Relationship between apurinic endonuclease 1 Asp148Glu polymorphism and gastrointestinal cancer risk: An updated meta-analysis

Abstract

AIM: To evaluate the relationship between apurinic endonuclease 1 (APE1) Asp148Glu polymorphism and the susceptibility to gastrointestinal (GI) cancers.

METHODS: We searched PubMed, ISI Web of Knowledge, and Chinese National Knowledge Infrastructure (CNKI) databases updated on July 15, 2014 for relevant studies. Only case-control studies comparing APE1 Asp148Glu polymorphism and GI cancer risk were included. We excluded studies reporting only standardized incidence ratios without control groups and those without detailed genotyping data. Meta-analysis was performed on 17 studies involving 4856 cancer patients and 6136 cancer-free controls. Review Manager version 5.1 was used to perform the meta-analysis. The pooled odds ratios (ORs) and 95% confidence intervals (CIs) were estimated under the allele contrast, homozygous, heterozygous, dominant and recessive genetic models. We also conducted subgroup analyses stratified by ethnicity and cancer type. Publication bias was evaluated using Begg’s test.

RESULTS: The meta-analysis showed a significant association between APE1 Asp148Glu polymorphism and GI cancer risk in three genetic models in the overall population (G vs T: OR = 1.18; 95%CI: 1.05-1.32; TG vs TT: OR = 1.28; 95%CI: 1.08-1.52; TG + GG vs TT: OR = 1.32; 95%CI: 1.10-1.57). Stratified analysis by ethnicity revealed a statistically increased GI cancer risk in Asians (G vs T: OR = 1.27; 95%CI: 1.07-1.51; GG vs TT: OR = 1.58; 95%CI: 1.05-2.38; TG vs TT: OR = 1.30; 95%CI, 1.01- 1.67; and TG + GG vs TT: OR = 1.38; 95%CI: 1.07-1.78), but not in Caucasians. Further subgroup analysis by cancer type indicated that APE1 Asp148Glu polymorphism may contribute to gastric cancer risk. However, Asp148Glu has no significant association with colorectal or esophageal cancer risk in any genetic model.

CONCLUSION: This meta-analysis suggests that the APE1 Asp148Glu polymorphism G allele is associated with an increased GI cancer risk, especially in gastric cancer.

Key Words: Apurinic endonuclease 1; Single nucleotide polymorphism; Gastrointestinal cancers; Cancer risk; Meta-analysis

Core tip: Apurinic endonuclease 1 (APE1) plays an important role in the DNA repair system and therefore has been implicated in human carcinogenesis. Many studies have suggested an association between the APE1 Asp148Glu polymorphism and gastrointestinal cancer susceptibility. However, the results remained inconclusive. We performed a meta-analysis on pooled data from previously published studies. The results showed that the APE1 Asp148Glu polymorphism G allele is associated with an increased gastrointestinal cancer risk, especially in gastric cancer.

INTRODUCTION

Gastrointestinal (GI) cancers, especially esophageal, gastric, and colorectal cancers, are the leading causes of cancer-related death worldwide[1]. GI cancers are multifactorial diseases caused by complex interactions between many genetic and environmental factors[1,2]. Allelic variations in oncogenes are candidate genetic risk factors that may alter the onset and outcome of GI cancers[3].

Apurinic/apyrimidinic endonuclease 1 (APE1) is an essential enzyme in the base excision repair pathway[4]. APE1 plays an important role in the DNA repair system and therefore has been implicated in human carcinogenesis[5]. The human APE1 is located on chromosome 14q11.2 and consists of five exons, spanning roughly 2.5 to 3 kb of DNA[6]. Many single nucleotide polymorphisms in the APE1 gene have been reported, including the commonly occurring Asp148Glu in the fifth exon and -141T/G in the promoter region[7]. These nonconservative amino acid alterations have been reported to reduce the DNA repair activity of APE1 and consequently increase cancer risk[8]. Our previous study has suggested that the APE1 -141T/G but not the Asp148Glu polymorphism may influence the susceptibility to and progression of breast cancer in the Chinese population[9]. However, a recent study reported that the APE1 148 GG genotype is associated with an increased risk of colorectal cancer (CRC)[10].

