer’s instructions. A SF cutoff of > 2000 ng/ml was used to define iron overload. Bone marrow biopsy, done within 30 days before transplant was scored by the same hematopathologist who assigned an iron score (bone marrow Fe) of 0 to 4 for each sample based on intensity of Prussian blue staining in bone marrow macrophages. The score of 0 indicated no stainable iron, 1 : trace iron, 2: normal iron, 3: increased iron, 4: extensive iron deposition.
Outcome Definitions
OS was defined as the time from UCB-HCT to death from any cause, or censored on the last known date patient was alive. Disease-free survival (DFS) and relapse were defined as per CIBMTR criteria.15 NRM was defined as death from causes not related to disease relapse/progression, and relapse/disease progression was considered as a competing risk and vice versa. Both NRM and relapse were censored at the time of last follow-up if patients were alive and remained relapse/progression free. Time to neutrophil engraftment was defined as the first documented day of achieving an absolute neutrophil count of ≥0.5 × 109/L for 3 consecutive days. Time to platelet engraftment was defined as the first documented day of achieving a platelet count of >20 × 109/L and platelet transfusion independence for 7 days. Acute graft-versus-host disease (GvHD) was graded by the Glucksberg scale,16 and chronic GvHD was classified as limited or extensive according to the Seattle criteria.17 Infections within the first 100 days post UCB-HCT were defined as the first detection of CMV viremia, blood stream infection (BSI), or invasive fungal (mold) infection and death was considered a competing risk.
Statistical Analysis
Baseline patient and HCT-related characteristics were entered in contingency tables and tested using two-sample Wilcoxon and chi-square tests, whenever appropriate. OS and DFS stratified by SF level were examined using Kaplan-Meier curves and log-rank tests in the univariate analysis and Cox proportional hazards model in the multivariable analyses. NRM, relapse rate, acute and chronic GvHD, neutrophil and platelet engraftment, and infections were assessed using cumulative incidence curves and Gray test in the univariate analysis and the proportional subdistribution hazards model for competing risks in the multivariable analyses. The assumptions of proportionality for both Cox regression and Fine and Gray models were checked by corresponding tests and plots of the scaled Schoenfeld residuals or the cumulative sums of residuals whenever appropriate.18–20
SAS 9.4 (SAS Institute, Cary, NC) and R 3.3.3 (R Foundation for Statistical Computing, Vienna, Austria) software were used to perform the statistical analyses. All tests were 2-sided at a significance level of 0.05.
RESULTS
Patient and HCT Characteristics
From 2006 to 2015, 150 patients underwent UCB-HCT at our center. Patient and transplant characteristics are summarized in Table 1. Briefly, median age at the time of HCT for the whole cohort was 34 years (range: 0-70), with 23% of patients being younger than 18 years of age. The most common indication for HCT was acute leukemia (n=104, 69%). DRI was high or very-high in 38% of the patients (n=57), with 20% having HCT comorbidity index of >2. The majority of patients received double UCB unit HCT (80%). Median total nucleated cell count was 4.9 × 107/kg (range: 1.8-26.6) and median CD34+ cell count was 1.8 × 105/kg (range: 0.2-122.4). HLA matching status was 4/6, 5/6, or 6/6 in 93 (62%), 48 (32%), and 9 (6%) of the patients, respectively. MAC was delivered to 58% of the patients and 79% received mycophenolate-based GVHD prophylaxis.
Table 1.
