The role of the DNA polymerase epsilon (POLE) in the correct replication of cellular DNA has been studied intensively (1). As a consequence of this research, the prognostic value of specific mutations in the POLE gene, so called “hotspot mutations,” has recently recognized, and this has revolutionized the management of endometrial and other cancers (2–4). A single missense mutation in a hotspot region of the gene can guide a clinician to reconsider the need for adjuvant therapy in cases, which, until recently, would have received treatments known to be associated with a high risk of complications (3). POLE mutant tumors are described in endometrial, ovarian, colorectal, and urological cancers (5–8). They lead to a ultramutated tumor phenotype and consistently demonstrate an excellent clinical outcome, especially in colorectal cancer (CRC) and endometrial cancer (EC) (5, 9).

Historically, standard treatment of EC consisted of hysterectomy, bilateral salpingo-oophorectomy, and pelvic lymph node dissection followed by adjuvant therapy in the form of radiotherapy and/or chemotherapy based on final histology (10). However, management of EC has become more patient-specific over the past 10 years: differences in the histo-molecular classification predict prognosis and dictate whether adjuvant therapies are required or to be avoided (10) ( Figure 1 ).

In 2013, The Cancer Genome Atlas (TCGA) identified a novel subgroup of ECs with unique mutations in POLE and associated unfavorable histomorphological features but, nevertheless, showing good survival outcomes. This new histo-molecular group was studied in the PORTEC-3 trial for patients with high-risk EC with Fédération International de Gynécologie et d’Obstétrique (FIGO) stage I–III, which investigated the benefit of adjuvant chemotherapy during and after radiotherapy over pelvic radiotherapy alone. The results of this trial showed that addition of adjuvant chemotherapy showed improved 5-year overall survival, especially in stage III patients. Crucially, molecular profiling revealed that POLE-mutated patients showed superior outcome, especially in stages I and II, irrespectively of their adjuvant treatment. This led to the author’s recommendation of de-escalation of adjuvant therapy for POLE-mutated patients (11, 12). These findings are limited by the relatively low number of POLE-mutated patients (12%) in general, and the fact that patients with stage III tumor disease showed superior outcome, if they had received a combined treatment. Furthermore, stage IV patients were not included in this trial. In 2016, the TCGA molecular classification was integrated into the ESMO (European Society for Medical Oncology) guideline, and this resulted in an updated risk classification for recurrence in stage I POLE-mutated ECs (13). The subgroup of POLE-mutated tumors was further integrated into the European Society of Gynecological Oncology (ESGO)/European Society for Radiotherapy and Oncology (ESTRO) and European Society of Pathology (ESP) guidelines for EC, resulting in specific changes in the recommendations for adjuvant treatment (3). The ESGO/ESTRO/ESP guidelines recommended that omission of adjuvant treatment should be considered for patients with stage I–II POLE-mutated EC. For the rare patients at stage III–IVA ECs with pathogenic POLE mutations, there are no reliable data on survival regarding omission of adjuvant treatment (3). In CRC as well, genetic testing for POLE was incorporated into the National Comprehensive Cancer Network guidelines in 2022 (14). Recently, this molecular classification has been integrated into the FIGO staging classification system of 2023. This shows the high clinical impact achieved by molecular characterization, including testing for POLE mutation, for optimal treatment of patients with EC. Although POLE mutant tumors tend to have a favorable outcome, there remain a number of questions to be answered: What is the physiological function and role of Pol ϵ? Is the good prognosis linked to the high mutational burden of POLE-mutated tumors? How is the correct annotation of the “POLE hotspot” mutations achieved? Is each POLE mutation pathogenic? What screening methods are available to determine POLE pathogenicity? What is the role of immunotherapy in POLE-mutated tumors and do POLE variants impact other tumor entities comparably? This review aims to discuss these fundamental questions and to highlight the current controversies related to this topic.