It is important to summarize inconclusive results from different studies to further validate the association of one polymorphism with cancer risk[11]. To clarify the role of the APE1 Asp148Glu polymorphism in GI cancer risk, we performed a meta-analysis on all eligible case-control studies to estimate the overall cancer risk associated with the APE1 Asp148Glu polymorphism. Furthermore, we conducted subgroup analyses stratified by ethnicity and cancer type.

MATERIALS AND METHODS

Methods

The procedures performed in this meta-analysis are in accordance with recent guidelines for the reporting of meta-analyses (PRISMA guidelines).

Publication search

We searched the electronic databases of PubMed, Web of Knowledge and Chinese National Knowledge Infrastructure databases to collect articles reporting case-control studies related to the association of APE1 Asp148Glu polymorphisms with GI cancer risk. The keywords used for search were as follows: apurinic/apyrimidinic endonuclease-1/APE1/APEX/HAP1/REF-1, gastrointestinal/esophageal/gastric/colorectal, cancer/carcinoma/tumor/neoplasm, polymorphism/genotype/SNP/variation. The latest search was updated on July 15, 2014. Furthermore, reference lists of main reports and review articles were also reviewed manually to identify additional relevant publications.

Selection criteria

The following criteria were used to select studies for further meta-analysis: (1) case-control studies; (2) studies that evaluated the associations between APE1 Asp148Glu polymorphism and GI cancer risk; (3) studies that contained at least two comparison groups (cancer vs control); and (4) studies that included detailed genotyping data.

Data extraction and synthesis

Articles were reviewed independently by two reviewers and data with discrepancies were discussed by all authors. For each included study, the following information was collected: first author, year of publication, country of origin, ethnicity, source of control, total numbers of cases and controls, genotyping methods as well as numbers of cases and controls with the different genotypes. Different ethnic groups were categorized as Caucasian, Asian, African, and “mixed”. All the case and control groups were well controlled. The non-cancer controls had no history of gynecologic disease, and there was no present evidence of any malignant disease.

Statistical analysis

The associations between APE1 Asp148Glu polymorphism and GI cancer risk were measured by odds ratio (OR) with 95% confidence interval (CI). The significance of the pooled OR was determined by the Z-test.

The meta-analysis assessed association by using five different genetic models: (1) allele contrast genetic model - A vs a (where ‘‘a’’ is the wild-type allele and ‘‘A’’ is the variant allele); (2) homozygous genetic model-comparison between the 2 homozygous genotypes (AA vs aa); (3) heterozygous genetic model-comparison between the heterozygous and homozygous wild-type genotype groups (Aa vs aa); (4) dominant genetic model-comparison between the wild-type homozygous genotype vs the variant allele-positive genotypes (AA + Aa vs aa); and (5) recessive genetic model-comparison between the variant homozygous genotype vs the rest (AA vs aa + Aa).

Statistical heterogeneity among studies was assessed with the Q and I² statistics. If the P-value of heterogeneity test was more than 0.1 (P≥ 0.1), the pooled OR estimate of the study was calculated using the fixed-effects model. Otherwise, the random-effects model was used[11]. The value of the I index was used to assess the degree of heterogeneity (I² < 25%: no heterogeneity; 25% < I² < 50%: moderate heterogeneity; 50% < I² < 75%: high heterogeneity; I² > 75%: extremely high heterogeneity). Publication bias was evaluated by the funnel plot and further assessed by the method of Egger’s linear regression test. All statistical analyses were carried out with the Review Manager version 5.1 (Revman; The Cochrane Collaboration, Oxford, United Kingdom).

RESULTS

Characteristics of the included studies

As shown in Figure 1, a total of 126 potential publications were initially extracted. After reading the abstracts, we excluded 67 irrelevant studies, 20 studies with insufficient data, and 11 duplicated studies. After reading the full-texts, we excluded 5 articles with no detailed genotyping data, 2 non-case-control studies, 2 laboratory studies and 3 review articles. Finally, 17 studies from 16 articles were included in this meta-analysis.

Figure 1 Flow diagram of study identification and selection.