Patient and HCT characteristics
| Serum Ferritin (SF) Level | ||||
|---|---|---|---|---|
| Total (N=150) | <=2000 (N=73) | >2000 (N=23) | P value* | |
| Age at HCT, years | 0.81 | |||
| Median (Range) | 34 (0-70) | 43 (4-70) | 48 (7-67) | |
| Interquartile range | 18, 51 | 31, 57 | 27, 60 | 0.39 |
| <18 | 34 (22.7%) | 3 (4.1%) | 2 (8.7%) | |
| ≥18 | 116 (77.3%) | 70 (95.9%) | 21 (91.3%) | |
| Recipient sex | 0.003 | |||
| Male | 82 (54.7%) | 31 (42.5%) | 18 (78.3%) | |
| Female | 68 (45.3%) | 42 (57.5%) | 5 (21.7%) | |
| Year of HCT | 0.90 | |||
| 2006-2010 | 65 (43.3%) | 20 (27.4%) | 6 (26.1%) | |
| 2011-2015 | 85 (56.7%) | 53 (72.6%) | 17 (73.9%) | |
| Primary Diagnosis at Transplant | 0.30 | |||
| Bone Marrow Failure | 13 (8.7%) | 4 (5.5%) | 3 (13%) | |
| Hemoglobinopathies | 2 (1.3%) | 1 (1.4%) | 0 (0%) | |
| Lymphoma | 23 (15.3%) | 14 (19.2%) | 1 (4.3%) | |
| Acute leukemia | 104 (69.3%) | 48 (65.8%) | 18 (78.3%) | |
| Chronic leukemia | 8 (5.3%) | 6 (8.2%) | 1 (4.3%) | |
| Cord units | 0.64 | |||
| Single | 30 (20%) | 7 (9.6%) | 3 (13%) | |
| Double | 120 (80%) | 66 (90.4%) | 20 (87%) | |
| Patient CMV Status Pre-Transplant | 0.36 | |||
| Negative | 29 (19.3%) | 12 (16.4%) | 2 (8.7%) | |
| Positive | 121 (80.7%) | 61 (83.6%) | 21 (91.3%) | |
| Karnofsky performance status % | 0.12 | |||
| 90-100 | 100 (66.7%) | 56 (76.7%) | 13 (56.5%) | |
| <90 | 31 (20.7%) | 14 (19.2%) | 7 (30.4%) | |
| Unknown | 19 (12.7%) | 3 (4.1%) | 3 (13%) | |
| DRI score | 0.032 | |||
| Low | 16 (10.7%) | 12 (16.4%) | 1 (4.3%) | |
| Intermediate | 69 (46%) | 38 (52.1%) | 8 (34.8%) | |
| High | 50 (33.3%) | 18 (24.7%) | 11 (47.8%) | |
| Very High | 7 (4.7%) | 2 (2.7%) | 3 (13%) | |
| NA | 8 (5.3%) | 3 (4.1%) | 0 (0%) | |
| HCT comorbidity index | 0.27 | |||
| 0 | 77 (51.3%) | 36 (49.3%) | 9 (39.1%) | |
| 1-2 | 25 (16.7%) | 19 (26%) | 4 (17.4%) | |
| >2 | 30 (20%) | 15 (20.5%) | 7 (30.4%) | |
| Unknown | 18 (12%) | 3 (4.1%) | 3 (13%) | |
| Conditioning Regimen | 0.99 | |||
| Myeloablative | 87 (58%) | 35 (47.9%) | 11 (47.8%) | |
| Non-myeloablative/Reduce intensity | 63 (42%) | 38 (52.1%) | 12 (52.2%) | |
| Conditioning regimen | 0.34 | |||
| FTBI based | 81 (54%) | 36 (49.3%) | 10 (43.5%) | |
| Fludarabine/Cytoxan based | 14 (9.3%) | 8 (11%) | 3 (13%) | |
| Fludarabine/Melphalan | 49 (32.7%) | 29 (39.7%) | 9 (39.1%) | |
| Busulfan/Cytoxan | 6 (4%) | 0 (0%) | 1 (4.3%) | |
| HLA match out of 6 | 0.85 | |||
| 4 | 93 (62%) | 50 (68.5%) | 16 (69.6%) | |
| 5 | 48 (32%) | 22 (30.1%) | 7 (30.4%) | |
| 6 | 9 (6%) | 1 (1.4%) | 0 (0%) | |
| Total TNC, x107 | 0.33 | |||
| Median (Range) | 4.9 (1.8-26.6) | 5.0 (2.9-26.6) | 4.7 (2.6-12.9) | |
| Interquartile range | 3.8, 6.7 | 3.9, 6.3 | 3.4, 6.7 | |
| 0.28 | ||||
| ≤4.9 | 75 (50%) | 35 (47.9%) | 14 (60.9%) | |
| >4.9 | 75 (50%) | 38 (52.1%) | 9 (39.1%) | |
| Total CD34, x105 | 0.27 | |||
| Median (Range) | 1.8 (0.2-122.4) | 1.9 (0.5-122.4) | 2.0 (0.6-54.2) | |
| Interquartile range | 1.2, 2.9 | 1.2, 2.8 | 1.6, 4.2 | |
| 0.38 | ||||
| ≤1.8 | 75 (50%) | 33 (45.2%) | 8 (34.8%) | |
| >1.8 | 75 (50%) | 40 (54.8%) | 15 (65.2%) | |
| GVHD prophylaxis | 0.15 | |||
| Methotrexate-based | 6 (4%) | 0 (0%) | 1 (4.3%) | |
| Mycophenolate-based | 119 (79.3%) | 59 (80.8%) | 16 (69.6%) | |
| Sirolimus-based | 25 (16.7%) | 14 (19.2%) | 6 (26.1%) | |
| SF level prior to HCT | ||||
| Median | 1210.0 | 950.8 | 2786.1 | |
| Range | (11.9-12957.3) | (11.9-1964.0) | (2110.0-12957.3) | |
Based on chi-square or Wilcoxon test whenever appropriate.
