Interleukin-10-1082 G/a polymorphism and acute renal graft rejection: a meta-analysis

Abstract

Objective: The aim of this study was to investigate the association between interleukin (IL)-10-1082 (G/A) promoter polymorphism and acute rejection (AR) in renal transplant recipients. Methods: We searched MEDLINE, EMBASE, Web of Science, and Cochrane Central Register from the inception to March 2015 for relevant studies. Data concerning publication information, population characteristics, and transplant information were extracted. Odds ratios (ORs) was calculated for the association between IL-10-1082 GG genotype (or IL-10-1082 G allele) and AR risk. Results: This meta-analysis included 22 case–control studies including 2779 cases of renal transplant recipients. The pooled estimate showed that the IL-10-1082 GG genotype was not significantly associated with AR risk (OR_random=1.07, 95% CI 0.80–1.43, p = 0.64). Similarly, the pooled estimate showed that the IL-10-1082 G allele was not significantly associated with AR risk (OR_fixed=1.02, 95% CI 0.90–1.16, p = 0.74). None of subgroup analyses yielded significant results in the association between IL-10-1082 GG genotype (or IL-10-1082 G allele) and AR risk. Meta-regression confirmed that there was no significant correlation between the pre-selected trial characteristics and our study results. Conclusions: This meta-analysis suggests that IL-10-1082 G/A polymorphism is not significantly associated with AR risk in renal transplant recipients.

• Renal transplantation
• acute rejection
• interleukin-10
• gene polymorphism
• meta-analysis

Introduction

Kidney transplantation is often the best medical treatment for patients with end-stage renal disease. Despite recent improvement in immunosuppressive medication and HLA-matching, acute rejection (AR) continues to be problematic and remains the leading cause of graft loss after renal transplantation.1 AR is mainly caused by T-cell-mediated immune responses. It has been suggested that graft microenvironment and cytokines play an important role in the precipitation of rejection.2 In the last decade, several single-nucleotide polymorphisms (SNPs) in the genes encoding for cytokines have been identified.3,4 The production of cytokines is under genetic control and varies among individuals as a function of polymorphisms within the regulatory regions of the various genes that determine the transcriptional activation.5 Although there is evidence that cytokine gene polymorphisms are associated with varying quantities of cytokine protein production, the data on the correlation between these gene polymorphisms and the risk of AR remains unclear.6,7

Cytokine interleukin-10 (IL-10) is thought to play a critical role in allograft tolerance, predominantly through its ability to suppress inflammation. IL-10 is encoded by IL-10 gene located on chromosome 1, while its promoter is a highly polymorphic region.8 The promoter of the IL-10 gene contains three biallelic polymorphisms at positions -1082 (base G to A, dbSNP no. rs1800896), -819 (base C to T, dbSNP no. rs1800871), and -592 (base C to A, dbSNP no. rs1800872) from the transcription start site, and these influence the capacity of cells to produce IL-10.5 In 2008, a review by Nickerson9 examines the accumulated literature implicating various cytokine (including IL-10) SNPs with renal or liver allograft outcomes, but it did not compute a pooled estimate results. Thakkinstian et al.10 published findings from a meta-analysis suggesting possible association between a 3-SNP-haplotype of IL-10 and poor outcomes including chronic allograft nephropathy (CAN), graft rejection (GR), and graft failure (GF) in renal transplantation, but this needs to be confirmed in larger studies.

Over the last two decades, although a number of studies11–32 have shown varying results in the association between IL-10–1082 (G/A) promoter polymorphism and AR risk in renal transplant recipients, the results are inconsistent and inconclusive. Moreover, the conclusion of a single study is limited by many factors such as the small size, ethnicity, age, gender, donor type, immunosuppressive regimens, and Hardy–Weinberg equilibrium (HWE) deviations. In order to derive a more comprehensive estimation of the association between IL-10–1082 polymorphism and AR risk in renal transplant recipients, we conducted a meta-analysis to evaluate the association.