Overall, 17 studies on APE1 Asp148Glu polymorphism and GI cancer risk were identified[10,12-26], including a total of 4856 cases and 6136 case-free controls. The characteristics of the included studies are listed in Table 1. Among the eligible 17 studies, nine were carried out in Caucasians from United States, Italy, Czech, Spain, Poland and Turkey, seven were based on Asian background and carried out in China and Japan, and one was based on mixed ethnic groups. All studies were case-controlled, including 11 CRC studies, 4 gastric cancer (GC) studies and 2 esophageal cancer (EC) studies. All GI cancers were confirmed by histology or pathology. The histological type of cancers in the included studies was adenocarcinoma except one EC study[26]. Moreover, controls were matched mainly by age. Eleven studies were population-based and six were hospital-based. Several genotyping methods were used in the studies, including polymerase chain reaction-restriction fragment length polymorphism, PCR-ligase detection reaction, TaqMan, MassARRAY, and Arrayed primer extension.

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

Ref.	Year	Country	Ethnicity	Cancer type	Genotyping method	Source of controls	Total sample size (cases/controls)
Zhang et al[10]	2014	China	Asian	CRC	PCR-CTPP	PB	247/300
Li et al[12]	2013	China	Asian	CRC	PCR-RFLP	HB	451/631
Canbay et al[13]	2011	Turkey	Caucasian	CRC	PCR-RFLP	PB	79/247
Gu et al[14]	2011	China	Asian	GC	PCR-RFLP	PB	572/547
Li et al[15]	2011	China	Asian	GC	PCR-RFLP	PB	126/156
Canbay et al[16]	2010	Turkey	Caucasian	GC	PCR-RFLP	PB	40/247
Brevik et al[17]	2010	United States	Caucasian	CRC	TaqMan	HB	304/359
Jelonek et al[18]	2010	Poland	Caucasian	CRC	PCR-RFLP	PB	153/273
Palli et al[19]	2010	Italy	Caucasian	GC	TaqMan	PB	314/548
Ye et al[20]	2010	China	Asian	CRC	MassARRAY	HB	123/158
Kasahara et al[21]	2008	Japan	Asian	CRC	PCR-RFLP	HB	68/121
Pardini et al[22]	2008	Czech	Caucasian	CRC	TaqMan	HB	532/532
Tse et al[23]	2008	United States	Caucasian	EC	TaqMan	HB	312/454
Berndt et al[24]	2007	United States	Caucasian	CRC	TaqMan	PB	720/725
Berndt et al[24]	2007	United States	Mixed	CRC	TaqMan	PB	47/48
Moreno et al[25]	2006	Spain	Caucasian	CRC	Arrayed primer extension	PB	359/312
Hao et al[26]	2004	China	Asian	EC	PCR-RFLP	PB	409/478

CRC: Colorectal cancer; GC: Gastric cancer; EC: Esophageal cancer; PCR-CTPP: Polymerase chain reaction with confronting two-pair primers; PCR-RFLP: Polymerase chain reaction-restriction fragment length polymorphism; PB: Population based; HB: Hospital based.

Meta-analysis results

As shown in Table 2, the frequency of the G allele varied widely across the 12 studies, ranging from 0.31 to 0.54. The average frequency of the G allele in the overall population, Caucasian population and Asian population was 0.47, 0.46, and 0.47, respectively. There was no significant difference between Asians and Caucasians (P > 0.05). The average frequency of the G allele in CRC, GC and EC was 0.46, 0.50, and 0.46, respectively.

Table 2 Apurinic endonuclease 1 Asp148Glu polymorphism genotype distribution and allele frequency in cases and controls.

