Prospects of POLD1 in Human Cancers: A Review

Simple Summary Polymerase delta 1 (POLD1), a catalytic and proofreading subunit of the human DNA polymerase delta (Polδ), plays a vital role in maintaining the stability of the genome. In recent years, POLD1 germline and somatic mutations, as well as gene-expression patterns, have been extensively studied in cancers. Nevertheless, the complex regulatory mechanisms of POLD1 expression and their clinical relevance are yet to be elucidated. This review aims to summarize the current research status regarding the role of POLD1 in neoplastic processes. Abstract Cancer is the second leading cause of death globally, exceeded only by cardiovascular disease. Despite the introduction of several survival-prolonging treatment modalities, including targeted therapy and immunotherapy, the overall prognosis for the metastatic disease remains challenging. Therefore, the identification of new molecular biomarkers and therapeutic targets related to cancer diagnosis and prognosis is of paramount importance. DNA polymerase delta 1 (POLD1), a catalytic and proofreading subunit of the DNA polymerase δ complex, performs a crucial role in DNA replication and repair processes. Recently, germline and somatic mutations of the POLD1 gene have been acknowledged in several malignancies. Moreover, diversified POLD1 expression profiles have been reported in association with clinicopathological features in a variety of tumor types. With this review, we aim to summarize the current knowledge on the role of POLD1 in cancers. In addition, we discuss the future prospects and clinical applications of the assessment of POLD1 mutation and expression patterns in tumors.


Introduction
Within the next twenty years the number of new cancer cases is expected to nearly double globally [1]. In addition, the observed increase in incidence will presumably be paralleled by a rise in cancer-related deaths [1]. The projected increase is expected to be driven by both aging and the growth of the human population worldwide [1]. Despite the rapid progress in cancer diagnosis and treatment, when it is diagnosed at a metastatic stage the prognosis remains poor [2]. Therefore, the identification of novel molecular diagnostic, prognostic, and predictive biomarkers may be highly warranted.
The polymerase delta 1 (POLD1) gene encodes the major catalytic and proofreading subunit of the DNA polymerase δ (Polδ) holoenzyme [3]. It is crucial that POLD1 expression is precisely regulated during the cell cycle [4]. With regard to the Polδ complex, it plays an essential role in DNA replication as it synthesizes the DNA at the lagging strand. Furthermore, the important error-correcting ability provided by the exonuclease activity of this enzyme contributes significantly to replication fidelity [5]. In addition, Polδ is involved in DNA repair upon exposure to mutagens [6].
Both germline and somatic alterations in the proofreading domain of POLD1 have been implicated in various diseases, including cancers [7]. Moreover, POLD1 mutation is thought to be strongly associated with genomic instability, mutator phenotype, and tumorigenesis [7][8][9].
Although POLD1 germline mutation was primarily linked to colorectal and endometrial cancers, more recent data have also found a connection between this mutation and an increased risk of many other tumor types [7,10]. Notably, two clinical conditions regarding the cancer-predisposing nature of the POLD1 mutation have been mentioned in the literature: the first, defined as polymerase proofreading-associated polyposis [11] and the second, the recently described POLE/POLD1-associated tumor syndrome, which have expanded the spectrum of possible tumors affecting people with inherited DNA polymerase-related mutations [12]. Nonetheless, somatic mutations of the POLD1 gene are inadequately studied in cancer [7]. Thus, they warrant further investigation to be appropriately addressed clinically and practically.
Past articles have thoroughly reviewed the role of POLD1 in tumors in 2016 [7,9]. Therefore, our aim is to summarize and update the current understanding of the POLD1 mutations and expression changes in human cancers. Moreover, we address possible therapeutic implications and future challenges in POLD1-mutated cancers.

