Introduction

Cancer is a global public health problem, and incidence and mortality are rapidly growing worldwide. According to the data of the International Agency for Research on Cancer, GLOBOCAN 2020 investigation results showed 19.3 million new cancer cases and 10.0 million cancer deaths in 2020 [1]. In 2023, it is estimated that there will be 1,958,310 new cancer cases and 609,820 cancer-related deaths in the USA [2]. With the rapid growth and aging of the world population, the predominance of cancer is a leading cause of death. Current evidence suggests that factors, such as irregular lifestyles, smoking, alcohol intake, environmental factors, and genetic factors, are closely associated with the occurrence of cancer [1, 2]. Accumulative evidence has demonstrated that genetic factors may be associated with the etiology of cancer and the individual’s risk of cancer development, especially whole-genome association studies (GWAS) have identified various genes that may be involved in cancer development [3, 4].

Vitamin D, an essential fat-soluble vitamin, is mainly come from ultraviolet exposure and diet metabolism [5]. Meanwhile, it plays critical roles in cellular growth and anti-proliferative activities [6]. Clinical studies have indicated that vitamin D deficiency contributed to cancer risk, including prostate cancer, breast cancer, and thyroid carcinoma [7]. 25 hydroxy vitamin D (25(OH)D) is the main circulating form of vitamin D. In addition, 1,25 dihydroxy vitamin D (1,25(OH)2D3), an active form of vitamin D, which is associated with cell functions and gene expression. In the process of vitamin D metabolism, 25(OH)D and 1,25(OH)2D3 are converted to 24,25 dihydroxy vitamin D (24,25(OH)2D3) and 1,24,25 trihydroxy vitamin D (1,24,25(OH)3D3), respectively, which are degraded by 25-hydroxyvitamin D 24-hydrolase (encoded by CYP24A1 gene) [8]. Mutation of CYP24A1 may influence the metabolism of Vitamin D and anti-proliferative effects [9, 10].

Single-nucleotide polymorphisms (SNPs) are the most common form of variation in the human genome, which can alter the expression level or function of genes or their encoded products and thus determine the phenotype of the organism [11, 12]. Therefore, it is increasingly recognized that SNPs play a crucial role in the mechanisms of cancer [13]. Epidemiological studies have demonstrated that several common SNPs of CYP24A1 are involved in the concentration of circulating 25(OH)D [14]. To date, five common SNPs (rs4809960, rs6068816, rs2296241, rs4809957, rs2762939) were found to be associated with cancer risk, including esophageal squamous cell carcinoma prostate cancer, breast cancer, and lung cancer [5, 14, 15]. However, controversial results were reported and the association was not yet well established. Therefore, a comprehensive meta-analysis was performed to better explore the associations of CYP24A1 polymorphisms with cancer risk.

Materials and methods

This study was performed under the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) and was registered in PROSPERO (CRD42023446451).

Search strategy

The relevant paper was identified (published until Feb. 2023) through Embase, PubMed, and Cochrane Library using the following strategy: (CYP24A1 or rs2296241 or rs4809957 or rs2762939 or rs4809960 or rs6068816) and (polymorphism or SNP or variant or variation or mutation or genotype) and (cancer or carcinoma or tumor or neoplasm). In addition, other potential publications were also searched by scanning the reference list. The details of the search strategy can be found in Supplementary Table 1.

Inclusion and exclusion criteria

Relevant studies were included according to the following criteria: (1) case-control studies, (2) evaluated the association between CYP24A1 polymorphism and cancer risk, (3) provided sufficient data to calculate the OR with 95%CI, and (4) control group conform to the Hardy–Weinberg equilibrium (HWE). The exclusion criteria were (1) review, abstract, comment, or letter; (2) duplication publications; and (3) relevant data not reported. In addition, for studies with repeat data, the study with the largest sample size was included. Each ethnicity was regarded as a separate study when different ethnicities were reported in a study.

Data extraction and quality assessment

Two authors independently extracted relevant data from the included studies. The extraction parameters included the first author, publication year, country, ethnicity, sample size, cancer type, genotype, and allele distribution in cases and controls, and the P value of HWE in the control group, methodology quality of each study was assessed according to the Newcastle–Ottawa Scale (NOS).