Methods

Literature search strategy

Two reviewers (Q.W. Hu and P. Liao) independently searched the MEDLINE, EMBASE, Web of Science, and Cochrane Central Register up to 2 March 2015. To maximize the data search, the search strategy used relevant keywords and medical subject heading (MeSH) terms including “interleukin-10” or “IL-10” or “cytokine gene polymorphism”; “kidney transplant” or “renal transplant” or “allograft nephropathy”; “acute rejection” or “early graft rejection”. We also hand searched the reference lists of retrieved articles, case reports for relevant articles. The search was performed with restriction on English, and conducted on human subjects.

Inclusion and exclusion criteria

Two reviewers independently screened the titles and abstracts of studies to identify those that fulfilled the inclusion criteria: (1) evaluation of the association between IL-10–1082 G/A polymorphism and AR risk in renal transplant recipients and (2) IL-10–1082 GG and GA/AA genotype data (or IL-10–1082 G and A allele data) in AR and non-AR transplant recipients were respectively provided. Major reasons for exclusion of studies were: (1) an abstract, comment, review, or editorial and (2) no sufficient data were reported.

Data extraction and quality assessment

Three reviewers (Q.W. Hu, P. Liao, and H. Tian) independently performed the data extraction that met the inclusion criteria. The titles and abstracts were scanned to exclude any clearly irrelevant studies. The full texts of the remaining articles were read to determine whether they contained information on the useful topic. Any disagreement was resolved by discussion to reach a consensus among the investigators. The following data were collected from each study: (1) publication data: first author, year of publication, and country of origin; (2) characteristics of the study population: ethnicity, age, gender, IL-10–1082 GG and GA/AA genotype data in AR and non-AR transplant recipients, and IL-10–1082 G and A allele data in AR and non-AR transplant recipients; and (3) transplant information: donor type, AR definition, and HWE deviations.

The methodological quality of the included studies was assessed by three independent investigators using the Newcastle–Ottawa Quality Assessment Scale (NOS)33 for observational trials, respectively. The NOS consists of eight questions with nine possible points. A points-scoring system was used to judge the data according to the selected populations, the comparability of the groups and exposure/outcome of interest (0–5: poor quality, 6–9: good quality).

Statistical analysis

Statistical analyses were performed using RevMan 5.0 (The Cochrane Collaboration, Copenhagen, Denmark) and STATA software 12.0 (StataCorp, College Station, TX). The outcome measure was biopsy-proven AR during the early post-transplantation period. There were two main outcome measurements in this meta-analysis: (1) odds ratios (OR) and 95% confidence interval (CI) of individual studies and pooled data for the association between IL-10–1082 GG genotype and AR risk; (2) ORs and 95% CI of individual studies and pooled data for the association between IL-10–1082 G allele and AR risk. We assessed for the heterogeneity between studies by calculating a Cochran Q statistic and an I² statistic.34 If significant heterogeneity (I² > 50% or p < 0.05) was observed between studies, a random-effects model was used to pool the data; otherwise, a fixed-effects model was used. Sensitivity analysis and subgroup analysis for the primary outcome measurement were performed in a predefined manner to assess the impact of the following: (1) ethnicity, (2) region of nations, (3) donor type, (4) AR definition, and (5) HWE deviations. To explore the impact of pre-selected covariates on the association between IL-10–1082 GG genotype (or IL-10–1082 G allele) and AR risk, we performed a random-effects (or fixed-effects) meta-regression analysis. The logarithm of OR for IL-10–1082 GG genotype (or IL-10–1082 G allele), weighted by the inverse variance of each study, was regressed against age, male ratio, ethnicity, region of nations, donor type, AR definition, and HWE deviations. Begg’s rank correlation method34 and Egger’s weighted regression method35 were used to statistically assess publication bias (p < 0.05 was considered statistically significant). All p values were 2-tailed, and a p value < 0.05 was considered statistically significant.

