Sep 30, 2017 - the INK4 locus (ANRIL) that locates within the. CDKN2A-CDKN2B gene cluster [9-11]. Through immunologic technology, we can detect differ-.
Int J Clin Exp Med 2017;10(9):12986-12994 www.ijcem.com /ISSN:1940-5901/IJCEM0055878
Review Article Association between CDKN2B-AS gene rs4977574 polymorphism and susceptibility to coronary heart disease: evidence from 124,752 subjects Yufeng Jiang*, Min Chen*, Nannan Zhang, Huajia Yang, Yafeng Zhou Department of Cardiology, The First Affiliated Hospital of Soochow University, Suzhou City, Jiangsu Province, P. R. China. *Co-first authors. Received April 20, 2017; Accepted September 3, 2017; Epub September 15, 2017; Published September 30, 2017 Abstract: Objectives: To investigate the association between the CDKN2B-AS rs4977574 polymorphism and coronary heart disease. Methods: We performed a computerized comprehensive search on PubMed, Embase, OVID, Web of Science, CNKI, and Wan Fang databases up to December of 2016. Two of the authors individually extracted study data and assessed the study quality using NOS scale. Odds ratios (ORs) and 95% confidence intervals (CIs) were combined for evaluation using a random effect model or fixed effect model according to study heterogeneity. Results: There were totally twenty case-control studies of 51,522 patients and 73,230 healthy controls included. Significant associations were found between CDKN2B-AS rs4977574 polymorphism and CHD in overall population (OR = 1.26, 95% CI = 1.21-1.32, p 50% indicating heterogeneity among studies. Otherwise the fixed effect model should be used. Sensitivity analysis was
Int J Clin Exp Med 2017;10(9):12986-12994
CDKN2B-AS polymorphism and CHD Table 1. Characteristics of the studies included for meta-analysis Source of control
Genotyping Pa- Risk Cases Controls method tients allele
Study design
NOS
HWE (Y/N)
G
Case-control
8
Y
G
Case-control
7
Y
MI
G
Case-control
8
Y
Author
Year
Country
Ethnicity
Cao [18]
2016
China
Asian
Hospital-based
PCR-RFLP
565
541
CHD
Asian
Hospital-based
TaqMan
250
252
CHD
TaqMan
1560
1751
AbdulAzeez [19] 2016 Saudi Arabia Zheng [20]
2016
US
Caucasian Population-based
Tang [21]
2016
China
Asian
Hospital-based
TaqMan
289
338
CHD
G
Case-control
8
Y
Matsuoka [22]
2015
Japan
Asian
Hospital-based
RT-PCR
1822
2284
MI
G
Case-control
8
Y
Hindy [23]
2014
Sweden
TaqMan
3164
20785
CHD
G
Case-control
8
Y
Wang [24]
2014
China
Asian
Population-based
TaqMan
2365
2678
MI
G
Case-control
8
Y
Shanker [25]
2014
India
Asian
Population-based
TaqMan
1034
1034
CHD
G
Case-control
8
Y
Huang [26]
2014
China
Asian
Hospital-based
RT-PCR
590
482
CHD
G
Case-control
8
Y
Lee [29]
2013
Korea
Asian
Population-based
Affymetrix
2293
4302
CHD
G
Case-control
8
Y
Tragante [30]
2013 Netherlands Caucasian Population-based
TaqMan
3788
2015
CHD
G
Case-control
8
Y
Caucasian Population-based
Lin [31]
2012
China
Asian
Hospital-based
PCR-RFLP
142
192
MI
G
Case-control
7
Y
Qi [32]
2011
US
Caucasian
Hospital-based
TaqMan
1989
2096
MI
G
Case-control
8
Y
Peden1 [33]
2011
UK
Caucasian
Hospital-based
Illumina
8424
7268
CHD
G
Case-control
8
Y
Peden2 [33]
2011
UK
Asian
Hospital-based
Illumina
6996
7794
CHD
G
Case-control
8
Y
Erdmann [34]
2011
Germany
RT-PCR
1157
1748
CHD
G
Case-control
8
Y
Saade [35]
2011
Lebanon
Asian
Hospital-based
Illumina
1524
425
CHD
G
Case-control
8
Y
Davies [36]
2010
Canada
Caucasian
Hospital-based
Affymetrix
3323
2319
CHD
G
Case-control
8
Y
Kathiresan [37] 2009
US
Caucasian
Hospital-based
Affymetrix
2967
3075
MI
G
Case-control
8
Y
Helgadottir [38] 2007
Iceland
Caucasian Population-based
Illumina
4479
7269
MI
G
Case-control
8
Y
UK
Caucasian Population-based
Affymetrix
2801
4582
CHD
G
Case-control
8
Y
Samani [39]
2007
Caucasian Population-based
Case-control design was used in all the included studies. year = publication year; NOS = Newcastle-Ottawa scale; HWE = Hardy-Weinberg equilibrium, CHD = coronary heart disease, MI = myocardial infarction.