Statistical analyses

HWE was assessed by the chi-square test. The ORs and 95%CIs were calculated to evaluate the strength under allelic, recessive, dominant, homozygous, and heterozygous models. The P value of < 0.05 was considered as statistically significant. The chi-square test and I2 statistics were calculated to evaluate the heterogeneity across studies. If heterogeneity was found (P<0.10 or I2> 50%), the random-effect model was adopted. Otherwise, the fixed-effect model was adopted. Bonferroni correction was performed to adjust multiple-test P value [16]. Sensitivity analyses were performed to evaluate the stability of the results. Stratified analyses were performed by cancer type and ethnicity. Begg’s and Egger’s test was used to assess publication bias. Statistical analyses were completed using Stata 12.0 software (StataCorp, College Station, TX).

Results

Characteristics of the included studies

A total of 258 articles were retrieved in the initial search. Finally, a total of 18 articles [14, 15, 17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32] (19,017 cancer patients and 21,623 controls) were identified (Fig. 1). Among these 18 articles, nine publications about rs2296241 polymorphism, four publications focused on rs4809957 polymorphism, four on rs2762939 polymorphism, six on rs4809960 polymorphism, and six on rs6068816 polymorphism. In addition, four studies focused on prostate cancer, three on lung cancer, five on breast cancer, one on thyroid carcinoma, three on colorectal cancer, one on esophageal squamous cell carcinoma, and one on pancreas cancer. The characteristics of the included studies were described in Table 1.

Fig. 1
figure 1

Flow chart of studies selection process

Full size image
Table 1 The main characteristics of the included studies
Full size table

Meta-analysis of rs4809960

Six publications [21, 26, 30,31,32,33] including seven studies (4509 cancer patients and 5210 controls) examined rs4809960 polymorphism. As shown in Table 2, no significant association between rs4809960 polymorphism and overall cancer susceptibility (Table 2). Subgroup analyses by ethnicity indicated that rs4809960 polymorphism was related to Caucasian population (TT vs. TC+CC: OR 1.18, 95%CI 1.01~1.37, P=0.035; C vs. T: OR 0.88, 95%CI 0.79~0.98, P=0.016) and Asian population (CC vs. TC+TT: OR 1.52, 95%CI 1.01~2.28, P=0.044; TT vs. TC+CC: OR 0.79, 95%CI 0.65~0.97, P=0.021; CC vs. TT: OR 1.52, 95%CI 1.08~2.48, P=0.020; C vs. T: OR 1.24, 95%CI 1.06~1.46, P=0.008). Subgroup analyses by cancer type revealed that rs4809960 polymorphism decreased breast cancer risk (TT vs. TC+CC: OR 1.19, 95%CI 1.05~1.36, P = 0.007; TC vs. TT: OR 0.82, 95%CI 0.72~0.94, P=0.004; C vs. T: OR 0.89, 95%CI 0.80~0.99, P=0.033). However, we only observed that rs4809960 polymorphism was significantly associated with the risk of breast cancer after Bonferroni correction.

Table 2 Summary of meta-analysis of association of rs4809960 polymorphism and cancer risk
Full size table

Significant heterogeneity was found in all genetic models. Sensitivity analysis suggested that a significant association between rs4809960 polymorphism and overall cancer susceptibility was found (TT vs. TC+CC: OR 1.20, 95%CI 1.03~1.39, P=0.020, I2 = 53.1%; TC vs. TT: OR 0.84, 95%CI 0.70~0.99, P=0.043, I2 = 60.2%; C vs. T: OR 0.88, 95%CI 0.81~0.95, P=0.001, I2 = 37.2%) when after removed Yi et al. (Fig. 2). No visual publication bias was detected under the allelic genetic model. In addition, Egger’s test showed that there was no publication bias under the allelic genetic model (P=0.347).