POLD1 Gene and a Polymerase Delta Holoenzyme in Health
The DNA polymerase delta 1 catalytic subunit (POLD1) gene is located on chromosome 19 at q13.3-q13.4 [27,28]. The human POLD1 gene is composed of 27 exons and 26 introns. It encodes a large 125-kDa protein known as the POLD1/p125 subunit, a catalytic and proofreading subunit of the human DNA polymerase delta (Polδ). The fluctuations of the POLD1 promoter activity imply that the transcriptional mRNA and protein levels are regulated during the cell cycle [4]. The POLD1 gene promoter is activated by Sp1, Sp3 transcription factors, cell cycle-related E2F1, and CCCTC-binding factor (CTCF) [29,30]. On the other hand, the p53 protein, the "genome guardian", and cyclin-dependent kinase inhibitor p21 are able to inhibit the Sp1-stimulated POLD1 promoter activity, thus repressing POLD1 gene expression [31,32].
Song et al. discovered that a cell cycle-dependent element (CDE)/cell cycle gene homology region (CHR) element-is an active functional element in the POLD1 promoter, crucial for the cell cycle regulation of the POLD1 gene [32]. They also showed that E2F1 and p21 regulate the POLD1 promoter via binding to the CDE/CHR element [32]. Interestingly, p53-mediated inhibition of POLD1 transcription seems to be indirect [33]. p53 does not directly bind to the CDE-CHR motifs. Instead, p53 triggers indirect p21-dependent POLD1 downregulation via the p53-p21-DREAM-CDE/CHR pathway [33]. In addition, CTCF mainly binds to the POLD1 promoter in site 3 and 4 regions [30]. It has been shown that these two regions play crucial roles in CTCF-mediated POLD1 regulation [30]. Additionally, microRNA-155-mediated suppression of the forkhead transcription factor FOXO3a subsequently contributes to the decreased expression of POLD1 at the mRNA and protein levels [34].
Moreover, POLD1 expression has been causally linked to age-related cellular senescence and two POLD1-related mechanisms of aging have been proposed [26,30]. Gao et al. showed that the attenuation of the binding affinity of E2F1 for the POLD1 promoter, via agerelated decline in E2F1 expression and increased CpG island 3 methylation in the POLD1 promoter region, causes a reduction in POLD1 expression in replicative senescence [26]. Hou et al. proposed, however, that an age-related decline in CTCF levels affects the binding affinity of CTFC to the POLD1 promoter, thus driving down POLD1 expression [30]. It is, therefore, well established that down-regulation of POLD1 is a characteristic age-related change occurring in humans [26,30,35].
Human Polδ holoenzyme has been believed to be a heterotetramer in unstressed, non-dividing cells composed of a catalytic subunit (POLD1) and three accessory subunits, POLD2 (p50), POLD3 (p66/p68), and POLD4 (p12) [36]. However, it has recently been suggested that the POLD4 subunit may occur as a dimer in the human Polδ complex [37]. Therefore, human Polδ potentially exists under a pentameric structure [37]. Through the catalytic POLD1 subunit, in association with both replication factor C (RFC) and proliferating cell nuclear antigen (PCNA), Polδ plays a pivotal role in discontinuous DNA synthesis at the lagging strand during genome replication. Moreover, its 3 -5 exonuclease domain provides an error-correcting activity during DNA synthesis [38]. It has also been shown that Polδ is required for break-induced telomere synthesis [39].
A pivotal role for POLD1 in maintaining the enzymatic function of the Polδ complex suggests that adequate cellular POLD1 expression levels should be sustained for the smooth functioning of genomic DNA replication [40]. Novel data suggest that POLD1 is protected from degradation while being interconnected with the three different subunits in the Polδ complex, whereas excessive non-associated POLD1 is degraded to intercept the formation of incomplete and inactive Polδ units [40]. Moreover, results obtained in an experimental study using a murine model revealed the overall low levels of POLD1-3 subunits in POLD3deficient cells [41].
Surprisingly, it has recently been shown that POLD1 plays a unique role in both the nucleus and the cytoplasm [42]. Subcellular fractionation has revealed that more than 60% of the POLD1 content occurs in the cytoplasm [42]. Regarding its cytoplasmic function, the POLD1 protein localized at the Golgi complex controls microtubule growth [42]. Shen et al. demonstrated that POLD1 undergoes bidirectional nucleocytoplasmic transport [43]. In addition, they also substantially unraveled the mechanisms underlying both the nuclear import and export of the POLD1 protein [43]. Figure 1 summarizes the mechanisms of POLD1 gene regulation. Moreover, it outlines the physiological role of the POLD1 protein within the cytoplasm and nucleus.