Outcomes of UCB-HCT
With a median follow-up of 67.0 months (range: 5.8- 148.9), 2- and 5-year DFS for the whole cohort were 41.8% (95%CI: 33.8%-49.5%) and 36.8% (95%CI: 28.9%-44.6%), respectively. OS at 2 and 5 years were 47.8% (95%CI: 39.6%-55.5%) and 38.7% (95%CI: 30.6%-46.7%), respectively. Cumulative incidence of NRM at 2 and 5 years were 35.5% (95%CI: 27.8%-43.1%) and 39.7% (95%CI: 31.6%-47.7%) and Cumulative incidence of relapse at 2 and 5 years were 22.8% (95%CI: 16.4%-29.8%) and 23.5% (95%CI: 17.0%-30.6%), respectively. Day 100 grade II-IV and grade III/IV acute GvHD were 56.7% (95%CI: 48.3%-64.2%) and 28.7% (95%CI: 21.6%-36.1%), respectively. Cumulative incidence of chronic GvHD at 2 years was 52.3% (95%CI: 43.9%-60.0%) with 39.5% (95%CI: 31.6%-47.2%) of patients experiencing extensive chronic GvHD.
In multivariable analysis, intensity of the conditioning regimen significantly affected OS (HR: 1.70, 95%CI: 1.12-2.58; p=0.011), DFS (HR: 1.84, 95%CI: 1.22-2.77; p=0.003) and relapse rate (HR: 3.88, 95%CI: 1.80-8.38; p<0.001). Conditioning regimen had no significant effect on the NRM (P=0.98). At day +100 post-HCT, rates of grade II-IV acute GvHD were 64% and 46% in recipients of MAC and RIC, respectively (p=0.009). (Table 2 and Supplementary Table 1)
Table 2.
Association between serum ferritin level prior to HCT and outcomes
| Serum ferritin level prior to HCT | ||||||
|---|---|---|---|---|---|---|
| ≤2000 (n=73) | >2000 (n=23) | |||||
| 2 Yr (95%CI) | 2 Yr (95%CI) | HR (95%CI) | P* | Adjusted HR (95%CI) † | P† | |
| OS | 0.521(0.401,0.628) | 0.261(0.106,0.447) | 2.25(1.28,3.93) | 0.003 | 2.26(1.28,4.00) | 0.005 |
| DFS | 0.438(0.323,0.548) | 0.261(0.106,0.447) | 1.84(1.06,3.20) | 0.025 | 1.82(1.04,3.18) | 0.037 |
| Relapse | 0.260(0.165,0.365) | 0.174(0.051,0.358) | 0.50(0.16,1.55) | 0.23 | ||
| NRM | 0.301(0.200,0.409) | 0.565(0.333,0.744) | 2.18(1.10,4.31) | 0.025 | ||
| Acute GVHD 100-day | 0.562(0.439,0.667) | 0.478(0.261,0.667) | 0.83(0.44,1.59) | 0.54 | ||
| Chronic GVHD | 0.575(0.452,0.681) | 0.435(0.216,0.636) | 0.70(0.35,1.42) | 0.36 | ||
| Neutrophil Engraftment | ||||||
| ≤28 days | 58(79.5%) | 14(60.9%) | 0.17 | |||
| 29-42 days | 9(12.3%) | 4(17.4%) | ||||
| >42 days | 3(4.1%) | 2(8.7%) | ||||
| No | 3(4.1%) | 3(13.0%) | ||||
| 28-day | 0.795(0.680,0.872) | 0.609(0.373,0.779) | 0.58(0.35,0.96) | 0.034 | ||
| 42-day | 0.918(0.821,0.964) | 0.783(0.535,0.908) | ||||
| Platelet Engraftment | ||||||
| 3-month | 0.808(0.695,0.883) | 0.522(0.294,0.708) | 0.48(0.23,0.98) | 0.044 | ||
Based on univariate analysis, log-rank test for OS and DFS, Gray’s test for relapse, NRM, aGVHD, cGVHD, neutrophil engraftment, and platelet engraftment, and Fisher’s exact test for neutrophil engraftment in 4 groups
