Polymorphism in the Interleukin-7 Receptor-alpha and Outcome after Allogeneic Hematopoietic Cell Transplantation with Matched Unrelated Donor

Abstract

Interleukin-7 (IL-7) is essential for T cell development in the thymus and maintenance of peripheral T cells. The α-chain of the IL-7R is polymorphic with the existence of SNPs that give rise to nonsynonymous amino acid substitutions. We previously found an association between donor genotypes and increased treatment related mortality (TRM) (rs1494555G) and acute graft versus host disease (aGvHD) (rs1494555G and rs1494558T) after hematopoietic cell transplantation (HCT). Some studies have confirmed an association between rs6897932C and multiple sclerosis. In the present study we evaluated the prognostic significance of IL-7Rα SNP genotypes in 590-recipient/donor pairs that received HLA matched unrelated donor HCT for hematological malignancies. Consistent with the primary studies, the rs1494555GG and rs1494558TT genotypes of the donor were associated with aGvHD and chronic GvHD in the univariate analysis. The T allele of rs6897932 was suggestive of an association with increased frequency of relapse by univariate analysis (p=0.017) and multivariate analysis (p=0.015). In conclusion, this study provides further evidence of a role of the IL-7 pathway and IL-7Rα SNPs in HCT.

Introduction

Interleukin-7 (IL-7) is essential for T cell development in the thymus [1] and maintenance of peripheral T cells [2]. IL-7 receptor (IL-7R) consists of the common gamma-chain (CD132) as well as an α-chain (CD127). The α-chain of the IL-7R is polymorphic with the existence of 4 non-synonymous single nucleotide polymorphisms (SNPs) in the exons; rs1494558 (+510C/T in exon 2), rs1494555 (+1237A/G in exon 4), rs6897932 (+2087T/C in exon 6), and rs3194051 (+3101A/G in exon 8) that all give rise to amino acid substitutions [3;4]. The α-chain is also used by the receptor of thymic stromal lymphopoitin (TSLP), a cytokine with complex effects on cytokine profiles, including stimulation of TNF production by dendritic cells (DC) and the induction of Th2 cytokines [3;5;6]. Furthermore, it has been shown that in human thymus TSLP activates DC which induces the generation of regulatory T cells (Tregs) [7].

Recipients of hematopoietic stem cell transplantation (HCT) suffer from a prolonged post-transplant immune deficiency that results in significant morbidity and mortality [8]. Reconstitution of the T cell population involves both thymus-dependent de novo T cell generation as well as extra-thymic expansion of mature, donor derived T cells and studies in mice indicate that IL-7 may be critically involved in both of these processes [9].

Based on the known functions of IL-7 and TSLP, we hypothesized that polymorphisms in exons of the IL-7Rα gene might influence the process of immune reconstitution after HCT impacting the risk of infections, acute and chronic graft versus host disease (GvHD), and treatment related mortality (TRM). In a previously published study of a Danish HCT cohort, we found an association between donor rs1494555G and rs1494558T and increased TRM after HLA matched unrelated donor (MUD) HCT [10]. The aim of the present study was to validate these findings in an independent, larger and more homogeneous cohort of adults receiving MUD HCT for hematological malignancies. In addition, we evaluated the significance of rs6897932 genotypes in relation to HCT since this SNP has previously been associated with autoimmune disease and allergy [11;12].

Patients, Material and Methods

Data collection

Established in 2004, the Center for International Blood and Marrow Transplant Research (CIBMTR) is a research affiliation of the International Bone Marrow Transplant Registry (IBMTR), Autologous Blood and Marrow Transplant Registry (ABMTR) and the National Marrow Donor Program (NMDP), and is comprised of a voluntary working group of more than 450 transplantation centers worldwide that contribute detailed data on consecutive allogeneic and autologous HCT to a Statistical Center at the Medical College of Wisconsin in Milwaukee (WI, USA) and the NMDP Coordinating Center in Minneapolis (MN, USA). Participating centers are required to report all transplants consecutively; compliance is monitored by on-site audits. Patients are followed longitudinally, with yearly follow-up. Computerized checks for discrepancies, physicians’ review of submitted data, and on-site audits of participating centers ensure data quality.