Fig. 2
figure 2

Sensitivity analysis for association between rs4809960 polymorphism and cancer risk (C vs. T)

Full size image

Meta-analysis of rs2296241

Nine publications [14, 15, 17, 19, 21,22,23,24,25] including 5831 cancer patients and 6179 controls were used to calculate pooled ORs and 95%CIs. As shown in Table 3, there was no significant association between rs2296241 polymorphism and overall cancer susceptibility in all genetic models. Subgroup analysis was performed according to ethnicity and cancer type. Stratification by ethnicity indicated that rs2296241 polymorphism was not related to ethnicity. In addition, subgroup analyses by cancer type revealed that rs2296241 polymorphism increased the risk of esophageal squamous cell carcinoma (AA vs. GG+AG: OR 1.34, 95%CI 1.04~1.74, P = 0.023) and decreased risk in prostate cancer (A vs. G: OR 0.91, 95%CI 0.84~0.99, P=0.022) (Fig. 3) (Table 3). However, these associations were no longer significant after the Bonferroni correction.

Table 3 Summary of meta-analysis of association of rs2296241 polymorphism and cancer risk
Full size table
Fig. 3
figure 3

Forrest plot for association between rs2296241 polymorphism and cancer risk (A vs. G)

Full size image

Significant heterogeneity was found under the recessive, homozygous, and allelic models. Sensitivity analysis showed that the initial result was not changed by removing each study respectively. No visual publication bias was detected under the allelic genetic model (Fig. 4). In addition, Egger’s test showed that there was no publication bias under the allelic genetic model (P=0.066).

Fig. 4
figure 4

Begg’s funnel plot for association between rs2296241 polymorphism and cancer risk (A vs. G)

Full size image

Meta-analysis of rs4809957, rs2762939 and rs6068816

Four publications [18, 20, 27, 28] (1851 cancer patients and 2570 controls) about rs4809957 polymorphism, four publications including five studies [15, 21, 22, 26] (2731 cancer patients and 2736 controls) about rs2762939 polymorphism and six publications [21, 26, 29,30,31, 33] (4095 cancer patients and 4829 controls) about rs6068816 polymorphism. As shown in Table 4, subgroup analyses revealed that rs4809957 polymorphism was significantly associated with Caucasian, especially pancreas cancer patients (GG vs. GA+AA: OR 0.80, 95%CI 0.66, 0.98, P=0.029; GA vs. GG: OR 1.27, 95%CI 1.04, 1.56, P=0.022). Furthermore, a significant association was found in breast cancer (AA vs. GG+GA: OR 1.70, 95%CI 1.22~2.58, P=0.012; AA vs. GG, OR 1.80, 95%CI 1.15~2.82, P=0.010; A vs. G: OR 1.27, 95%CI 1.03~1.55, P=0.024) (Fig. 5). In addition, there was no association between rs2762939 polymorphism with cancer risk (Table 5). For rs6068816, we found that rs6068816 polymorphism significantly decreased lung cancer (CC vs. CT+TT: OR 1.45, 95%CI 1.07~1.97, P = 0.016; TT vs. CC: OR 0.58, 95%CI 0.35~0.99, P = 0.044; CT vs. CC: OR 0.71, 95%CI 0.52~0.98, P = 0.036; T vs. C: OR 0.76, 95%CI 0.61~0.95, P = 0.016) and breast cancer risk (TT vs. CC+CT: OR 0.52, 95%CI 0.27~0.98, P = 0.043; TT vs. CC: OR 0.52, 95%CI 0.25~0.97, P = 0.039) (Fig. 6) (Table 6). However, these associations were no longer significant after the Bonferroni correction.

Table 4 Summary of meta-analysis of association of rs4809957 polymorphism and cancer risk
Full size table
Fig. 5
figure 5

Forrest plot for association between rs4809957 polymorphism and cancer risk (GG vs. GA+AA)

Full size image
Table 5 Summary of meta-analysis of association of rs2762939 polymorphism and cancer risk
Full size table
Fig. 6
figure 6

Forrest plot for association between rs6068816 polymorphism and cancer risk (TT vs. CC+CT)

Full size image
Table 6 Summary of meta-analysis of association of rs6068816 polymorphism and cancer risk
Full size table

Sensitivity analysis showed that removing each study respectively from the meta-analysis did not change the initial result. No publication bias was detected in the studies about rs4809957 and rs2762939 polymorphism meta-analysis.

