Prostate cancer (PCa), which is highly prevalent in developed countries, is the most common cancer in men and the third most common cause of cancer-related deaths in Europe after lung and colorectal cancer (1, 2). In the majority of cases, cancer is efficiently eradicated and cured by surgical prostatectomy, but late-diagnosed or metastatic aggressive cancers have a five-year relative survival rate of less than 30% (3). The tumor suppressor gene TP53 plays a pivotal role in the prevention of tumor development. Not surprisingly, it has also been described as the most commonly mutated gene in human cancers (4). Somatic TP53 mutations occur more frequently as later events and are drivers of aggressive and metastatic cancers, including PCa (5).

Several hereditary genetic variations, found in approximately 5-20% of different cancer types, are associated with pathogenic variants of genes, such as BRCA2 and ATM. In PCa, pathogenic germline variants of the TP53 gene have been described with a relative risk of 4.7-8.6, which is comparable to the frequencies described for those genes (6).

Germline variant alleles, known as single-nucleotide polymorphisms (SNPs), are present in over 1% of the population and occur naturally. Although many of these variants are considered to have minor or not impact on human health, recent significant studies indicate otherwise, particularly in PCa (7, 8). Some common SNPs can indeed affect the structure and function of proteins, thereby contributing to severe health outcomes. In PCa, specific SNPs, such as those in the BTNL2 gene, have been associated with elevated prostate cancer risk, and structural variants have been linked to poor clinical outcomes and disease progression (9, 10).

Despite significant progress in understanding PCa genetics, the understanding of some controversial germline genetic variations is still incomplete. In particular, the P72R single nucleotide polymorphism (rs1042522) has shown conflicting results in its association with prostate cancer risk and aggressiveness in various reports (11–14). These studies suggest that P72R may influence PCa susceptibility, although the exact nature of this influence remains debated within the scientific community.

By investigating the association of the rs1042522 SNP of the TP53 gene in cancer tissues, primary tumor cultures, and serum samples, our study aims to provide additional evidence of the association of this SNP with prostate cancer risk in a European Caucasian population. Here, we describe the identification of the rs1042522 in the majority of hnPCs analyzed. Its presence was confirmed in a cohort of 11 radical prostatectomy (RP) tissues and 94 serum samples from Caucasian patients with prostate cancer. The association between the variant allele and prostate cancer risk was also investigated by comparing with the healthy (non-cancerous) European (non-Finnish) population from GnomAD v2.1.1.

These results highlight the potential value of rs1042522 assessment for identifying patients at risk for aggressive metastatic prostate cancer.

The tumor suppressor gene TP53, reported to be mutated at a high frequency (53.3%) in aggressive metastatic castration resistant prostate cancer, CRPC (4, 17), was the focus of this investigation. The finding that 9 out of 12 primary tumor cultures from patients with aggressive metastatic PCa had a p53 protein with arginine (CGC) at codon 72, but that only 2 cultures had a proline (CCC) was intriguing. An initial explanation for the enrichment of the R72 variant in hnPCs from cancer tissues was the loss of heterozygosity (LOH), particularly in exon 4, which includes codon 72 of the protein. In fact, loss of the proline (C) allele and preferential retention of the arginine (G) allele have been described in primary tumors and metastatic tissues of squamous cell carcinoma of the vulva (18), head and neck (19), colorectal tumors (20), and lung cancer (21). In addition, a recent meta-analysis using the TCGA pan-cancer database and patients selected for P72R heterozygosity reported that 31% (127/409) of heterozygotes had lost the P72 allele in the corresponding tumor tissue (22), indicating that the G allele is preferentially selected for tumor development, possibly due to enhanced p73-induced apoptosis (16) and the modulation of tumor metabolism by regulating PGC-1α (23). Although proper LOH screening was not performed in our cancer cohort, genotyping analysis of serum samples showed a similar frequency of the R72 variant compared to that of hnPCs and cancer tissues from prostatectomy, suggesting that this variant was obtained through germline, instead of LOH.