Interestingly, during normal cell cycle progression, conversion of heterotetrameric Polδ (Polδ4) to the heterotrimer (Polδ3) takes place via POLD4/p12 subunit degradation [40,41]. It has been shown that POLD4 undergoes proteasomal degradation initiated by its polyubiquitination, in synchrony with the S phase [44,45]. Contrarily, no alteration in POLD1, POLD2, and POLD3 levels with the progression of the cell cycle was observed in Zhang et al.'s studies [45]. Although both Polδ4 and Polδ3 are active polymerases, Polδ3 seems to be the predominating form of human Polδ during the S phase of the cell cycle [6,45]. Moreover, a direct comparison of Polδ3 with Polδ4 has shown that the heterotrimeric complex displays a higher ratio of exonuclease to polymerase activity in vitro than foursubunit Polδ4 [6]. In addition, POLD4 degradation in response to DNA damage after the application of ultraviolet, methyl methanesulfonate, hydroxyurea, and aphidicolin, causes the concomitant conversion of Polδ4 to a heterotrimeric form [46]. Therefore, exposure to the agents causing DNA damage following Polδ4 to Polδ3 conversion suppresses DNA synthesis, while at the same time favoring proofreading activity under conditions of DNA replication stress. Interestingly, during normal cell cycle progression, conversion of heterotetrameric Polδ (Polδ4) to the heterotrimer (Polδ3) takes place via POLD4/p12 subunit degradation [40,41]. It has been shown that POLD4 undergoes proteasomal degradation initiated by its polyubiquitination, in synchrony with the S phase [44,45]. Contrarily, no alteration in POLD1, POLD2, and POLD3 levels with the progression of the cell cycle was observed in Zhang et al.'s studies [45]. Although both Polδ4 and Polδ3 are active polymerases, Polδ3 seems to be the predominating form of human Polδ during the S phase of the cell cycle [6,45]. Moreover, a direct comparison of Polδ3 with Polδ4 has shown that the heterotrimeric complex displays a higher ratio of exonuclease to polymerase activity in vitro than four-subunit Polδ4 [6]. In addition, POLD4 degradation in response to DNA damage after the application of ultraviolet, methyl methanesulfonate, hydroxyurea, and aphidicolin, causes the concomitant conversion of Polδ4 to a heterotrimeric form [46]. Therefore, exposure to the agents causing DNA damage following Polδ4 to Polδ3 conversion suppresses DNA synthesis, while at the same time favoring proofreading activity under conditions of DNA replication stress.

POLD1 Gene and a Polymerase Delta Holoenzyme in Non-Oncogenic Processes
Numerous studies have documented the physiological role of both Polδ holoenzyme and POLD1 protein. As they are involved in multiple biological processes, including DNA replication and DNA repair, a growing body of studies in recent years has highlighted the role of germline and somatic mutations in POLD1-POLD4 genes in the genesis of human pathologies. Mandibular hypoplasia, deafness, progeroid features, and progressive lipodystrophy (MDPL) syndrome is a rare autosomal dominant disorder caused by heterozygous de novo mutations in the POLD1 gene [47,48]. Notably, exome sequencing revealed that mutations in different POLD1 domains may lead to a major phenotypic variability in MDPL syndrome [48]. Recent data have also revealed that POLD1 mutations causing reduced Polδ polymerase activity with a maintained exonuclease activity can lead to a non-

POLD1 Gene and a Polymerase Delta Holoenzyme in Non-Oncogenic Processes
Numerous studies have documented the physiological role of both Polδ holoenzyme and POLD1 protein. As they are involved in multiple biological processes, including DNA replication and DNA repair, a growing body of studies in recent years has highlighted the role of germline and somatic mutations in POLD1-POLD4 genes in the genesis of human pathologies. Mandibular hypoplasia, deafness, progeroid features, and progressive lipodystrophy (MDPL) syndrome is a rare autosomal dominant disorder caused by heterozygous de novo mutations in the POLD1 gene [47,48]. Notably, exome sequencing revealed that mutations in different POLD1 domains may lead to a major phenotypic variability in MDPL syndrome [48]. Recent data have also revealed that POLD1 mutations causing reduced Polδ polymerase activity with a maintained exonuclease activity can lead to a non-syndromic sensorineural hearing loss in an autosomal recessive manner [49]. In addition, several cases of autosomal recessive POLD1-linked combined immunodeficiency have been reported [50][51][52]. Additionally, Nichols-Vinueza et al. described a potential role for POLD1 in B-cell maturation [52].