Discussion

CYP24A1, a member of the cytochrome P450 enzyme family, is located on the long arm of chromosome 20 (20q13.2). It is a key gene that converted 1,25(OH)2D3 to 1,24,25(OH)2D3 by 24-hydroxylation25-hydroxyvitamin D 24-hydrolase [34]. Albertson et al. [35] first identified the 20q13 gene amplification in breast cancer and identified the CYP24A1 gene as a candidate oncogene using array comparative genomic hybridization. CYP24A1 has been identified as a potential biomarker for cancer [36]. Numerous studies have suggested the expression level of the CYP24A1 was abnormally increased in several cancers, such as breast cancer, ovarian cancer, cervix carcinoma, lung cancer, and colon cancer [7, 37, 38]. Kong et al. [39] revealed that the rs6068816 and rs4809957 polymorphisms were associated with NSCLC risk. For breast cancer, Wei et al. [27] reported a significant association between the rs4809957 and breast cancer risk. Anderson et al. [18] revealed no significant correlation between rs4809957 with pancreas cancer. Among these publications reported the associations of CYP24A1 polymorphisms with cancer susceptibility, while the results remain controversial. The previous meta-analysis was performed by Zhu et al. [40], but they had not controlled the type I error rate through Bonferroni correction and had a smaller sample size. Therefore, the present meta-analysis aimed to re-evaluate the associations of CYP24A1 polymorphisms with cancer risk.

The present study indicated that there was no association between CYP24A1 polymorphisms (rs4809960, rs2296241, rs4809957, rs2762939, rs6068816) and overall cancer risk. For rs4809960 polymorphism, it was related to the Caucasian and Asian populations and decreased breast cancer risk. Moreover, our results suggested that rs2296241 polymorphism increased esophageal squamous cell carcinoma risk and decreased prostate cancer risk. For rs4809957 polymorphism, it was associated with pancreas cancer and breast cancer risk. In addition, we found that rs6068816 polymorphism significantly decreased lung cancer and breast cancer risk. However, rs4809960 polymorphism was associated with a decreased breast cancer risk after Bonferroni correction. A previous meta-analysis also reported CYP24A1 rs2296241 polymorphism was associated with prostate cancer risk [41]. Although our work found rs2296241 polymorphism was associated with an increased esophageal squamous cell carcinoma risk and decreased prostate cancer risk, these results could not withstand the Bonferroni correction.

The improvements of our meta-analysis are as follows: Firstly, more case-control studies about rs4809960, rs6068816, and rs2296241 polymorphism were included in the meta-analysis. Secondly, this is the first meta-analysis to assess the relationship between CYP24A1 (rs4809957, rs2762939) polymorphism and cancer risk. Thirdly, all included studies conform to the HWE, which may improve the reliability and stability of our study. In addition, all CYP24A1 polymorphisms were considered at the beginning. Ultimately, due to a lack of eligible articles and overlapping studies, our further evaluation of other CYP24A1 polymorphisms was limited. Therefore, in this meta-analysis, we only focused on five polymorphisms.

There are several limitations should be noted in the present study. First, the sample size of the included studies was relatively small, which might weaken the strength of the results. Second, the number of included studies in the subgroup analysis was also relatively small, which might lead to statistical bias. Third, not sufficient data to analyze whether environmental factors may influence the statistical result. Four, the association of CYP24A1 polymorphism with different types or stages, drinking, smoking, age, gender, exposure factors, or other risk factors was not considered in this study. Five, the patients of included studies mainly come from Caucasians. The African and Asian populations were relatively small.

In conclusion, this meta-analysis indicated that rs4809960 polymorphism was associated with a decreased breast cancer risk. No association between rs4809957, rs2296241, rs2762939, rs4809957 polymorphism, and overall cancer risk was found after Bonferroni correction. Considering the above limitations, more large-scale and large-sample studies are necessary to confirm these results.