Ethics Statement

Observational studies conducted by the CIBMTR are performed in compliance with the Privacy Rule (HIPAA) as a Public Health Authority, and in compliance with all applicable federal regulations pertaining to the protection of human research participants as determined by continuous review of the Institutional Review Boards (IRB) of the NMDP and the Medical College of Wisconsin.

Study Population

The study population consisted of 590 donor-recipients pairs receiving a bone marrow (BM) or growth factor mobilized peripheral blood stem cell (PBSC) transplant following a myeloablative conditioning regimen between 1988 and 2004 facilitated through the National Marrow Donor Program (NMDP). All donors and recipients were Caucasian and over 18 years old. Diagnoses included acute myeloid leukaemia (AML) (n=111), acute lymphoblastic leukaemia (ALL) (n=76), chronic myeloid leukaemia (CML) (n=373), or myelodysplastic syndrome (MDS) (n=30) (Table 1). Early disease was defined as patients with ALL and AML in first complete remission, CML in first chronic phase and MDS with refractory anemia, or refractory anemia with ringed sideroblasts. Intermediate was defined as ALL and AML in second or greater complete remission, CML in accelerated phase, or second or greater chronic phase. Because patients with advanced disease have high treatment-related mortality and relapse rates even in the fully matched setting, CIBMTR usually excludes these patients from analyses focused on testing the association of HLA and other genetic factors with clinical outcomes.

Table 1.

Characteristics of Caucasian patients with AML, ALL, CML or MDS receiving a myeloablative conditioning regimen and a bone marrow or peripheral blood transplant from an unrelated Caucasian donor allele-level matched for HLA-A, -B, -C, -DRB1 and -DQB1 from the National Marrow Donor Program.

Variable	N Eval	N (%)
Number of patients		590
Number of centers		78
Age, median (range), years	590	38 (18–62)
Age at transplant	590
18–19 yr		18 (3)
20–29 yr		126 (21)
30–39 yr		175 (30)
40–49 yr		202 (34)
50 and older		69 (12)
Male sex	590	344 (58)
Karnofsky prior to transplant ≥90	575	470 (82)
Disease at transplant	590
AML		111 (19)
ALL		76 (13)
CML		373 (63)
MDS		30 (5)
Disease stage at transplant	590
Early		426 (72)
Intermediate		164 (28)
Use of ATG	590	49 (8)

All transplantation pairs were 10/10 allele-matched at HLA-A, B, C, DRB1 and DQB1 with HLA typing validated through the ongoing NMDP retrospective high resolution typing program [13]. All surviving unrelated recipients included in this analysis were retrospectively contacted and provided informed consent for participation in the NMDP/CIBMTR research program. Approximately 9% of surviving patients would not provide consent for use of the research data. To adjust for the potential bias introduced by exclusion of non-consenting surviving patients, a corrective action plan modeling process randomly excluded appropriately the same percentage of deceased patients using a biased coin randomization with exclusion probabilities based on characteristics associated with not providing consent for use of the data in survivors [14]. Patient-, disease-, and transplant- related characteristics are listed in Table 1.

Endpoints

The objective of this study was to evaluate the impact of IL-7Rα polymorphisms in the donor and recipient on the outcomes of HCT. The main outcomes analyzed were TRM, relapse, acute and chronic GvHD, disease free survival (DFS), and overall survival (OS). Relapse consisted of leukemia recurrence, whereas TRM was death in the absence of relapse. The acute GvHD (aGvHD) end-point referred to the development of grades II-IV and grades III-IV according to the Glucksberg criteria [15]. Chronic GvHD (cGvHD) was diagnosed following the standard definitions [16]. DFS was defined as survival in complete remission after HCT. For OS, from any cause was considered an event. All living patients were censored at last follow-up.