POLD1 in Cancers
Knowing that Polδ is crucial in genome maintenance and is involved in the DNA proofreading processes, it is not surprising that mutations in the POLD1 gene have been linked to genomic instability, mutator phenotype, and carcinogenesis [7][8][9]. In this paragraph, we aim to discuss the influence of both germline and somatic POLD1 mutations on different malignancies. Moreover, the changes in the POLD1 expression level in relation to clinicopathological parameters will be addressed.

Colorectal Cancer
Colorectal cancer (CRC) is the third most frequently diagnosed cancer worldwide but is second in terms of mortality [1]. Almost a third of all CRC cases have a familial component, however, only about 5% of these develop as a consequence of well-defined inherited syndromes, whereas the majority of them do not possess a known genetic background [53,54]. To date, colorectal cancer (CRC) seems to be the most extensively studied malignancy concerning POLD1 mutations.
Among patients with a family history of CRC and multiple or large colorectal polyps, a minor fraction of carriers of POLD1 heterozygous germline mutations can be found [55,56]. Regarding somatic pathogenic variants affecting POLD1 in CRC, they are thought to be extremely rare findings [57]. Unfortunately, the overall incidence of POLD1 mutation in CRC patients remains unknown. Since the first description in 2013 of a correlation between the POLD1 mutation and dominantly inherited intestinal adenomas and carcinomas [55], two terms regarding this phenomenon have been coined. Originally, this syndrome was described as "polymerase proofreading-associated polyposis" [11], however, the term "POLE/POLD1-associated tumor syndrome" was recently proposed instead [12], as among some patients with a germline mutation in DNA polymerase subunits POLE and/or POLD1, expanded extracolonic tumor spectrum and absence of colonic polyposis have been noted. Interestingly, tumors with POLE/POLD1 mutations were found to be hypermutated, chromosomally unstable, and, surprisingly, microsatellite stable [55]. Moreover, recent data suggested the occurrence of an abundance of tumor-infiltrating lymphocytes (TILs) in POLE/POLD1-mutant CRCs [58]. In general, lymphocytes located in the tumor tissue region have emerged as the activators of a patient's pre-existing intratumoral immunity after exposure to ICI treatment [58].
Unfortunately, data regarding POLE/POLD1 status to predict the efficacy of immunotherapy in CRC treatment are scarce, thus conducting research in this field would be reasonable.
Recently, Siraj et al. investigated the correlation of immunohistochemical POLD1 protein expression with clinicopathological features in Middle Eastern CRCs [13]. They observed that low POLD1 expression, found in more than half of CRC cases, was correlated with larger tumor size, adenocarcinoma histology, and stage III tumors [13]. Moreover, in our study focused on patients with CRC using IHC analysis, lower POLD1 nuclear expression in CRC cells compared with normal epithelial colon cells was observed. POLD1 expression levels in the tumor cells did not correlate with clinicopathological factors and the prognosis of CRC patients (data not published yet).
In Figure 2, we present representative immunohistochemical staining patterns of POLD1 within healthy, benign polypoid, and malignant tissues.

Endometrial Cancer
With its incidence continuing to rise, endometrial cancer (EC) is the most common gynecological malignancy in developed countries [1,59]. Although the POLD1 mutations in general are very scarce in EC [14,60], it is now well established that POLD1 germline mutations predispose people to EC [11,14]. Presumably, the frequency of POLD1 mutations in EC may be population dependent [60,61]. Among 47 South East Asian women with grade 3 (according to FIGO grades) endometrioid ECs, a presence of two pathogenic germline POLD1 mutations was detected [61]. Whereas Church et al.'s data revealed only one germline POLD1-mutated endometrioid EC among 154 patients with different FIGO grades enrolled in a British study [60].