Accurate replication of DNA prior to cell division is essential for maintaining genomic stability and for suppressing mutagenesis and tumor development (15). The high fidelity of eukaryotic DNA replication is due to a combination of highly accurate base incorporation and 3′-5′ exonuclease proofreading by the replicative DNA polymerases Pol δ and Pol ϵ and post-replication surveillance of the newly synthesized DNA by the mismatch repair (MMR) apparatus (5). In humans, Pol ϵ belongs to the B family polymerases, comprises four subunits, and is encoded by POLE (5). The proofreading function of Pol ϵ requires highly conserved motifs in their exonuclease domain (EDM), named exo-motifs, within which lie the catalytic site residues that are essential for exonuclease activity (5, 15). Misincorporation of a base into the leading strand leads to pausing of the polymerase ϵ and, consequently, to a switch from the catalytic site to the exonuclease domain, where the incorrect base is excised and replaced by the correct base (16).

Considering the close correlation between POLE mutations and increased mutation rates, it is important to define the tumor mutational burden (TMB) when describing tumor biology. TMB indicates the number of single mutations per megabase presented in the specimen. A high TMB describes a highly mutated tissue, which could be considered to have more aggressive biological behavior. More specifically, hypermutation is defined by a mutational load of 10 or more mutations per megabase (≥10 mut/Mb) (17), whereas ultramutated tumors show a frequency of base substitution mutations that is equal to or higher than 100 mutations per megabase (ultramutation ≥100 mut/Mb) (1).

Given how the presence of a POLE-EDM mutation impacts the outcome and especially the therapeutic approach in EC, this raises the question—does this impact and approach generalize to other tumor entities? The oncological literature on this topic is expanding rapidly, and, almost in every field, there is an effort to identify possible pathogenic POLE-mutated variants.

Evaluation of a pathogenic POLE mutation remains challenging, as parameters and methods that allow a standard procedure in clinical practice have not been validated as yet. Whole-exome or whole-genome sequencing (WES/WGS) by Sanger or next-generation-sequencing can be used to identify POLE mutations in the exonuclease domain (exons 9–14). However, these methods are time-consuming, not widely available, and expensive; require expertise; and, therefore, limit routine use in current clinical practice. Estimation of pathogenicity of somatic POLE mutations in the absence of exome and genome sequencing has been carried out by some authors by using in silico prediction tools (1). Although this seems to be a feasible technique, the setting prognosis relies on the sequencing tools that have been used. Clinical practice requires a POLE testing method that is not only affordable, with a fast turnaround time, but also easy to interpret and implement. Therefore, sequencing methods restricted to the analysis of the hotspot POLE exonuclease domain mutations could present an alternative technique and have been developed recently by several research groups. Deveraux et al. (37, 38), for example, use a single-gene POLE hotspot SNaPshot assay in their routine prospective molecular classification of ECs. This technique involves an initial PCR amplification of the relevant gene target regions of the POLE-EDMs, followed by multiplexed single-nucleotide primer extension (38). Van den Heerik et al. (39) created a quantitative polymerase chain reaction (qPCR) assay for pathogenic POLE mutations (QPOLE). So far, there is no standardized method that allows determination of POLE mutations in ECs in clinical practice. Moreover, a workflow that reports both the molecular and histologic findings in an integrative manner is still not available. However, the integration of molecular classification together with clinicopathologic features into the ESGO/ESTRO/ESP guidelines and into the novel 2023 FIGO staging classification system shows the high clinical impact that testing of POLE mutation has in the patients’ treatment and management. This issue cannot be ignored any longer. It is critically necessary that the clinical assay used in daily practice reliably identifies POLE mutations in the hotspot POLE-EDM as their role in tumor biology and their therapeutical consequences are known. However, as new hotspot POLE-EDM mutations continue to emerge, their clinical role must be rapidly evaluated, and they should be incorporated into validated assays as appropriate. One indirect approach to identify POLE-mutated cancers is by looking at the number of tumor-infiltrating lymphocytes (TILs), as we know that a highly mutated microenvironment expresses more antigens and, therefore, activates the host’s immune system. For example, in CRC cases, MSI tumors were TIL-high (≥4 lymphocytes per high-power field), in 68% of cases with a TMB of 54 mut/Mb, whereas MSS CRCs were only TIL-high in 4.5% of cases. In contrast, MSS CRC tumors with POLE/POLD1 pathogenic variants were TIL-high in 82% of cases and had a TMB over 150 mut/Mb. These differences in tumoral immunity provide the rationale for immunotherapy (40). A possible way to screen for patients who might benefit from immunotherapy even if they are MMR-intact could be by using an immune microenvironment evaluation system such as that described by Galon et al. (41). In CRC, the Immunoscore (IS) has been shown to be a prognostic factor superior to the previous tumor, node, metastasis (TNM) classification of malignant tumors. There are ongoing validation and promotion initiatives to increase the use of IS in routine clinical settings (41). There are several ongoing clinical trials assessing the efficacy of ICI therapy for treating patients with POLE/POLD1 mutations, especially for metastatic CRC (40, 42–44). Different systems of immunoscoring have also found application in gastric and endometrial cancer (45), but further prognostic studies are needed to validate the routine use of ISs in routine clinical practice.