Results

Studies selection

Figure 1 is a flow diagram of the selection of eligible studies. Two hundred twenty records were identified by the primary literature search. However, after screening the titles and abstracts, 178 irrelevant studies were excluded. We therefore retrieved 42 potentially relevant articles for a detailed review. Of these studies, 14 studies having no sufficient data, three studies reporting donor IL-10 gene polymorphism, two reviewed studies, and one study having duplicate data were further excluded. Consequently, the remaining 22 studies were included for this meta-analysis (Figure 1).

Figure 1. Flow diagram of trial selection process.

Study characteristics

All of the 22 included studies, a total of 2779 renal transplant patients were grouped for case–control design. One study27 is divided into Caucasian group and African-American group, so there were 23 sets of data altogether. The baseline characteristics of the included studies are summarized as follows: first author; year of publication (range, 2000–2014); country of origin (10 countries in Europe, 6 countries in Asia, 5 countries in America, 1 country in Africa); ethnicity (Caucasian or African-American); average age (range, 25.6–50.8 years); male ratio (range, 47–84%); number of IL-10–1082 GG and GA/AA genotype in AR and non-AR transplant recipients; number of IL-10–1082 G and A allele in AR and non-AR transplant recipients (data in two studies14,20 were not informed); donor type (living, deceased, living or deceased); AR definition (biopsy except two studies14,22); and HWE deviations (HWE deviations were studied using appropriate Chi-square tests and p-values < 0.05 were considered significant) (Table 1). Average NOS scores of the included 22 studies were 6.36 (range, 5–8 points), and only one study23 scored below 6 points was rated as poor quality study (Supplementary Table available online at http://informahealthcare.com/doi/suppl/10.3109/0886022X.2015.1106770).

Table 1. Characteristics of the studies included in the meta-analysis.

					AR		Non-AR		AR		Non-AR
Author (Year), Ref.	Country	Ethnicity	Age (years) AR/non-AR	Male (%)	GG	GA + AA	GG	GA + AA	G	A	G	A	Donor type	AR definition	HWE
Chen (2014)11	Netherlands	Caucasian	50.8/48.4	61.8	14	36	63	212	53	47	246	304	Living/deceased	Biopsy	No
Gaafar (2014)12	Saudi Arabia	NI	34/40.5	60.9	5	12	14	61	20	14	63	87	Living	Clinical and biopsy	No
Dhaouadi (2013)13	Tunisia	NI	NI	62.3	16	64	29	122	67	93	130	172	Living/deceased	Histological findings	NI
Mandegary (2013)14	Iran	NI	32/41	62	4	7	42	47	NI				Living	Serum creatinine	NI
Seyhun (2012)15	Turkey	NI	31.2/31.3	61.1	2	17	8	63	12	26	44	98	Living	Clinical and biopsy	NI
Kocierz (2011)16	Poland	Caucasian	40/40	62.3	41	102	2	54	135	151	53	69	Deceased	Clinical and biopsy	No
Azarpira (2009)17	Iran	NI	32.8/31.1	47	2	44	0	54	45	47	50	58	Living/deceased	Clinical or biopsy	Yes
Grinyó (2008)18	Spain	Caucasian	NI	63.6	13	99	33	75	51	71	146	172	Living/deceased	Biopsy	No
Manchanda (2008)19	India	NI	34.8/34.9	84	9	9	41	41	20	16	98	66	Living	Biopsy	NI
Breulmann (2007)20	Germany	NI	NI	59.2	31	83	20	90	NI				Living/deceased	Biopsy	No
Canossi (2007)21	Italy	NI	NI	NI	8	17	12	49	27	23	46	76	Deceased	Biopsy	No
Tajik (2006)22	Iran	NI	NI	69	3	8	4	27	11	11	27	35	Living	Clinical	NI
Gendzekhadze (2006)23	Venezuela	NI	40/40	58.7	2	28	1	32	12	48	16	50	Living/deceased	Serum creatinine  or biopsy	No
Dmitrienko (2005)24	Canada	Caucasian	45/43	57	14	36	19	31	48	52	58	42	Living/deceased	Biopsy	No
Ligeiro (2004)25	Portugal	NI	NI	NI	4	27	5	30	26	36	29	41	Deceased	Biopsy	NI
Plothow (2003)26	Brazil	Caucasian	34.5/34.5	53.8	4	35	3	23	26	52	17	35	Living/deceased	Biopsy	No
Hutchings 1 (2002)27	USA	Caucasian	23/23	76	2	22	4	22	19	29	21	31	Living/deceased	Biopsy	NI
Hutchings 2 (2002)27	USA	African–American	25.6/25.6	70.6	2	15	4	30	18	16	21	47	Living/deceased	Biopsy	NI
Cartwright (2001)28	UK	NI	NI	NI	9	16	8	16	30	20	28	20	Deceased	Biopsy	NI
George (2001)29	UK	NI	NI	NI	5	29	18	53	23	45	73	69	Living/deceased	Serum creatinine  and biopsy	NI
Marshall (2000)30	UK	NI	NI	NI	43	71	33	62	130	98	111	79	Deceased	Serum creatinine and biopsy	NI
Hahn (2000)31	USA	Caucasian	36.7/36.7	65	6	26	24	57	26	38	76	86	Living/deceased	Biopsy	NI
Pelletier (2000)32	UK	Caucasian/  African-American	45.9/45.9	62.4	9	23	12	55	32	32	53	81	Living/deceased	Biopsy	No