Detection of IL-7Rα SNPs by Sequence Specific Primer-PCR (SSP-PCR)

IL-7Rα polymorphisms (rs1494558, rs1494555, rs6897932 and rs3194051) were determined by an SSP-PCR system in genomic DNA extracted from banked pre-transplant donor and recipient blood samples from the NMDP Research Repository (Minneapolis, MN). The genomic DNA extraction was performed by Maxwell™ 16 blood DNA Purification Kit (Promega).

The SSP-PCR reactions were set up in a total volume of 10 µl with control primer (0.2 µM) and specific primer (0.5 µM) as described previously [10]. A GeneAmp PCR System 9700 (Applied Biosystems) and Mastercycler eppendorf (AH diagnostics) was used. To each PCR sample, 2 µl loading buffer was added, and the samples was ran for 30 min at 150V in gel electrophoresis of 2.5 % agarose (Medionova) stained with ethidium bromide (EtBr) (Sigma).

Statistics

Medians and ranges are reported for continuous variables and percentages for categorical variables. Probabilities for overall survival and disease-free survival were calculated using the Kaplan-Meier estimator. All other outcomes used the cumulative incidence estimator. All outcomes were compared using a point wise p-value at a specific point in time. Cox proportional hazards regression models were fit to the other outcomes. The proportional hazard assumption was assessed for each variable using a time-dependent approach. Variables used in the analysis include recipient age, Karnofsky performance score, use of ATG, disease, disease stage, stem cell source, GvHD prophylaxis, time from diagnosis to transplant for CML, CMV matching, year of transplant, donor sex, and number of donor pregnancies (Table 4). Stepwise model selection procedures were applied to build the models from the prognostic variables under consideration. We adopted a level of threshold (p-value <0.05) for variable selections. Each genetic marker was forced into the models that were built in the initial step and tested for association separately. Recipient genetic markers and donor genetic markers were treated separately in the analysis. Due to multiple testing, the p-values in the range 0.01 – 0.05 should be interpreted with caution test. For pair-wise linkage disequilibrium analysis the Lewontin’s D was used.

Table 4.

Univariate probability of outcomes based on donor IL-7Rα genotype^a)

SNP rs6897932	CC	TC	TT	P value
Overall survival - 5 years	47 (41–52)%	48 (41–54)%	42 (28–57)%	0.78
TRM - 5 years	43 (37–49)%	37 (31–44)%	40 (25–55)%	0.40
Relapse - 5 years	12 (9–16)%	20 (15–26)%	25 (13–40)%	0.0168
Grades 2–4 aGvHD - 100 days	56 (51–62)%	55 (48–61)%	44 (30–59)%	0.32
Grades 3–4 aGvHD - 100 days	26 (21–31)%	24 (19–30)%	12 (4–23)%	0.0442
Chronic GvHD - 1 year	51 (45–56)%	52 (45–58)%	49 (34–64)%	0.93
SNP rs1494558	CC	CT	TT	P value
Overall survival - 5 years	44 (38–50)%	49 (43–55)%	50 (37–64)%	0.47
TRM - 5 years	42 (36–48)%	39 (33–45)%	39 (26–53)%	0.74
Relapse - 5 years	17 (13–22)%	16 (12–21)%	13 (6–24)%	0.80
Grades 2–4 aGvHD - 100 days	51 (45–57)%	56 (50–62)%	74 (61–85)%	0.0047
Grades 3–4 aGvHD - 100 days	25 (20–31)%	23 (18–28)%	28 (17–41)%	0.65
Chronic GvHD - 1 year	51 (45–57)%	48 (42–55)%	73 (60–84)%	0.0022
SNP rs1494555	AA	AG	GG	P value
Overall survival - 5 years	45 (39–51)%	48 (42–54)%	50 (37–63)%	0.65
TRM - 5 years	41 (36–47)%	38 (32–45)%	42 (29–55)%	0.77
Relapse - 5 years	17 (13–21)%	17 (12–22)%	11 (4–20)%	0.40
Grades 2–4 aGvHD - 100 days	51 (46–57)%	56 (49–62)%	72 (59–83)%	0.0101
Grades 3–4 aGvHD - 100 days	25 (20–30)%	23 (18–29)%	26 (15–38)%	0.85
Chronic GvHD - 1 year	51 (45–57)%	48 (42–54)%	71 (58–82)%	0.0038

^a)Probabilities of acute, chronic GVHD, TRM, and relapse were calculated using the cumulative incidence estimate. Overall survival was calculated using the Kaplan-Meier product limit estimate.