Based on the multivariable Cox regression model adjusting for age, conditioning regimen, HLA matching, and TNC for OS and DFS; based on the proportional subdistribution hazards model for competing risks and adjusting for age, DRI, conditioning for relapse, and adjusting for age, KPS, HLA, total TNC, conditioning, and disease type for NRM; Based on the proportional subdistribution hazards model for competing risks and adjusting for age, KPS, HLA, total TNC, conditioning, and disease type for engraftment.
UCB HCT Outcomes and Iron Parameters
High level of SF (>2000 ng/ml) was significantly associated with worse OS, shorter DFS and higher NRM, compared to lower SF levels (≤2000 ng/ml) in both univariate and multivariable analyses (Table 2). Higher SF levels was significantly associated with shorter DFS in both univariate (HR: 1.84; 95% CI: 1.06-3.20; p= 0.025) and multivariable analysis (adjusted HR: 1.82; 95% CI: 1.04-3.18; p= 0.037). Figure 1 shows association of SF levels with two-year OS and NRM. OS rate at 2 years (Figure 1a) was significantly higher in patients with SF levels of ≤2000 ng/ml (52%, 95%CI: 0.40-0.63), compared with patients with SF levels of >2000 ng/ml, (26%, 95%CI: 0.11-0.45, P value = 0.003). By multivariate analysis, 2-year OS remained significant with adjusted HR of 2.26 (95% CI: 1.28-4.00, P value = 0.005). (Table 2) Two-year NRM rate (Figure 1b) was significantly higher among patients with SF levels of >2000 ng/ml (56%, 95%CI: 0.33-0.74) compared to lower SF levels of <2000 ng/ml (30%, 95%CI: 0.20-0.41, P value=0.006). NRM remained significant with adjusted HR of 2.18 (95%CI: 1.10-4.31, P value = 0.025) (Table 2).
Figure 1.
Kaplan-Meier curves comparing survival outcomes in recipients of UCB-HCT with serum ferritin levels of ≤2000 ng/ml and >2000 ng/ml. (a) Overall survival, and (b) Non-relapse mortality
Iron Overload and Engraftment
We examined the association between iron overload and neutrophil and platelet engraftment after UCB-HCT. Cumulative incidence of neutrophil engraftment at 42 days post-HCT was 78% (95%CI: 0.53-0.91) among patients with SF levels of >2000 ng/ml and 92%, (95%CI: 0.82-0.96) among patients with SF levels of ≤2000 ng/ml, with HR=0.58 (95%CI: 0.35-0.96), with an adjusted P value of 0.034. (Figure 2 and Table 2). A significant difference was also detected in platelet engraftment based on SF levels (Figure 2). At 3 months post-HCT, cumulative incidence of platelet engraftment in patients with SF levels of >2000 ng/ml and ≤2000 ng/ml were at 52% (95%CI: 0.29-0.71) and 81% (95%CI: 0.69-0.88) respectively, with HR of 0.48 (95%CI: 0.23-0.98) and adjusted P-value of 0.044 (Table 2). Association between engraftment and baseline characteristics are detailed in Supplementary Table 2.
