p.Arg72Pro polymorphism of P53 and breast cancer risk: a meta-analysis of case-control studies

Background The effect of the p.Arg72Pro variant of the P53 gene on the risk of development ofbreast cancer remains variable in populations. However, the use ofstrategies such aspoolingage-matched controls with disease may provide a consistent meta-analysis. Our goal was to perform a meta-analysis in order to assess the association of p.Arg72Pro variant of P53 gene with the risk of breast cancer. Methods Databases such as PubMed, Genetics Medical Literature, Harvard University Library, Web of Science and Genesis Library were used to search articles. Case-control studies with age-matched on breast cancer havingevaluated the genotype frequencies of the TP53 p.Arg72Pro polymorphism were selected. The fixed and random effects (Mantel-Haenszel) were calculated using pooled odds ratio of 95% CI to determine the risk of disease. Inconsistency was calculated to determine heterogeneity among the studies. The publication bias was estimated using the funnel plot. Results Twenty-one publications with 7841 cases and 8876 controls were evaluated in this meta-analysis. Overall, our results suggested that TP53 p.Arg72Pro was associated with the risk of breast cancer for the dominant model (OR = 1.09, 95% CI = 1.02–1.16, P = 0.01) and the additive model (OR = 1.09, 95% CI = 1.01–1.17, P = 0.03), but not for the recessive model (OR = 1.07, 95% CI = 0.97–1.18, P = 0.19). According to the ethnic group analysis, Pro allele was associated with the risk of breast cancer in Caucasians for the dominant model and additive model (P = 0.02), and Africans for the recessive model and additive model (P = 0.03). Conclusions This meta-analysis found a significant association between TP53 p.Arg72Pro polymorphism and the risk of breast cancer. Individuals carrying at least one Pro allele were more likely to have breast cancer than individuals harboring the Arg allele.


Background
Breast cancer is a multifactorial disease which constitutes a major public health problem [1]. In 2018, the World Health Organization reported that 2.09 million new cases of breast cancer were detected [1] compared to 1.38 million cases in 2008 [2]. It is the leading cause of death in women around the world. It should be noted that the incidence of breast cancer differs among different populations around the world [1]. Over the past decades, major advances have been made in understanding the pathology of breast cancer at the molecular level, including the involvement of certain genes associated with the development of the disease such as BRCA1, BRCA2 and P53 which produce tumor suppressor proteins and participate in damaged DNA repair [3][4][5]. P53 plays a key role in the regulation of cell proliferation and apoptosis. The P53 protein is essential for maintaining the integrity of the cell and its components. In human cancers, mutated P53 produces abnormal proteins that alter or inhibit transcriptional regulation [6]. As a result, the stress response, cell cycle as well as apoptosis are affected. Inactivation or mutation of P53 gene would lead in linkage disequilibrium in the DNA sequence, which, associated with chromosomal aberrations induce the appearance of genomic instability and later the development of cancer [7,8]. The molecular signature of human cancers shows that this gene is frequently observed in its mutated form [9]. P53 has been mapped on chromosome 17p13 and contains 11 exons. Several single nucleotide polymorphisms (SNP) have been identified and the most studied variant is the substitution of Arginine by Proline at position 72 in exon 4. Studies carried out on different populations around the world have shown that this SNP is associated with the development of numerous diseases including cancers [10,11]. It should be noted that, many association studies have examined the relationship between the SNP p.Arg72Pro of P53 gene and the risk of breast cancer, however, the reports from these studies remain conflicting as some studies have shown that p.Arg72Pro is associated with the risk of breast cancer, while others found no association. Menzel et al. 2004 [12] and Akkiprik et al. 2009 [13] in their investigations showed a link between p.Arg72Pro and the risk of breast cancer. However, other authors who carried out a case-control study in which participants ages were not matched in a similar population, have concluded that p.Arg72Pro was not associated with the risk of breast cancer [14]. These different results with diverging conclusions can be explained by a very strong heterogeneity in allele and genotype distribution of p.Arg72Pro of the P53 gene. This heterogeneity may be related not only to the geographic and ethnic origin [15][16][17] but also to the study design such as non-agematched case-control studies. Based on these above observations, we hypothesized that the p.Arg72Pro polymorphism of P53 gene may represent a potentially important genetic marker, contributing to breast cancer susceptibility in Caucasian, Asians and Africans. The present meta-analysis included only age-matched casecontrol studies in order to statistically decrease the heterogeneity between the studies, to qualitatively assess the effect of p.Arg72Pro on the risk of breast cancer. We have performed an independent two-stage meta-analysis; overall and sub-group analysis.