Results Study characteristics The search of the six databases identified 225 records in total. After removing duplicated studies, there were 182 studies left for screening and 25 of records were excluded. 39 studies were read by full-text, and 19 of full-text articles were excluded, including 6 metaanalyses, 6 reviews, 3 articles not relevant to rs497Figure 1. The flow diagram of the 7574 and 4 articles not relstudy selection and exclusion. evant to CHD. There were eventually twenty studies [18-26, 29-39] of 51,522 cases and 73,230 performed by combining ORs repeatedly with omission of each study to identify potential controls eligible for this meta-analysis on the alternation of the overall meta result. We have relationship between CDKN2B-AS gene rs497also investigated publication bias vias calculat7574 polymorphism and CHD. All of these artiing Begg’s and Egger’s test and drawing Begg’s cles were published in English. The sample funnel plot. P > 0.05 was considered that there sizes ranged from 334 to 23,949 of all eligible was no statistically significant bias of publicastudies. The races of the included studies were tion. Meta-analysis was performed using Stata Asian (n = 11) and Caucasian (n = 10). (Peden’s version 14.0 (Stata Corporation, USA). study included different stages of Asian and
12988
Int J Clin Exp Med 2017;10(9):12986-12994
CDKN2B-AS polymorphism and CHD
Figure 2. Forest plot from the meta-analysis on the association of CDKN2B-AS rs4977574 polymorphism and CHD risk in allele model (G vs A allele distribution frequency of CDKN2B-AS gene rs4977574 polymorphism). CHD = coronary heart disease, CI = confidence interval, OR = odds ratio.
Caucasian, so we divided it into two parts for meta-analysis and subgroup analysis). All the included studies fitted in with the HWE test. The NOS scores of all eligible studies in our meta-analysis were higher than 6 points, representing a good study quality. Characteristics of the studies included for meta-analysis are shown in Table 1. Figure 1 shows the complete procedure of the study selection and exclusion. Quantitative synthesis Combining the 20 included studies on the association between the CDKN2B-AS gene rs4977574 polymorphism and susceptibility to CHD provided 51,522 patients and 73,230 controls for this meta-analysis. To analyze the whole datasets, a randomized effect model was used
12989
to perform the pooled analysis regarding to I2 > 50%. A significant increased risk of CHD was found to be related to the CDKN2B-AS gene rs4977574 polymorphism in overall population (OR = 1.26; 95% CI = 1.21 to 1.32; p 50%. Therefore, we conducted stratified analyses to explore the source of heterogeneity. In subgroup analysis by ethnicity, we found significant association between the CDKN2B-AS rs4977574 polymorphism and CHD risk in both Asian (OR = 1.28, 95% CI = 1.19-1.38) and Caucasian (OR = 1.25, 95% CI = 1.18-1.32). A randomized effect model was used to perform this pooled analysis regarding to I2 > 50%. Figure 3 shows the forest plot of subgroup meta-analysis result of ethnicity. We
Int J Clin Exp Med 2017;10(9):12986-12994
CDKN2B-AS polymorphism and CHD
Figure 3. Subgroup meta-analysis by ethnicity of the relationship between CDKN2B-AS rs4977574 polymorphism and CHD risk in allele model. CHD = coronary heart disease, CI = confidence interval, OR = odds ratio.
have also performed subgroup analysis according to source of control, sample size and genotyping method. The detailed information is presented in Table 2. When processing the subgroup analysis by sample size, no significant association was found between CDKN2B-AS rs4977574 polymorphism and CHD risk in sample size