Figure 2.
Kaplan-Meier curves comparing engraftment outcomes in recipients of UCB-HTB with serum ferritin levels of ≤2000 ng/ml and >2000 ng/ml. (a) cumulative incidence of neutrophil engraftment, and (b) cumulative incidence of platelet engraftment
Iron overload and infectious and other complications after HCT
Of the 69 patients, 39 patients (72%) developed BSI, 13 patients had invasive fungal infection, and 53 had CMV reactivation within 100 days after UCB-HCT. Among whom 45 patients developed either BSI or fungal infection, 7 had both infections, 32 had BSI only, and 6 had fungal infection only. Significantly higher rate of BSI or fungal infection occurred in patients with SF level >2000 ng/ml compared to patients with SF levels of ≤2000 ng/ml (60.9% vs. 42.5% at day 100, HR=1.88; 95%CI: 1.01-3.50, P=0.047). (data not shown). Significantly higher proportion of causes of death due to infection was observed in patients with SF levels of >2000 ng/ml compared to patients with SF levels of ≤2000 ng/ml (55.6% vs. 26.2%, p=0.04) (Table 3). Other causes of death including relapse, GVHD, multiple organ failure (MOF) were not significantly different when patients were stratified by SF. (Table 3)
Table 3.
Association of serum ferritin with cause of death post-HCT
| Serum Ferritin Level | ||||
|---|---|---|---|---|
| ≤2000 (N=42) | >2000 (N=18) | Total (N=60) | p value | |
| Cause of death | 0.30 | |||
| Infection* | 11 (26.2%) | 10 (55.6%) | 21 (35%) | |
| Relapse | 16 (38.1%) | 3 (16.7%) | 19 (31.7%) | |
| GvHD | 5 (11.9%) | 2 (11.1%) | 7 (11.7%) | |
| Unknown | 4 (9.5%) | 2 (11.1%) | 6 (10%) | |
| MOF | 3 (7.1%) | 1 (5.6%) | 4 (6.7%) | |
| Other** | 3 (7.1%) | 0 (0%) | 3 (5%) | |
| Infection* | 0.040 | |||
| Yes | 11 (26.2%) | 10 (55.6%) | 21 (35%) | |
| No | 31 (73.8%) | 8 (44.4%) | 39 (65%) | |
Infection refers to both bacterial and fungal infections.
graft failure (n=1), second malignancy (n=1) and intracranial bleed (n=1)
Association between bone marrow Fe and SF
Pre-HCT SF level was available in 96 patients (65%), with a median value of 1210 ng/ml (range: 11.9-12957.3). SF level of >2000 ng/ml was detected in 23 patients (24%). Younger patients who underwent HCT in earlier years of this study period were less likely to have SF due to the change in our standard practice of routinely measuring SF prior to HCT. Bone marrow Fe levels were scored by the same hematopathologist on pre-HCT bone marrow biopsies, which was available in 60 patients (40%). Bone marrow Fe scores were 3 and 4 in 25 and 22 patients, respectively; and 6, 5, and 2 patients had scores of 0, 1, and 2, respectively; showing that the majority of the patients in this study were iron overloaded. A modest correlation was detected between bone marrow Fe score and SF levels (Spearman correlation coefficient =0.40; p value =0.007) among the 45 patients who had both measurements available.
DISCUSSION
Iron overload is an adverse prognostic factor for allogeneic HCT, 1–4 but the impact of iron overload on outcomes of UCB-HCT has not yet been examined. UCB-HCT has unique characteristics of engraftment delay and higher infection risk that could be considered as a consequence of iron overload on the bone marrow microenvironment and hematopoietic progenitor cell function. Notwithstanding the limitations inherent in a retrospective study of this size, our analysis indicates that iron overload defined as SF levels >2000 ng/ml impacts OS and NRM at 2 years post-UCB-HCT. These adverse impacts may, at least in part, be related to the deleterious effects of iron overload on neutrophil engraftment and the consequent increase in risk of bacterial/fungal infection.