Ref.	Genotype (n)								Allele frequency (n)				MAF
	Cases				Controls				Cases		Controls
	Total	TT	TG	GG	Total	TT	TG	GG	T	G	T	G
Zhang et al[10]	247	87	90	70	300	121	137	41	264	230	381	219	0.47
Li et al[12]	451	123	247	81	631	186	335	110	493	409	707	555	0.45
Canbay et al[13]	79	28	43	8	247	151	63	33	99	59	365	129	0.37
Gu et al[14]	338	69	185	84	362	110	183	69	323	353	403	321	0.52
Li et al[15]	126	26	64	36	156	56	70	30	116	136	182	130	0.54
Canbay et al[16]	40	14	18	8	247	151	63	33	46	34	365	129	0.42
Brevik et al[17]	304	102	137	65	359	108	167	84	341	267	383	335	0.44
Jelonek et al[18]	113	49	59	5	273	70	141	62	157	69	163	143	0.31
Palli et al[19]	298	103	147	48	546	208	243	95	353	243	659	433	0.41
Ye et al[20]	123	37	86	0	158	52	106	0	160	86	210	106	0.35
Kasahara et al[21]	68	23	45	0	121	70	51	0	91	45	191	51	0.33
Pardini et al[22]	531	140	261	130	530	157	267	106	541	521	581	479	0.49
Tse et al[23]	311	75	162	74	454	123	228	103	312	310	474	434	0.50
Berndt et al[24]	692	175	364	153	710	204	335	171	714	670	743	677	0.48
Berndt et al[24]	47	11	23	13	48	18	22	7	45	49	58	36	0.48
Moreno et al[25]	359	95	177	87	312	99	147	66	367	351	345	279	0.49
Hao et al[26]	409	126	211	72	477	149	243	95	463	355	541	433	0.43

MAF: Minor allele frequency.

Overall, there was evidence of an association between GI cancer risk and the variant genotypes when all the eligible studies were pooled into the meta-analysis. As show in Figure 2 and Table 3, there was a significant association between APE1 Asp148Glu polymorphism and GI cancer risk in three genetic models in the overall population (G vs T: OR = 1.18, 95%CI = 1.05-1.32, P = 0.004; TG vs TT: OR =1.28, 95%CI = 1.08-1.52, P = 0.004; TG + GG vs TT: OR = 1.32, 95%CI =1.10-1.57, P = 0.002). However, there was no significant association in the other two genetic models (GG vs TT: OR = 1.25, 95%CI = 0.98-1.59, P = 0.07; GG vs TT + TG: OR = 1.10, 95%CI = 0.90-1.34, P = 0.33).

Figure 2 Forest plot of association of apurinic endonuclease 1 Asp148Glu polymorphism with gastrointestinal cancer risk stratified by ethnicity (TG + GG vs TT).

Table 3 Meta-analysis results.

Comparison	OR	95%CI	P value	Heterogeneity		Effects model
				I2	P value
G vs T	1.18	1.05-1.32	0.0041	71%	< 0.00001	Random
Caucasian	1.09	0.94-1.26	0.24	72%	0.0004	Random
Asian	1.27	1.07-1.51	0.0071	70%	0.003	Random
Colorectal cancer	1.15	0.99-1.33	0.06	73%	< 0.0001	Random
Gastric cancer	1.41	1.09-1.83	0.0091	70%	0.02	Random
Esophageal cancer	1.01	0.88-1.16	0.87	0%	0.40	Fixed
GG vs TT	1.25	0.98-1.59	0.07	73%	< 0.00001	Random
Caucasian	1.04	0.77-1.41	0.81	72%	0.0003	Random
Asian	1.58	1.05-2.38	0.031	76%	0.002	Random
Colorectal cancer	1.12	0.79-1.59	0.51	79%	< 0.00001	Random
Gastric cancer	1.77	1.11-2.84	0.021	63%	0.04	Random
Esophageal cancer	1.02	0.77-1.35	0.90	0%	0.34	Fixed
TG vs TT	1.28	1.08-1.52	0.0041	68%	< 0.0001	Random
Caucasian	1.26	0.98-1.63	0.07	76%	<0.0001	Random
Asian	1.30	1.01-1.67	0.041	63%	0.01	Random
Colorectal cancer	1.23	0.98-1.54	0.08	73%	< 0.0001	Random
Gastric cancer	1.66	1.20-2.31	0.0021	51%	< 0.0001	Random
Esophageal cancer	1.04	0.83-1.31	0.72	0%	0.41	Fixed
GG+TG vs TT	1.32	1.10-1.57	0.0021	73%	< 0.00001	Random
Caucasian	1.24	0.95-1.61	0.11	80%	< 0.00001	Random
Asian	1.38	1.07-1.78	0.011	66%	0.007	Random
Colorectal cancer	1.23	0.98-1.55	0.08	75%	< 0.0001	Random
Gastric cancer	1.77	1.40-2.24	< 0.00011	19%	0.29	Fixed
Esophageal cancer	1.02	0.81-1.29	0.84	9%	0.29	Fixed
GG vs TT+TG	1.10	0.90-1.34	0.33	70%	< 0.0001	Random
Caucasian	0.95	0.75-1.20	0.66	64%	0.004	Random
Asian	1.36	0.95-1.95	0.10	77%	0.002	Random
Colorectal cancer	1.05	0.77-1.43	0.76	79%	< 0.00001	Random
Gastric cancer	1.25	0.99-1.56	0.06	33%	0.21	Fixed
Esophageal cancer	0.96	0.75-1.21	0.71	0%	0.38	Fixed

¹Represent a significant association between APE1 Asp148Glu polymorphism and GI cancer risk in three genetic models in the overall population.