Results

Genotype distribution

The IL-7Rα genotype frequencies of patients and donors were comparable (Table 2), and corresponded to previously reported gene frequencies [10;17]. The SNPs are in strong linkage disequilibrium (Table 3).

Table 2.

IL-7Rα SNP genotypes in patients, donors and controls (percentages). No significant differences were observed (p>0,05).

SNP Position	Patients		Donors
	N	n (%)	N	n (%)
rs1494558	587		584
CC		276 (47)		277 (47)
CT		251 (43)		254 (43)
TT		60 (10)		53 (9)
rs1494555	588		584
AA		277 (47)		279 (48)
AG		241 (41)		248 (42)
GG		70 (12)		57 (10)
rs6897932	587		583
CC		311 (53)		319 (55)
TC		224 (38)		221 (38)
TT		52 (9)		43 (7)
rs3194051	586		585
AA		335 (57)		294 (50)
AG		216 (37)		245 (42)
GG		35 (6)		46 (8)

Table 3.

Pair-wise linkage disequilibrium analysis: D’ (p-value)

Patients
Markers	rs1494558C/T	rs3194051A/G	rs6897932T/C
rs1494555A/G	0.86(<0.0001)	0.75(<0.0001)	0.73(<0.0001)
rs1494558C/T		0.98(<0.0001)	0.97(<0.0001)
rs3194051A/G			0.98(<0.0001)
Donors
Markers	rs1494558C/T	rs3194051A/G	rs6897932T/C
rs1494555A/G	0.96(<0.0001)	0.98(<0.0001)	0.96(<0.0001)
rs1494558C/T		1.00(<0.0001)	0.96(<0.0001)
rs3194051A/G			0.96(<0.0001)

IL-7Rα rs1494558 and rs1494555 and outcome

In the univariate analysis, IL-7Rα rs1494558 was found to be associated with grade II-IV aGVHD as well as cGVHD at 1 year, the probability being highest in patients receiving transplants from donors with TT genotype (Table 4 and Figure 1). A similar pattern was observed for IL-7Rα rs1494555, where the G allele was significantly associated with increased grade II-IV aGVHD and cGVHD. By multivariate analysis, however, these associations were not significant. Neither rs1494558 nor rs1494555 were associated with overall survival or TRM (Table 5).

Figure 1. Cumulative incidence of acute and chronic graft versus host disease (GVHD) based on IL-7R genotype.

(A) Acute GVHD grades II-IV at 100 days based on donor rs1494558 genotype (univariate p=0.005, multivariate p=0.09). (B) Chronic GVHD at one year based on donor rs1494558 genotype (univariate p=0.002, multivariate p=0.08). (C) Acute GVHD grades II-IV at 100 days based on donor rs1494555 genotype (univariate p=0.01, multivariate p=0.20). (D) Chronic GVHD at one year based on donor rs1494555 genotype (univariate p=0.004, multivariate p=0.15)

Table 5.