Endometrial Cancer
With its incidence continuing to rise, endometrial cancer (EC) is the most common gynecological malignancy in developed countries [1,59]. Although the POLD1 mutations in general are very scarce in EC [14,60], it is now well established that POLD1 germline mutations predispose people to EC [11,14]. Presumably, the frequency of POLD1 mutations in EC may be population dependent [60,61]. Among 47 South East Asian women with grade 3 (according to FIGO grades) endometrioid ECs, a presence of two pathogenic germline POLD1 mutations was detected [61]. Whereas Church et al.'s data revealed only one germline POLD1-mutated endometrioid EC among 154 patients with different FIGO grades enrolled in a British study [60].
Moreover, EC was found to be the most common malignancy among female POLD1 germline mutation carriers [62]. Palles et al.'s studies revealed that EC is the most frequent extraintestinal cancer in polymerase proofreading-associated polyposis syndrome [62]. Surprisingly, one recent study on Chinese genomic cancer population data suggests that EC may be the most common malignancy with POLD1 mutations among all cancers [63].
Even though POLD1-mutated cancers are believed to be primarily microsatellite stable [55], some ECs and CRCs possess microsatellite instability (MSI) [63][64][65][66][67]. Notably, Moreover, EC was found to be the most common malignancy among female POLD1 germline mutation carriers [62]. Palles et al.'s studies revealed that EC is the most frequent extraintestinal cancer in polymerase proofreading-associated polyposis syndrome [62]. Surprisingly, one recent study on Chinese genomic cancer population data suggests that EC may be the most common malignancy with POLD1 mutations among all cancers [63].
Even though POLD1-mutated cancers are believed to be primarily microsatellite stable [55], some ECs and CRCs possess microsatellite instability (MSI) [63][64][65][66][67]. Notably, Haraldsdottir et al. have suggested that the deficient DNA mismatch repair system in endometrial and colorectal cancers resulting in MSI phenotype, may occur mainly in tumors with somatic rather than germline POLD1/POLE mutations [64]. In addition, one study that assessed an endometrial dedifferentiated carcinoma molecular profile suggested that POLD1 and MLH1 (a gene involved in DNA mismatch repair), as well as high microsatellite instability (MSI-H) and increased tumor mutation burden (TMB), may be peculiar to this rare distinct subtype of EC [68].
As it is widely known that patients with MSI-H tumors can benefit from immunotherapy [69], routine clinical genetic testing for both POLD1/POLE and MSI-related mutations may help identify patients with endometrial and colorectal cancers who could respond well to immunotherapy [70]. To date, only one study evaluated POLD1 protein immunohistochemical expression in EC and its association with the clinicopathological profile [14]. Low expression of POLD1 was noted in almost 60% of Saudi EC cases and significantly correlated with grade 1 tumors [14].
Nevertheless, conducting research to establish both prognostic and predictive implications of POLD1 mutation and/or POLD1 protein expression for EC patients is justified.

Renal Cancer
Clear cell renal cell carcinoma (ccRCC) is the most common histopathological subtype of renal cancer, accounting for up to 85% of all kidney cancers [71]. Current studies have highlighted the role of DNA damage repair pathways in ccRCC. Two research groups have recently analyzed DNA polymerase epsilon and delta and their possible prognostic significance for patients with ccRCC [15,72]. Our team was the first to report that, unexpectedly, POLD1 nuclear protein expression levels correlated with better prognosis in patients with ccRCC [15]. Moreover, we have also demonstrated that the level of POLD1 immunoexpression in tumor cells does not correlate with the demographic and clinicopathological characteristics of ccRCC patients [15]. This unanticipated data demonstrating the positive correlation of POLD1 nuclear immunoreactivity with longer overall survival among ccRCC patients, in contrast to several reports regarding other malignancies, may suggest that the potential role of POLD1 can be considered as cancer specific [14,17].
Eventually, subsequent molecular studies will be crucial to clarify the role of POLD1 in ccRCC tumorigenesis.