In the stratified analysis by population, as shown in Figure 2 and Table 3, meta-analysis showed that Asp148Glu polymorphism had no association with GI cancer risk in Caucasians in all genetic models (allele contrast genetic model: OR = 1.09, 95%CI = 0.94-1.26, P = 0.24; homozygote comparison: OR = 1.25, 95%CI = 0.98-1.59, P = 0.81; heterozygote comparison: OR = 1.26, 95%CI = 0.98- 1.63, P = 0.07; the dominant model: OR = 1.24, 95%CI = 0.95-1.61, P = 0.11; and the recessive model: OR = 0.95, 95%CI = 0.75-1.20, P = 0.66).

There were 7 studies with 1587 cases and 1913 controls for assessing the relationship between Asp148Glu polymorphism and GI cancer susceptibility in Asians. As shown in Figure 2 and Table 3, Asp148Glu polymorphism was significantly associated with GI cancer risk in four genetic models (allele contrast genetic model: OR = 1.27, 95%CI = 1.07-1.51, P = 0.007; homozygote comparison: OR = 1.58, 95%CI = 1.05-2038, P = 0.03; heterozygote comparison: OR = 1.30, 95%CI = 1.01-1.67, P = 0.04; and the dominant model: OR = 1.38, 95%CI = 1.07-1.78, P = 0.01.

In the stratified analysis by cancer type, 11 studies including 3083 cases and 3706 controls were used to evaluate the relationship between APE1 Asp148Glu polymorphism and CRC risk. As shown in Table 3 and Figure 3, there was no significant association between APE1 Asp148Glu polymorphism and CRC risk under all genetic models (allele contrast genetic model: OR = 1.15, 95%CI = 0.99-1.33, P = 0.06; homozygote comparison: OR = 1.12, 95%CI = 0.79-1.59, P = 0.51; heterozygote comparison: OR = 1.23, 95%CI = 0.98-1.54, P = 0.08; the dominant model: OR = 1.23, 95%CI = 0.98-1.55, P = 0.08; and the recessive model: OR = 1.05, 95%CI = 0.77-1.43, P = 0.76).

Figure 3 Forest plot of association of apurinic endonuclease 1 Asp148Glu polymorphism with GI cancer risk stratified by cancer type (TG + GG vs TT).

There were four studies including 1052 cases and 1498 controls used to evaluate the relationship between APE1 Asp148Glu polymorphism and GC risk. As shown in Table 3 and Figure 3, Asp148Glu polymorphism was significantly associated with an increased risk of GC under all genetic models (allele contrast genetic model: OR = 1.41, 95%CI = 1.09-1.83, P = 0.009; homozygote comparison: OR = 1.77, 95%CI = 1.11-2.84, P = 0.02; heterozygote comparison: OR = 1.66, 95%CI = 1.20-2.31, P = 0.002; and the dominant model: OR = 1.77, 95%CI = 1.40-2.24, P < 0.0001).

There were only two EC studies included in this meta-analysis. As shown in Table 3 and Figure 3, no significant association was detected between Asp148Glu polymorphism and EC risk.

Sensitivity analysis and publication bias

The influence of any single study on the overall estimate was analyzed by excluding one study at a time. No significant difference was observed when any of the studies was excluded. Therefore, our results were statistically reliable.

Funnel plot and Egger’s test were performed to evaluate the publication bias. As shown in Figure 4, the funnel plots failed to detect any obvious asymmetry in all genotypes in the overall population, and the Egger’s test revealed no publication bias (P > 0.05). Therefore, no significant publication bias was found in this meta-analysis.

Figure 4 Funnel plot for publication bias.