Literature search
The Pubmed Genetics Medical Literature Database, the Harvard University Library, and the Web of Science and Genesis Library were used to identify available articles published in English. The keywords "P53", "p.Arg72Pro" and "polymorphism" or "mutation" or "gene" and "breast cancer"cited in the genetic association studies were used to detect and select scientific manuscripts in these databases. We also reviewed references cited in these studies to identify additional articles that were not identified by our research in the databases.

Inclusion criteria
The inclusion criteria included: (1) published casecontrol studies as an original article to evaluate the association between p.Arg72Pro of the P53 gene and the risk of breast cancer, (2) full manuscript available, (3) casecontrol study with age-matched, (4) distribution of genotype respecting Hardy-Weinberg equilibrium (HWE) in controls, (5) availability of the three genotypic frequencies (Arg/Arg, Arg/Pro and Pro/Pro) in the case and control groups. (6) Study no influencing the pooled odd ratio (OR) values. Three investigators independently evaluated each study to determine eligibility.

Data extraction
The data were collected by an investigator and verified by a second investigator to reach consensus on all points. First author, year of publication, country, ethnicity of study population, sample size, age-matched, distribution of genotype andalleles, as well as the recalculation of HWE in controls were extracted from the eligible studies. A third reviewer made a contradictory assessment to reconcile the assumptions. The data of controls evaluated with p.Arg72Pro variant were included in this meta-analysis.

Statistical analysis
Chi 2 analysis with a significance level of P < 0.05 was used to evaluate whether p.Arg72Pro polymorphism distribution of the P53 gene in controls fits HWE. The association between the p.Arg72Pro and the risk of breast cancer was evaluated by the Odd ratio (OR) of 95% CI. We evaluated the strength of association between the p.Arg72Pro polymorphism of P53 gene and the risk of breast cancer using different genetic models, including the dominant (Pro/Pro + Arg/Pro vs. Arg/Arg), recessive (Pro/Pro vs. Arg/Arg + Arg/Pro) and the additive (Pro vs. Arg). Heterogeneity among the studies was assessed by I 2 statistical test [18,19]. If I 2 > 50% (presence of heterogeneity), the random effects model was used to calculate the overall OR, otherwise in case of lack of heterogeneity, the fixed effects method was used. We also have examined the funnel plot to determine publication bias [20]. All statistical analyses were performed with Review Manager Software version 5.1.

Sensitive analysis
To maintain the stability of the meta-analysis after the non-inclusion of deviant studies of HWE, we evaluated the influence of each study on pooled OR. After the exclusion of studies [39][40][41][42],no study has shown a significant influence of pooled OR effect and p-values for the different genetic models ( Table 2).

Sources of heterogeneity
To avoid large heterogeneity, we excluded studies in which the distribution of genotypes deviated from the HWE equilibrium. The sensitivity analysis overall, showed moderate heterogeneity (I 2 < 50%) in the recessive and dominant models. We noted the same tendency of heterogeneity when considering all the data (I 2 = 47%, P = 0.01) and among Asians (I 2 = 54%, P = 0.03) for the additive model (Table 2). In addition, we compared the pooled OR of the fixed and random effects, no statistically significant difference was found between the two effects, which supports strongly the consistency of the present study's data.

Publication Bias
The publication bias was assessed using the funnel plot.
After the exclusion of studies deviating from HWE and those influencing the pooled ORs values, no significant publication bias was found in the different genetic models (Fig. 7).