Liver Cancer
Hepatocellular carcinoma (HCC), accounting for 75-85% of primary liver cancers [1], is the second most lethal malignancy after pancreatic cancer, with a 5-year relative sur-vival of 18% [59]. The role of POLD1 in HCC was first presented in 2010 [16]. Sanefuji et al. demonstrated that POLD1 overexpression induced by tp53 mutation correlates with a high degree of vascular invasion, poor cellular differentiation, and a poorer prognosis in HCC [16]. They have also suggested that POLD1 may contribute to cancer cell dedifferentiation in human HCC [16]. Furthermore, apoptosis of HCC cells was shown to be related to the inhibition of POLD1 gene expression [73]. Similarly, apoptosis of triple-negative breast cancer (BC) cells was driven by the reduction in POLD1 expression [74].
Recently, Tang et al. conducted a thorough analysis of the predictive and prognostic value of POLD1 in HCC [17]. It was shown that POLD1 is significantly upregulated in HCC in comparison with adjacent healthy liver tissue [17]. Additionally, POLD1 overexpression correlated with increased levels of alpha-fetoprotein, high TNM stage (III and IV), and predicted worse prognosis in HCC [17]. In addition, aberrantly high expression of POLD1 in HCC tissue was demonstrated to be caused by DNA copy number gain, low POLD1 methylation, and downregulation of small noncoding miR-139-3p [17]. Lastly, POLD1 expression positively correlated with the immune infiltration levels of B cells, macrophages, dendritic cells, and CD4+ T cells in the HCC tumor microenvironment [17]. Moreover, an in vivo orthotopic liver injection model revealed that POLD1 knockdown in HCC cells diminished tumor incidence, size, and lung metastases in mice [75].
Taken together, an in-depth exploration of HCC tumorigenesis can aid in the development of novel clinical treatment modalities.

Breast Cancer
With an estimated 2.3 million new cases in 2020, breast cancer is the current leading cause of global cancer incidence, constituting 11.7% of all cancers diagnosed worldwide [1]. Patients harboring germline POLD1 mutations are at higher risk of developing BC com-pared with the general population [76,77]. Somatic mutations, although insufficiently documented, have also been reported in BC [78,79]. Furthermore, even though BC cases are infrequently associated with MSI [80], this particular group of tumors may be POLD1mutated as well [81].
Research conducted by Zhang et al. highlighted the mechanisms underlying changes in POLD1 expression in BC [18]. They have shown that POLD1 expression and the methylation level of the POLD1 promoter are both increased in BC tissues and cell lines [18]. Moreover, POLD1 expression and the methylation of the POLD1 gene promoter were found to be inhibited by the p53 protein through the suppression of DNA methyltransferase 1 and Sp1 activities in BC [18]. In the following studies, they observed that sirtuin 1, a member of the sirtuin family, enhances proliferative, migratory, and invasive properties of BC cells in vitro by inhibiting the p53 protein expression and, therefore, promoting the POLD1 upregulation [82].
To date, data regarding the expression patterns and clinical value of POLD1 among patients with BC are scarce. However, novel studies conducted by Qin et al. revealed that increased POLD1 expression is linked to poor prognosis in BC [19]. Elevated POLD1 levels were related to lymph node metastasis, histological grade, p53 status, and the ki-67 index [19]. Moreover, survival analysis showed that BC patients with high POLD1 levels had poorer disease-free survival in comparison with patients with low POLD1 expression [19]. In addition, POLD1 status likely affects the treatment response, especially in triple-negative breast cancer [83].
Thus, these data suggest that the POLD1 protein can be considered as a potential prognostic biomarker in BC [19].

Other Cancers
Although POLD1 mutations in cancers seem to be understudied, they may serve as promising prognostic/predictive biomarkers in the future and, importantly, future targets for novel POLD1-oriented therapies. Nevertheless, apart from malignancies mentioned in the previous paragraphs, several different cancers have been reported to harbor POLD1 mutation as well. Wang et al.'s studies that aimed to assess the clinical benefits of immunecheckpoint inhibitor (ICI) treatment have estimated that across almost 50,000 patients with different cancer types, POLD1 mutation occurred in 1.37% of cases [10].