The results in serum samples, confirmed the high frequency of the G allele, compared to a control European non-cancer population selected from the GnomAD v2.1, indicating a very significant association (p<0.0001) between this SNP and prostate cancer risk, with an odds ratio of 7.937 (95% CI 5.37-11.00). Several reports have described a similarly significant association between the R72 SNP and the risk of developing cutaneous melanoma, breast cancer, and prostate cancer (24–27). A study on PCa compared 187 Iranian patients with 185 control individuals negative for cancer (27). Although the frequency of the G allele in patients with cancer (65.77%) was not as high as the one reported here (73.93%), the difference was statistically significant compared to the control group (p<0.05).

Notably, several reports have shown controversial results (28–30) or failed to demonstrate a significant association between the rs1042522 variant and the risk of any type of cancer (14, 31–34), mostly because of limited sample size or selection bias. For example, in a colorectal cancer study in the Northern European population, no significant association between the presence of the SNP and the risk of cancer was reported (35), but the healthy control samples selected had a high frequency of the ARG allele (61%), in contrast to the estimated 26% frequency of the (non-cancer) European population from GnomAD v2.1.1. Similar to our findings, one study on prostate cancer reported an association between Pro/Pro and a lower risk of prostate cancer (11). However, several other reports have shown conflicting results (11–14, 36–38), suggesting that an appropriate representation of control samples is necessary for accurate results.

To address potential geographical sub-population biases, we further validated our results by analyzing the 1000 Genomes database, filtered by country, and used the Spanish population as a control. In the control population, we observed that the C allele is present at higher frequencies (0.71) compared to the G allele (0.29), similar to the results obtained with the control non-cancer European population from GnomAD ( Figure 2C ). This indicates that the high frequency of the G allele observed in our serum samples is not attributable to a sub-population effect but it rather reflects its association with aggressive prostate cancer.

These findings underscore the importance of considering population stratification in genetic association studies to avoid potential biases that may skew the results. The use of well-matched control populations is crucial for the accurate interpretation of genetic data. Future studies should continue to account for these variables to ensure the robustness of genetic association findings.

Beyond these conflicting findings regarding ARG frequency at codon 72 of p53, it is crucial to consider the potential impact of this variant on cancer development, although its full understanding is still not clear. Studies in mice have revealed that the presence of the R72 variant is associated with an increased incidence of mammary tumors (39). In addition, this variant is associated with increased phosphorylation of p53 and enhanced transactivation of CDKN1A (p21^WAF1) in response to starvation, with consequent augmented growth arrest and reduced apoptosis, a situation that favors survival (40). Furthermore, TP53 mutations tended to preferentially occur in the R72 allele. In the presence of mutant TP53 (e.g., R175H, R273H, and A138V) harboring the R72 SNP, an enhanced capacity for migration, invasion, and metastasis in various cancer cell lines, including PCa, has been reported (23). These observations provide strong evidence that the ARG at codon 72 affects the tumor-suppressive activity of p53.

Although no significant association between the P72R SNP and Gleason score was found in our study, the G allele (R72) was more frequent in patients with a high Gleason score (≥8, group 4-5), suggesting a potential association with more undifferentiated malignant PCa lesions. A few studies have also evaluated the association between this polymorphism and tumor grade and found no significant association with clinical stage or Gleason grade in PCa patients (36, 37, 41, 42). One study revealed a link between the modest influence of this SNP and shorter biochemical recurrence (BCR) after radical prostatectomy (38). Another study in the Japanese population (12) observed a particularly high frequency of the ARG allele in patients with metastases and in those with a high Gleason score (≥8). However, the latter study analyzed the association using the arginine allele as a reference, and the authors calculated the significance level between the Pro/Pro and Pro/Arg genotypes and arginine. As expected, they found no significant differences, as proline did not differ significantly from that in the control group.

To validate and extend our findings, further investigations with a larger sample size and consideration of family history of prostate cancer are needed. Nevertheless, our results suggest that a simple but powerful test (gDNA assessment for the P72R SNP) may be useful for the early identification of patients at risk for aggressive metastatic prostate cancer. Future studies should focus on developing detailed guidelines on how this SNP assessment could be integrated into current prostate cancer screening allowing clinicians to intervene at an early stage.