Notes: NI, not informed; AR, acute rejection; HWE, Hardy–Weinberg equilibrium.

Meta-analysis

The outcome measurements of this meta-analysis are summarized in Figures 2 and 3. Twenty-two studies reported IL-10–1082 GG genotype were analyzed for OR. There was significant heterogeneity among the studies (p_{heterogeneity}=0.03, I² = 40%); thus, a random-effects model was used to pool the data. The pooled estimate showed that the IL-10–1082 GG genotype was not significantly associated with AR risk (OR_random=1.07, 95% CI 0.80–1.43, p = 0.64) (Figure 2). Twenty studies reported IL-10-1082 G allele were analyzed for OR. There was no significant heterogeneity among the studies (p_{heterogeneity}=0.18, I² = 22%); thus, a fixed-effects model was used to pool the data. The pooled estimate showed that the IL-10-1082 G allele was not significantly associated with the AR risk (OR_fixed=1.02, 95% CI 0.90–1.16, p = 0.74) (Figure 3).

Figure 2. Odds ratios and 95% CI of individual studies and pooled data for the association between interleukin-10–1082 GG genotype and acute rejection risk in renal transplant recipients.

Figure 3. Odds ratios and 95% CI of individual studies and pooled data for the association between interleukin-10–1082 G allele and acute rejection risk in renal transplant recipients.

When stratified for ethnicity, no significant association was observed among either Caucasian group or non-Caucasian group (for IL-10-1082 GG genotype: Caucasian group, OR_random=0.88, 95% CI 0.41–1.89, p = 0.74; non-Caucasian group, OR_random=1.23, 95% CI 0.96–1.59, p = 0.11 and for IL-10-1082 G allele: Caucasian group, OR_fixed=0.95, 95% CI 0.78–1.16, p = 0.65; non-Caucasian group, OR_fixed=1.07, 95% CI 0.91–1.26, p = 0.42). None of the other subgroup (region of nations, donor type, AR definition, and HWE deviations) analyses yielded significant results in the association between IL-10-1082 GG genotype (or IL-10–1082 G allele) and AR risk (Table 2). Meta-regression confirmed that there was no significant correlation between the pre-selected trial characteristics (age, male ratio, ethnicity, region of nations, donor type, AR definition, and HWE deviations) and our study results.

Table 2. Subgroup analyses for the primary outcome measurement.