Discussion
Like other multifactorial diseases, the causes of breast cancer are not known. However, several factors combine their effects for the development of the disease. These factors are of clinical, biological, environmental and genetic origin [43,44]. From a genomic point of view, it has been reported that genetic polymorphisms of the p53 gene can influence the development of cancers [45]. However, the mechanism by which these polymorphisms affect cancer development remain unknown. Functionally, these polymorphisms alter alternative splicing and thus affecting mRNA stability and protein synthesis. The normal P53 gene produces a protein that plays a key role in DNA repair, cell cycle control and apoptosis [45]. Through this physiological role P53 acts as a guardian of the genome, preventing the malignant transformation of normal cells. In the event of a mutation, the function of p53 is impaired, leading to the appearance of malignant cells and later cancerous disease [46][47][48]. It has been reported that the R72 variant of the P53 mutant in addition to influencing the onset of cancer is also associated with a bad prognosis through the rapid onset of metastasis [49].
In the present meta-analysis, we examined the relationship between TP53 p.Arg72Pro polymorphism and the risk of breast cancer. Overall, our findings showed that the dominant and additive models were associated with an increased risk of breast cancer for the carriers of Fig. 4 Forest plots of the association between breast cancer and TP53 p.Arg72Pro polymorphism for the additive model. The black diamond denotes the pooled OR; blue squares indicate the OR in each study with square sizes inversely proportional to the standard error of the OR; and horizontal lines represent the 95% CI 72Pro allele. These results corroborate with the data reported from two recent meta-analyzes, the first covering eleven studies with 950 cases and 882 controls in the Asian population [50] and the second performed on the Indian population which covered seven studies with 1249 cases and 1838 controls [51]. These authors in their analysis showed that the dominant and the additive models were associated with the risk of breast cancer. Contrary to our results, other meta-analyzes found conflicting results [52,53].  the different genetic models applied [54][55][56][57]. These differences might be explained by the samples size, types of allelic variant and eligible studies included. In the subgroup analysis, our meta-analysis revealed a high risk of breast cancer with TP53 p.Arg72Pro in Caucasians (dominant model and additive model) and Africans (recessive and additive models). These trends were consistent with previous studies [12,13,22] but inconsistent with the findings of other studies [52,53]. However, Jafrin et al. 2020 concluded that TP53 p.Arg/Pro was associated with the risk of breast cancer in the South Asian population. The difference between the studies could be explained by the ethnicity and study design. The effects of ethnicity may be due to several factors, allelic heterogeneity, gene-gene and gene-environment interaction and linkage disequilibrium [58][59][60][61]. In the Fig. 6 Forest plots of the association between breast cancer and TP53 p.Arg72Pro polymorphism for the a dominant model, b recessive model and c additive model in Asians. The black diamond denotes the pooled OR; blue squares indicate the OR in each study with square sizes inversely proportional to the standard error of the OR; and horizontal lines represent the 95% CI previous meta-analyzes, the selection criteria of studies were not sufficiently robust such as inclusion of the ageunmatched case-control studies and the inclusion of studies with control groups not satisfying HWE [62][63][64][65][66][67][68][69][70][71][72][73][74][75]. The major advantage of the present meta-analysis was the inclusion of a large number of samples, including very selective criteria to measure the strength of the association between this polymorphism in exon 4 of TP53 gene and the risk of breast cancer using different genetic models. However, several limitations need to be highlighted, sample size and small number of casecontrol age-matched studies in ethnic groups.

Conclusion
In the light of this meta-analysis, we noticed that individuals carrying at least one Pro allele of the P53 gene are more likely to have breast cancer with dominant and additive models than individuals harboring the wild-type Arg allele. Our study further strengthened and confirmed the hypothesis that the P53 gene is usually mutated in about half of breast cancer cases. For the stability and homogeneity of results from meta-analysis, future similar studies should take into account selection criteria for articles such as no deviation from HWE in the control group and the matching of cases and controls according to age.