Multivariate analysis of the association of donor IL-7Rα genotype and outcome

SNP rs6897932	CC	TC	TT	Overall p-value
Overall survivala)	1.00	1.01 (0.80–1.28)	1.16 (0.78–1.74)	0.76
TRMb)	1.00	0.93 (0.71–1.22)	1.06 (0.64–1.77)	0.82
Relapsec)	1.00	1.56 (0.99–2.47)	2.63 (1.31–5.26)	0.0153
Grades 2–4 aGvHDd)	1.00	0.97 (0.77–1.23)	0.76 (0.47–1.22)	0.52
Grades 3–4 aGvHDe)	1.00	0.97 (0.68–1.38)	0.42 (0.17–1.04)	0.17
Chronic GvHDf)	1.00	1.04 (0.82–1.31)	0.96 (0.62–1.49)	0.92
SNP rs1494558	CC	CT	TT	Overall p-value
Overall survivala)	1.00	0.87 (0.69–1.09)	0.67 (0.45–1.00)	0.11
TRMb)	1.00	0.89 (0.68–1.16)	0.78 (0.49–1.24)	0.48
Relapsec)	1.00	0.82 (0.53–1.27)	0.67 (0.30–1.49)	0.49
Grades 2–4 aGvHDd)	1.00	1.15 (0.91–1.45)	1.48 (1.03–2.12)	0.09
Grades 3–4 aGvHDe)	1.00	0.87 (0.61–1.25)	0.95 (0.53–1.70)	0.76
Chronic GvHDf)	1.00	0.89 (0.70–1.13)	1.34 (0.94–1.92)	0.08
SNP rs1494555	AA	AG	GG	Overall p-value
Overall survivala)	1.00	0.91 (0.73–1.15)	0.69 (0.47–1.03)	0.18
TRMb)	1.00	0.91 (0.70–1.19)	0.85 (0.55–1.31)	0.68
Relapsec)	1.00	0.90 (0.58–1.38)	0.54 (0.23–1.27)	0.36
Grades 2–4 aGvHDd)	1.00	1.10 (0.87–1.40)	1.38 (0.97–1.95)	0.20
Grades 3–4 aGvHDe)	1.00	0.90 (0.63–1.29)	0.86 (0.48–1.54)	0.79
Chronic GvHDf)	1.00	0.91 (0.72–1.16)	1.30 (0.92–1.85)	0.15

^a)Adjusted for patient age, year of transplant and stratified by disease stage.

^b)Adjusted for disease stage and year of transplant.

^c)Adjusted for disease, Karnofsky score, time from diagnosis to transplant for CML, number of donor pregnancies.

^d)Adjusted for disease, disease stage, graft type and GvHD prophylaxis.

^e)Adjusted for disease, graft type, number of donor pregnancies and year of transplant.

^f)Adjusted for use of ATG in conditioning regimen, disease, donor sex, graft type, number of donor pregnancies and GvHD prophylaxis.

IL-7Rα rs6897932

By univariate and multivariate analysis IL-7Rα rs6897932TT genotype of the donor was suggestive of an association with increased frequency of relapse (overall p=0.015) compared to CC and CT donors (Figure 2, Table 4 and 5). The C allele was associated with increased risk of grade III-IV aGVHD by univariate analysis (Table 4), but the association did not hold in the multivariate model (Table 5). No association was found between IL-7Rα rs6897932 genotypes and OS or TRM.

Figure 2. Cumulative incidence of relapse at five years based on donor IL-7Rα SNP rs6897932 genotype (univariate p=0.02, multivariate p=0.015).

Other genotypes

The IL-7Rα rs3194051 genotypes of donors or recipients were not associated with any of the outcome parameters (data not shown).

Discussion

Reconstitution of the T cell population involves both thymus-dependent de novo T cell generation as well as extra-thymic expansion of mature, donor derived T cells. Based on the known functions of IL-7, we hypothesized that polymorphisms in exons of the IL-7Rα gene might influence the process of immune reconstitution after HCT, impacting the risk of GvHD and TRM. In a previously published study, we demonstrated an association between rs1494555 SNPs AG and GG genotypes of the donor and TRM in Danish patients receiving MUD HCT [10]. Moreover, in a recent study of a two centre British-French cohort of MUD and sibling donor HCT we found associations between both rs1494555GG and rs1494558TT genotypes of the donor and grade 3–4 aGvHD [17].

In the present study, univariate analysis was consistent with the previous finding of an association between the rs1494555GG and rs1494558TT genotype of the donor and aGvHD and indicated further that these genotypes are associated with increased risk of cGvHD. Although this did not reach significance in the multivariate analysis, these findings are of interest when considered in light of the previous results, because the bulk of other data appears to point towards an impact of these SNPs on adverse outcome in HCT. In addition to this a recently published paper showed increased risk of non-Hodgkin lymphoma with rs1494555GG [18], further indicating an impact of this SNP on T-cell homeostasis.