Outcome	Subgroup	No. of studies	OR (95% CI)	p	I^2 (%)
Association between IL-10-1082  GG genotype and AR risk	Caucasian	7	0.88 (0.41–1.89)	0.74	74.6
	Non-Caucasian	16	1.23 (0.96–1.59)	0.11	0
	Europe	10	1.21 (0.73–2.00)	0.46	68.6
	America	6	0.68 (0.40–1.16)	0.16	0
	Asia	6	1.23 (0.70–2.18)	0.47	0
	Africa	1	1.05 (0.53–2.08)	0.89	0
	Living donor	5	1.16 (0.65–2.08)	0.61	0
	Deceased donor	5	1.71 (0.82–3.58)	0.15	58.2
	Living or deceased donor	13	0.88 (0.60–1.30)	0.53	43.9
	Biopsy	21	1.07 (0.79–1.45)	0.66	43
	Non-biopsy	2	1.14 (0.30–4.31)	0.85	37.6
	No deviations from HWE	10	1.34 (0.76–2.36)	0.32	69.3
	Deviations from HWE	1	6.12 (0.29–130.87)	0.25	0
	Not informed	12	0.93 (0.69–1.25)	0.63	0
	Association between IL-10-1082  G allele and AR risk	Caucasian	7	0.95 (0.78–1.16)	0.65	0
Non-Caucasian		14	1.07 (0.91–1.26)	0.42	33.2
Europe		9	1.04 (0.87–1.23)	0.69	43.7
America		6	0.92 (0.69–1.22)	0.54	31.1
Asia		5	1.17 (0.85–1.62)	0.33	0
Africa		1	0.95 (0.65–1.40)	0.81	0
Living donor		4	1.21 (0.81–1.79)	0.35	0
Deceased donor		5	1.07 (0.85–1.36)	0.56	0
Living or deceased donor		12	0.97 (0.83–1.14)	0.73	39.6
Biopsy		20	1.02 (0.90–1.15)	0.79	25.4
Non-biopsy		1	1.30 (0.49–3.44)	0.6	0
No deviations from HWE		9	1.11 (0.93–1.33)	0.26	37.5
Deviations from HWE		1	1.11 (0.64–1.94)	0.71	0
Not informed		11	0.93 (0.78–1.12)	0.44	9.5

Publication bias

Publication bias was examined by evaluating funnel plots for studies of AR in renal transplant recipients. Egger and Begg tests showed that no potential publication bias existed among the included studies for the two end-points analyzed: Egger test p = 0.483, Begg test p = 0.712 for IL-10-1082 GG genotype (Figure 4A); Egger test p = 0.397, Begg test p = 0.194 for IL-10-1082 G allele (Figure 4B).

Figure 4. Funnel plot showing publication bias for the primary study ends of the association between interleukin-10–1082 GG genotype and acute rejection risk (A) and the association between interleukin-10–1082 G allele and acute rejection risk (B). Note: OR, odds ratios.

Discussion

The balance between pro- and anti-inflammatory cytokines determines the inflammatory response which mediates renal allograft outcome.36 In the context of allograft rejection, IL-10 as an anti-inflammatory cytokine can inhibit the production of tumor necrosis factor-α, IL-1, IL-6, IL-8, and IL-12 in monocytes/macrophages and interferon-γ in T cells.37 Some studies have suggested that IL-10 mRNA levels are increased just before a rejection episode.38 Suarez et al. showed that allelic variants at position -1082 were the most important genetic factor in the regulation of the constitutive mRNA levels of IL-10.39 Our study is the first meta-analysis ever published on this topic assessing the association between IL-10–1082 (G/A) promoter polymorphism and AR risk in renal transplant recipients. There were a total of 2779 renal transplant patients in 22 case–control studies enrolled in this meta-analysis, but it was unable to demonstrate any statistically significant association.