Most of the previous studies examining the impact of SF levels on HCT outcomes have used an SF level cutoff of 1000 ng/ml. However, outcomes of HCT have been shown to progressively worsen with higher levels of pre-HCT SF.1,2 In order to improve the specificity of SF level as a marker of clinically significant iron overload, in this study we used a SF level cutoff value of 2000 ng/ml. This cutoff value was chosen based on the available data indicating that this level correlates better with clinically significant liver iron concentration as well as iron deposition in organs such as heart.21 Choosing the SF value of >2000 ng/ml to define iron overload identifies almost all patients with liver iron concentration ≥7 mg Fe/g, which is generally considered clinically relevant.22 Bone marrow macrophages are a storage site for iron and studies have shown bone marrow iron score is associated with inferior survival as well as higher rates of bacterial infection after allogeneic HCT.23,24 Although we observed a modest correlation between bone marrow Fe score and SF levels, we did not find an association between bone marrow Fe score and HCT outcome, possibly because the number of cases in our cohort who had both pre-HCT bone marrow Fe score and SF levels available was limited (N=45, data not shown).
HCT causes profound alterations in iron metabolism during the aplastic period resulting from the conditioning regimen. It is important to mention that due to ablation of erythropoiesis by the conditioning regimen, there will be a rise in plasma non-transferrin-bound iron (NTBI) post-HCT. Normally, most of the plasma iron is bound to serum transferrin, which protects tissues from exposure to toxic NTBI. However, in conditions such as iron overload and post-HCT, NTBI and its highly reactive component (labile plasma iron) can appear in plasma, thus causing tissue damage.25 A significant correlation has been described between SF levels and NTBI in adult HCT recipients.26 Since erythropoiesis utilizes NTBI, ablation of erythropoiesis by the conditioning regimen leads to marked elevation in NTBI levels, which declines after red cell progenitor engraftment.27 In the case of UCB-HCT, the delayed red cell engraftment further exposes tissues to the deleterious effects of NTBI exposure for a longer time period compared to other graft sources. Moreover, prolonged period of NTBI availability to the bacteria and fungi (especially mold) and the prolonged neutropenia period observed in UCB-HCT could further increase the infection risk. Hence, UCB-HCT patients could be particularly susceptible to harmful effects of iron overload. In a study by Hilken et al, NTBI levels measured on day 0 was a better predictor, compared with SF level, for BSI risk after allogeneic HCT.28 In accordance to this study, our data indicated higher rates of BSI and fungal infection in the high SF group.
Iron overload can impair hematopoietic progenitor cell function and the BM microenvironment, both of which could have adverse consequences in UCB-HCT, particularly in adult patients with low progenitor cell numbers. Animal models have provided an insight to the possible mechanisms by which iron overload adversely impact the bone marrow function. In one study, iron overload in mice led to impaired clonigenic function of bone marrow progenitors by induction of reactive oxygen species.8 In another study, transplantation of normal hematopoietic progenitor cells into iron overloaded mice resulted in delayed hematopoietic reconstitution, which was associated with a reduction in gene expression of homing molecules in bone marrow stromal cells.9 Our observation of delayed neutrophil and platelet engraftment in patients with iron overload undergoing UCB-HCT is consistent with outcomes of these preclinical studies and was independent of HLA matching and total nucleated cell dose infused. Interestingly, in our cohort HLA disparities, total nucleated cell counts, and CD34 counts were not significant factor determining engraftment post-HCT; most likely due to the minimal matching criteria and the standard cell dose in our practice, which is aligned with established guidelines.29,30
At the present time, with the advances in alternative donor transplantation, graft source and donor selection for patients who lack a fully matched related/unrelated donor is made between mismatched unrelated, UCB or haploidentical donor. Although our study carries the inherent limitations of a single center retrospective study, our data suggest that iron overload should be considered as a strong adverse prognostic factor in patients transplanted with UCB as the graft source. Effective iron chelation prior to UCB-HCT should be considered especially for patients with non-malignant disorders and in whom HCT is not urgently required.
Supplementary Material
Highlights.
Impact of iron overload (SF>2000 ng/ml) on UCB-HCT outcomes was examined.
OS rate at 2 years was significantly lower in patients with higher pre-HCT SF levels (p=0.005)
NRM was significantly higher in patients with higher SF (p=0.02)
Significantly faster engraftment was seen in patients with lower SF levels.