DISCUSSION

The present meta-analysis, including 4856 cases and 6136 case-free controls from 17 case-control studies, was conducted to evaluate the association between APE1 Asp148Glu polymorphism and GI cancer risk. Our results indicated that the variant genotypes were associated with an increased risk of GI cancers, especially GC.

APE1 is a key multifunctional gene involved in the base excision repair pathway. It was reported that APE1 is associated with aggressive tumor biology and has an impact on survival of GC patients[27]. The Asp148Glu polymorphism is a common non-synonymous APE1 coding region variant. Previous studies on the association between the APE1 Asp148Glu polymorphism and GI cancer risk have shown controversial results[12-26].

In a previous meta-analysis, Gu et al[28] suggested that the APE1 Asp148Glu polymorphism may contribute to genetic susceptibility to cancers, especially CRC. However, a recent meta-analysis by Shen et al[29] failed to detect an association between APE1 Asp148Glu polymorphism and CRC risk. We found several worthwhile queries in Shen’s study. First, the ethnicity of Canbay’s study was identified as Asian in Shen’s meta-analysis[29]. While in the original article, the cases and controls were both based on Caucasians but not Asians[13]. Second, the study by Berndt et al[24] was carried out in Caucasians and a mixed non-Caucasian population, and therefore should probably be divided into two studies. Third, non-English literature database such as the CNKI should also be considered for the search of eligible case-control studies. We have found one eligible study published in Chinese and included it in this study[20].

In our study, 17 case-control studies were included. There was a significant association between APE1 Asp148Glu polymorphism and GI cancers risk in four genetic models in the overall population. Stratified analysis by ethnicity revealed that there was a statistically increased GI cancers risk in Asians. Further subgroup analysis by cancer type indicated that APE1 Asp148Glu polymorphism may contribute to GC risk.

There are some limitations of this meta-analysis that should be noted. First, this meta-analysis was based on pooled data while no individual data were available; thus, we could not assess the risk of cancer based on environmental factors, age, and other risk factors for GI cancers. Second, the small study effect, where the effects reported in small studies are larger, could not be avoided in some studies of a relative small size (< 200). Third, there was no significant association between Asp148Glu polymorphism and EC risk in this meta-analysis. Since only two EC studies with different pathological types[23,26] were included, this negative finding may result from a lack of statistical power. Larger scale multicenter studies are warranted to further validate the association between APE1 Asp148Glu polymorphism and GI cancer risk.

In conclusion, our present meta-analysis provides evidence for the association between the APE1 Asp148Glu polymorphism and GI cancer risk. Results suggest that the APE1 Asp148Glu G allele was associated with an increased GI cancer risk among Asian subjects. Furthermore, the APE1 Asp148Glu polymorphism was associated with an increased risk GC. Further large-population based studies are needed to confirm the association between APE1 Asp148Glu polymorphism and EC risk.

COMMENTS

Background

Epidemiological studies have suggested that Asp148Glu polymorphism in the apurinic endonuclease 1 (APE1) gene is associated with gastrointestinal (GI) cancer risk. However, the results are still controversial.

Research frontiers

APE1 plays an important role in the DNA repair system and therefore has been implicated in human carcinogenesis. Epidemiologic studies suggested that single nucleotide polymorphisms (SNP) in APE1 may confer individuals’ susceptibility to cancer. Recently, numerous studies have evaluated the association between APE Asp148Glu polymorphism and cancer risk. However, the results remain inconclusive.

Innovations and breakthroughs

The present meta-analysis was performed on all eligible case-control studies to estimate the association of the APE1 Asp148Glu polymorphism with GI cancer risk.

Applications

The present meta-analysis showed that the APE1 Asp148Glu G allele was associated with an increased GI cancer risk among Asian subjects. Furthermore, APE1 Asp148Glu polymorphism was associated with an increased risk of GC.

Terminology

SNP is DNA sequence variation occurring when a single nucleotide in the genome differs between members of a biological species or paired chromosomes. SNP in some genes may cause an increase or decrease in risk for some certain diseases.

Peer-review

This meta-analysis showed that the G allele of APE1 Asp148Glu polymorphism was associated with a higher gastrointestinal tract cancer risk. However, larger scale studies are warranted to further validate the association between this polymorphism and GI cancer risk.