Several large multi-center studies have demonstrated a protective effect of the T allele of rs6897932 on the development of multiple sclerosis [12;19]. In line with this, we previously found that rs6897932 T is associated with reduced risk of inhalation allergy [11]. These data indicate a protective effect of rs6897932 T towards the development of inflammatory disease. Furthermore, SNP rs6897932 has been shown to predispose to sarcoid inflammation [20].

In the present study the T allele of rs6897932 in the donor was suggestive of an association with increased risk of relapse and a trend towards reduced risk of aGVHD. Since the graft versus leukaemia effect, as well as aGvHD, is induced by pro-inflammatory T cell responses, these findings appear to be in line with the previous observations in multiple sclerosis [12;19] and allergy [11]. The rs6897932 in relation to HCT was included in our previous studies [10], but associations with clinical outcome did not reach significance. This apparent discrepancy is most likely due to the fact that the previous studies were relatively small, and therefore not sufficiently powered to evaluate any impact of the rs6897932 minor allele, which is relatively infrequent (4%). Thus the present finding of an association between rs6897932 and relapse is novel, and will require confirmation in other large HCT cohorts.

The potential biological impact of rs6897932 is not yet understood. This SNP affects the amino acid composition of the transmembrane region of IL-7R α, and it is therefore possible that rs6897932 genotype affects the anchorage of the receptor in the plasma membrane. Alternatively spliced transcripts of human IL-7Rα were reported in leukemic cells from children with acute lymphoblastic leukaemia (ALL) [21]. Another study observed increased production of the soluble form of the IL-7Rα protein due to a two-fold increase in alternatively spliced transcripts that eliminated exon 6 [19]. Moreover, serum levels of sIL-7Rα has been associated with the Hap2 haplotype (counting rs6897932T), also associated with autoimmune disease [22]. Investigation of health controls demonstrate that an increase in sIL-7Rα is associated with the rs6897932 SNP, also found to be related to relapse in the present study with an approximately 3-fold increase in the median levels between the TT and CC genotype, and intermediate levels for the CT genotype [23].

The functional impact of sIL-7Rα on IL-7 activity is not known in vivo, but it was recently shown that in vitro, the native sIL-7R does interfere in optimal IL-7/IL-7Rα-signalling by significant inhibition of STAT5 and Bcl-2 phosphorylation [24]. It is likely that increased levels of sIL-7Rα may be associated with reduced IL-7 activity due to diminished expression of IL-7Rα on the cell surface. In addition, the soluble form of IL-7Rα may bind IL-7 in solution and may therefore act as a decoy receptor [25]. This may affect the IL-7 dependent thymic production of T cells, including the rate of regulatory T cell production that has been associated with T cell alloreactivity in HCT [26]. The biological significance of this in relation to HCT, however, deserves further investigation since IL-7 levels have been shown to be considerably elevated during the early phase after HCT [27]. Recently it was demonstrated that IL-7Rα Hap 2 (counting rs6897932T) is associated with faster CD4+ T cell reconstitution following antiretroviral therapy (ART) for HIV infection and that these individuals have lower circulating soluble IL7Rα [28].

Furthermore, the potential of sIL-7Rα to influence TSLP-signalling should be explored in future studies. TSLP is important for the development of regulatory T cells. A reduction in TSLP signalling could lead to reduced production of Tregs and thereby increased GvHD and TRM.

In conclusion, there is accumulating evidence for an association between various IL-7Rα SNPs and adverse outcome in HCT. In this study we show for the first time that the donor type of IL-7Rα rs6897932 may be associated with the risk of relapse in patients undergoing HCT for hematological malignancies. In addition, the functional impact we know of rs6897932 on the release of sIL-7Rα in health controls and a potential biological mechanism for the immune-modulating function of the SNP. These data provide further evidence of a role of the IL-7 pathway in outcome of HCT and impact of non-synonymous SNPs on IL-7Rα function.