Iron overload is a strong adverse prognostic factor for cord blood transplant.
AKNOWLEDGMENTS
Authors thank City of Hope staff and nurses, as well as the patients and their families, without whom this work would not be possible. This study was partially supported by NIH P30 CA033572 (Biostatistics Core).
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
COMPETING INTERESTS
VP has served on Advisory Boards and Speakers Bureau for Novartis. The remaining authors have no relevant conflicts of interest to declare.
REFERENCES
- 1.Pullarkat V, Blanchard S, Tegtmeier B, et al. Iron overload adversely affects outcome of allogeneic hematopoietic cell transplantation. Bone Marrow Transplant. 2008;42(12):799–805. [DOI] [PubMed] [Google Scholar]
- 2.Armand P, Kim HT, Cutler CS, et al. Prognostic impact of elevated pretransplantation serum ferritin in patients undergoing myeloablative stem cell transplantation. Blood. 2007;109(10):4586–4588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Wermke M, Schmidt A, Middeke JM, et al. MRI-based liver iron content predicts for nonrelapse mortality in MDS and AML patients undergoing allogeneic stem cell transplantation. Clin Cancer Res. 2012;18(23):6460–6468. [DOI] [PubMed] [Google Scholar]
- 4.Yan Z, Chen X, Wang H, et al. Effect of pre-transplantation serum ferritin on outcomes in patients undergoing allogeneic hematopoietic stem cell transplantation: A meta-analysis. Medicine (Baltimore). 2018;97(27):e10310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Olthof AW, Sijens PE, Kreeftenberg HG, et al. Correlation between serum ferritin levels and liver iron concentration determined by MR imaging: impact of hematologic disease and inflammation. Magn Reson Imaging. 2007;25(2):228–231. [DOI] [PubMed] [Google Scholar]
- 6.Jacobi N, Herich L. Measurement of liver iron concentration by superconducting quantum interference device biomagnetic liver susceptometry validates serum ferritin as prognostic parameter for allogeneic stem cell transplantation. Eur J Haematol. 2016;97(4):336–341. [DOI] [PubMed] [Google Scholar]
- 7.Bullen JJ, Rogers HJ, Spalding PB, Ward CG. Iron and infection: the heart of the matter. FEMS Immunol Med Microbiol. 2005;43(3):325–330. [DOI] [PubMed] [Google Scholar]
- 8.Chai X, Li D, Cao X, et al. ROS-mediated iron overload injures the hematopoiesis of bone marrow by damaging hematopoietic stem/progenitor cells in mice. Sci Rep. 2015,5:10181. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Okabe H, Suzuki T, Uehara E, Ueda M, Nagai T, Ozawa K. The bone marrow hematopoietic microenvironment is impaired in iron-overloaded mice. Eur J Haematol. 2014;93(2):118–128. [DOI] [PubMed] [Google Scholar]
- 10.Zhang Y, Zhai W, Zhao M, et al. Effects of iron overload on the bone marrow microenvironment in mice. PLoS One. 2015;10(3):e0120219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jin X, He X, Cao X, et al. Iron overload impairs normal hematopoietic stem and progenitor cells through reactive oxygen species and shortens survival in myelodysplastic syndrome mice. Haematologica. 2018;103(10):1627–1634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Giralt S, Ballen K, Rizzo D, et al. Reduced-intensity conditioning regimen workshop: defining the dose spectrum. Report of a workshop convened by the center for international blood and marrow transplant research. Biol Blood Marrow Transplant. 2009;15(3):367–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Armand P, Kim HT, Logan BR, et al. Validation and refinement of the Disease Risk Index for allogeneic stem cell transplantation. Blood. 2014;123(23):3664–3671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sorror ML, Logan BR, Zhu X, et al. Prospective Validation of the Predictive Power of the Hematopoietic Cell Transplantation Comorbidity Index: A Center for International Blood and Marrow Transplant Research Study. Biol Blood Marrow Transplant. 2015;21(8):1479–1487. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Flomenberg N, Baxter-Lowe LA, Confer D, et al. Impact of HLA class I and class II high-resolution matching on outcomes of unrelated donor bone marrow transplantation: HLA-C mismatching is associated with a strong adverse effect on transplantation outcome. Blood. 2004;104(7):1923–1930. [DOI] [PubMed] [Google Scholar]