For instance, the overall frequency of POLD1 mutations was estimated to be 2.5% among Chinese patients with lung cancer [84]. Furthermore, a recent analysis revealed that POLD1 is associated with the development of non-small cell lung cancer (NSCLC) from chronic obstructive pulmonary disease [20]. Interestingly, high expression levels of POLD1 indicated poor prognosis in lung adenocarcinoma but not in squamous cell carcinoma [20]. Additionally, in both mesothelioma, a rare and deadly cancer, and esophageal squamous cell carcinoma, POLD1 has been found to play a significant role in resistance to platinumbased chemotherapy [85,86].
In papillary thyroid cancer, low POLD1 expression implies poor clinicopathological characteristics [21]. Recently, POLD1 protein level was shown to be negatively correlated with the expression of cadherin-16 in this cancer [87].
In gastric adenocarcinoma, POLD1 mutation occurred in 2.77% of 613 patients [88]. Zhu et al. have also shown that using immunotherapy in POLE/POLD1-mutated tumors may have potentially positive implications [88].
Comprehensive molecular profiling has identified the presence of POLD1 mutations in brain tumors [62,[89][90][91]. Additionally, POLD1 expression in this heterogeneous group of neoplasms was only described in 1p19q co-deleted lower-grade gliomas [24]. Yet, the clinical significance of POLD1 mutations and changes in the POLD1 protein level in central nervous system tumors is not fully understood.
Studies concerning uveal melanoma, the most common primary intraocular malignancy in adults, have discussed the changes in POLD1 expression patterns within patients with this cancer [22,23].
Much research has been conducted to investigate the possible clinical significance of POLD1 in gynecologic malignancies. Besides endometrial carcinoma, cervical and ovarian cancers were also assessed for the importance of POLD1 mutations and expression changes [92,93]. Interestingly, the POLD1 gene was more frequently altered among Chinese women with locally advanced cervical cancer than in Caucasian patients [92]. Furthermore, using cisplatin-resistant and cisplatin-sensitive ovarian cancer cell lines, Xing et al. demonstrated that POLD1 was highly expressed in the platinum-sensitive group, whereas chemoresistant cell lines possessed a reduced expression of POLD1 [93].
Intriguingly, in acute lymphoblastic leukemia (ALL), POLD1 upregulation may promote the relapse of this disease [25]. Thus, its expression could be proposed as the potential diagnostic marker and therapeutic target for the treatment of relapsed ALL [25].
A summary of studies discussing the relationship of POLD1 expression level with clinicopathological characteristics was provided in Table 1.

Conclusions and Future Directions
With cancer incidence rates continuing to rise globally, there is a strong need for the development of new biomarkers and treatment strategies [1]. The genetic mutations and expression profiles in POLD1-mutated cancers have recently gained remarkable attention. Nevertheless, the complex regulatory mechanisms of POLD1 expression in cancers remain to be fully characterized.
Both POLD1 germline and somatic mutations within the proofreading domain are considered rare in a variety of malignancies [9,10,63]. Additionally, the presence of pathogenic POLD1 germline alterations has been linked with some hereditary tumors, such as colorectal and endometrial cancers [55].
Based on the primary role of Polδ in genome maintenance, mutations in the catalytic POLD1 subunit lead to genomic instability, mutator phenotype, and malignant transformation [7,9]. Missense mutations are believed to be the most common mutations in POLD1-associated tumors, and no mutation hotspots have been identified so far [70,84].
It should be noted that POLD1 has been considered haploinsufficient in the literature [9,47,55]. Contrarily, a POLD1 haplosufficiency has been recently revealed in the yeast model [94] and in germline monoallelic POLD1 alteration carriers [95,96]. It has been speculated that a solitary heterozygous POLD1 mutation in the exonuclease domain without concurrent mutations in other repair systems causes only a modest increase in the mutation rate [95]. In addition, Schamschula et al. suggested that POLD1-mutated cancers acquire the ultra-high TMB only with concurrent mismatch repair deficiency [66]. It is hypothesized that there are no cancers with an exclusive heterozygous POLD1 mutation and only a complete loss of the two POLD1 alleles can lead to cancer [95]. Therefore, both germline and somatic POLD1 mutations triggering tumor formation may turn out to be biallelic [95]. Nevertheless, an attempt to organize the current knowledge and deepen understanding of POLD1 biology in carcinogenesis would be of paramount importance and could pave the way for novel anti-cancer drugs.