Most of the included studies revealed no significant correlation between IL-10–1082 (G/A) promoter polymorphism and AR risk in the recipients except for 3 articles.21,27,29 Canossi et al.21 reported that more incidence of AR episodes (ARE)-positive patients (35.8% vs. 18.2%) in the high producers group (GCC/GCC,GCC/ACC,GCC/ATA) was observed compared with the low producers. Hutchings et al.27 found African-American patients who were IL-10 A homozygous had a significantly lower probability of experiencing one or more AREs (p = 0.006), although there was no significant relationship in the Caucasian patients. Nevertheless, George et al.29 draw a different conclusion that the frequency of IL-10–1082 AA, a genotype associated with low production of IL-10, was high in the rejection group when compared to the rejection-free group (p = 0.01). Moreover, there were only two studies13,32 showed some correlation tendency between IL-10–1082 (G/A) polymorphism and AR risk. Dhaouadi et al.13 suggested that there was a greater prevalence of ARE among kidney transplant recipients carrying the IL-10–1082 GG genotype than those with other levels, albeit that the differences were no longer significant. Pelletier et al.32 found an increase in AR incidence as the IL-10 production phenotype increased (low, intermediate, high).

Lack of statistical power to detect small or moderate gene effects is a major limitation for many gene-association studies. Meta-analyses based on published data might be helpful to overcome this problem. However, some key factors may influence the results of meta-analyses, such as the AR definition. AR was defined as “biopsy proven AR” in most of the 22 included studies except for two studies14,22 in which AR was defined as “increased serum creatinine” or “clinical diagnostic criteria”. However, subgroup analyses for AR definition did not yield any significant results. In addition, the occurrence time of AR was also defined inconsistently. Among the included studies, eight studies14,15,18,20,21,27,28,32 did not refer to the occurrence time of AR, other studies defined it varies from one month to one year after transplantation. These different definitions and standards might make our result bias.

Another key factor “ethnicity” was also a potential source of bias. Although the article of Pallet et al.40 demonstrated that ethnic origin did not affect outcome after renal transplantation, previous research had shown that ethnicity can strongly influence the distribution of cytokine gene polymorphisms.41 For example, Hutchings et al.27 performed a study to determine the distribution of IL-10–1082 (G/A) promoter polymorphism in African-American and Caucasian renal transplant recipients, then he found that there was a significant link between IL-10 genotype and AREs, but only in African-American patients (p < 0.01). Among our included studies, six studies11,16,18,24,26,31 were in research of Caucasian recipients, two studies27,32 were in research of both Caucasian and African-American recipients, and the others did not refer to ethnicity. Nevertheless, when stratifying for ethnicity, no significant association was observed.

Different donor types were other factors affecting the results of this analysis. Among the 22 studies, living kidney donor was researched in five studies,12,14,15,19,22 deceased kidney donor was researched in five studies,16,21,25,28,30 and living or deceased kidney donor was researched in the remaining case studies. But subgroup analyses for donor type did not yield significant results. Moreover, in the 22 case–control studies, although all of the researches took an immunosuppressive regimen in patients after renal transplantation, differences in immunosuppressive regimens, such as type of induction therapy, the scheme of maintenance therapy and dosage of immunosuppressants, were also a potential source of bias. Other important factors for the association between IL-10–1082 polymorphism and AR risk in renal transplant recipients included age, gender, region of nations, and HWE deviations. Nevertheless, none of the subgroup analyses and meta-regression analyses for these factors yielded any significant results.

It is necessary to mention the relatively limited power of our meta-analysis to find a different result. For example, the country zones of the 22 studies were unevenly distributed, in which, 10 researches11,16,18,20,21,25,28–30,32 were from Europe, six researches12,14,15,17,19,22 were from Asia, 5 researches23,24,26,27,31 were from America, and only one research13 was from Africa. Another drawback that might influence our study results is the insufficient information among the included studies, encompassing ethnicity, age, gender, HWE deviations, etc. In summary, this meta-analysis suggests that the IL-10-1082 (G/A) promoter polymorphism is not significantly associated with AR risk in renal transplant recipients. In the future, larger and well-designed multicenter case–control studies should be performed to reevaluate the association.

Supplementary material available online

Acknowledgments

The authors thank all the participants in the study.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.