- 16.Glucksberg H, Storb R, Fefer A, et al. Clinical manifestations of graft-versus-host disease in human recipients of marrow from HL-A-matched sibling donors. Transplantation. 1974;18(4):295–304. [DOI] [PubMed] [Google Scholar]
- 17.Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med. 1980;69(2):204–217. [DOI] [PubMed] [Google Scholar]
- 18.Fine JP, Gray RJ. A Proportional Hazards Model for the Subdistribution of a Competing Risk. Journal of the American Statistical Association. 1999;94(446):496–509. [Google Scholar]
- 19.GRAMBSCH PM, THERNEAU TM. Proportional hazards tests and diagnostics based on weighted residuals. Biometrika. 1994;81(3):515–526. [Google Scholar]
- 20.Li J, Scheike TH, Zhang M-J. Checking Fine and Gray subdistribution hazards model with cumulative sums of residuals. Lifetime data analysis. 2015;21(2):197–217. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Meloni A, Casale M, Filosa A, et al. Association between serum ferritin and liver iron concentration with cardiac iron in pediatric thalassemia major patients. Journal of Cardiovascular Magnetic Resonance. 2016;18(1):P295. [Google Scholar]
- 22.Taher AT, Porter JB, Viprakasit V, et al. Defining serum ferritin thresholds to predict clinically relevant liver iron concentrations for guiding deferasirox therapy when MRI is unavailable in patients with non-transfusion-dependent thalassaemia. Br J Haematol. 2015;168(2):284–290. [DOI] [PubMed] [Google Scholar]
- 23.Sivgin S, Nazlim S, Zararsiz G, et al. Increased Bone Marrow Iron Scores Are Strongly Correlated With Elevated Serum Ferritin Levels and Poorer Survival in Patients With Iron Overload That Underwent Allogeneic Hematopoietic Stem Cell Transplantation: A Single Center Experience. Clin Lymphoma Myeloma Leuk. 2016;16(10):582–587. [DOI] [PubMed] [Google Scholar]
- 24.Ohmoto A, Fuji S, Miyagi-Maeshima A, et al. Association between pretransplant iron overload determined by bone marrow pathological analysis and bacterial infection. Bone Marrow Transplant. 2017;52(8):1201–1203. [DOI] [PubMed] [Google Scholar]
- 25.Cabantchik ZI, Breuer W, Zanninelli G, Cianciulli P. LPI-labile plasma iron in iron overload. Best Pract Res Clin Haematol. 2005;18(2):277–287. [DOI] [PubMed] [Google Scholar]
- 26.Goto T, Ikuta K, Inamoto Y, et al. Hyperferritinemia after adult allogeneic hematopoietic cell transplantation: quantification of iron burden by determining non-transferrin-bound iron. Int J Hematol. 2013;97(1):125–134. [DOI] [PubMed] [Google Scholar]
- 27.Sahlstedt L, Ebeling F, von Bonsdorff L, Parkkinen J, Ruutu T. Non-transferrin-bound iron during allogeneic stem cell transplantation. Br J Haematol. 2001;113(3):836–838. [DOI] [PubMed] [Google Scholar]
- 28.Hilken A, Langebrake C, Wolschke C, et al. Impact of non-transferrin-bound iron (NTBI) in comparison to serum ferritin on outcome after allogeneic stem cell transplantation (ASCT). Ann Hematol. 2017;96(8):1379–1388. [DOI] [PubMed] [Google Scholar]
- 29.Laughlin MJ, Eapen M, Rubinstein P, et al. Outcomes after transplantation of cord blood or bone marrow from unrelated donors in adults with leukemia. N Engl J Med. 2004;351(22):2265–2275. [DOI] [PubMed] [Google Scholar]
- 30.Wagner JE, Barker JN, DeFor TE, et al. Transplantation of unrelated donor umbilical cord blood in 102 patients with malignant and nonmalignant diseases: influence of CD34 cell dose and HLA disparity on treatment-related mortality and survival. Blood. 2002;100(5):1611–1618. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.