Whereas an overwhelming number of studies in the last decades have focused on POLD1 pathogenic variants in a wide range of neoplasms, only a few studies have validated the prognostic and predictive utility of POLD1 mRNA and protein expression in human cancers to date. So far, POLD1 expression and its clinical significance have been investigated in colorectal, endometrial, renal, liver, breast, lung, and thyroid cancers [13][14][15]17,[19][20][21]. Nevertheless, these data are mainly limited to single-institution experiences.
In recent years, various immunotherapy-based cancer treatments have been approved for clinical use and many additional drugs are still being developed, thus opening a new era in cancer treatment [97,98]. Moreover, ICI treatment is considered a new standard of care across many cancer indications [97]. There are several biomarkers for predicting the outcome of ICI treatment, among which expression of programmed death-ligand 1 (PD-L1), MSI status, and TMB are the most commonly used [63,97]. Moreover, high levels of TILs (especially the CD8+ T cells subpopulation [99]), common in MSI cancers, may predict the response to ICI treatment in various malignancies [100]. As some ICI recipients fail to respond to the treatment, it is therefore crucial to identify accessory molecular biomarkers for predicting the ICI response to prevent immunotherapy overtreatment [98][99][100][101][102].
Some reports have suggested that POLE/POLD1 mutations and/or POLE/POLD1 expression patterns may be useful to predict immunotherapy effectiveness [10,70]. Previous studies investigating the association of POLE/POLD1 mutation with MSI status in various cancers have shown ambiguous results. Although cancers harboring POLD1 mutation are believed to be primarily microsatellite stable, some of them may display an MSI-H phenotype [55,[63][64][65][66][67][68]. Wang et al.'s study on a cohort of almost 50,000 patients with different solid cancer types assessed the impact of POLE/POLD1 mutations on immunotherapy outcomes [10]. Their analysis demonstrated that among ICI-treated cancer individuals, patients with either POLE or POLD1 mutations have significantly longer overall survival in comparison with those without [10]. Notably, the ICI treatment outcomes were similar, regardless of whether these mutations were located within the exonuclease domain or outside of it [10]. Regrettably, POLE/POLD1 alterations cannot be independent factors for predicting the benefit from immunotherapy [101]. In addition, a study conducted by He et al. suggested that POLE/POLD1 mutations in the proofreading domain may result in DNA repair defects and an exceptionally high TMB, thereby generating high neoantigen load levels that, in turn, positively correlate with the sensitivity to ICI treatment [63]. Moreover, tumors expressing higher numbers of neoantigens are associated with a better response to immunotherapy in patients with CRC, melanoma, and NSCLC [63]. However, little is known about the influence of POLD1 expression level on the efficacy of ICI in cancer treatment.
In the last decade, whole-genome sequencing identified specific mutational signatures that describe the patterns of mutations that arise during tumorigenesis [103,104]. Several single-base substitution (SBS) signatures have been described concerning POLD1-mutated tumors; SBS10c and SBS10d are linked to defective POLD1 proofreading [105], whereas SBS20 is associated with concurrent POLD1 mutations and mismatch repair deficiency [65]. Groundbreaking studies have shown that the presence of specific SBS signatures may have prognostic and therapeutic implications [106]. Notably, Ma et al. reported that POLD1 functional mutations generating SBS signatures associated with POLD1 activity lead to augmented immunogenicity [107]. Moreover, patients with cancers generating POLD1mutation-related SBS signatures demonstrated an enhanced response to ICI treatment [107].
Given the variety of immuno-oncology treatments currently in use for therapies in patients with different cancer types, including melanoma, NSCLC, CRC, and HCC [97,108], novel research examining the relationship between the expression of the POLD1 protein and the clinical outcomes in patients receiving immunotherapy would be of great importance. Several clinical trials are ongoing to evaluate the impact of POLD1 mutation on treatment outcomes (NCT05103969, NCT03810339, NCT03428802, NCT03491345, NCT04969029).
With the advancement of knowledge regarding cancer biology, substantial progress in the development of new diagnostic and prognostic molecular biomarkers as well as targeted therapeutic strategies against oncological diseases can be expected within the next few years.