In EC, the ultramutated phenotype caused by POLE-EDM mutations has been shown to cause a “self-limiting” tumor progression with excellent prognosis after surgery. This has been demonstrated even without adjuvant treatment in patients previously classified as “high” or “intermediate-high” risk (10, 46). The Identification of pathogenic POLE mutations in early-stage EC also plays a crucial role when considering fertility sparing treatments (FSTs) such as hormonal therapy or hysteroscopic resection in young women (47, 48). Improving risk stratification for FST is one of the future targets, because the molecular classification and different molecular markers emerging are changing the risk profile assessment for patients. A preliminary molecular analysis of an endometrial biopsy is, therefore, necessary in patients with desire to conceive or in case of organ sparing. Pathologists should systematically perform a molecular analysis of hysteroscopic biopsy samples, including the sequencing for the presence of the POLE mutation. In OC of the endometrioid histological subtype, there is reason to believe that POLE mutations lead to a better prognosis, but further confirmation is needed (49). Although there are many molecular studies on POLE-mutated CRCs, data on the clinical implications of the POLE ultramutated phenotype are lacking. In some studies of POLE-EDM–mutated stage II CRCs, a robust intratumoral T-cell response was detected in a small subset of cases with excellent outcomes (50). However, a significantly better prognosis across all stages of CRC has not been confirmed as yet (as in the case of ECs). Therefore, the therapeutic management of CRC is currently not impacted by the presence of a POLE mutation. In other solid tumors such as pancreatic cancer, mutations in the hotspot regions of POLE are very rare events. In advanced pancreatic cancer, it is highly unlikely that POLE mutations contribute to genetic instability; therefore, POLE mutations do not serve as a relevant biomarker and should not be tested for on a regular basis (51). In a totally different oncologic entity, namely, high-grade gliomas (HGGs), a subgroup based on somatic POLE mutations has been identified. Such cases are genomically, histologically, and clinically different from the other HGGs and exhibit an improved prognosis (52). Most recently, two trials have clarified the role of ICI for the treatment of EC. These two big randomized phase III trials (RUBY and GY018) with still limited follow-up did so far not change the practice for POLE-mutated EC (53). Over the last several years, an enormous effort has been made with regard to EC to find new biomarkers that can accurately identify patients who can benefit from immunotherapy even if they are not MMR deficient. In silico analysis has proved that POLE mutant cancers display more antigenic neoepitopes than other ECs, providing a potential rational for POLE immunogenicity (54). Yet POLE-mutated tumors are currently not a recommended indication for ICIs even in advanced and metastatic EC cases. On the other hand, CRCs have recently received approval for treatment of MSI CRC with ICIs. Assessment of POLE status may help guide therapeutic decisions for tumors with high TMB and intact MMR. Recent reports have shown that, even in advanced CRC or multiresistent disease, patients can significantly benefit from ICIs, if they harbor a pathogenic POLE mutation (55–57). A case report of a high-grade CRC with POLE-EDM (P286R) mutation and TMB of 119 mut/Mb described triple-chemotherapy being ineffective, whereas ICIs had a significant impact on progression-free survival (PFS) (55). Another example involved the case of a 24-year-old male patient with an aggressive stage IV high-grade, poorly differentiated CRC, where response was complete and durable (over 48 months) with a single-agent ICI after rapidly progression with standard chemotherapy. Genetic testing of this case revealed a P286R POLE mutation and an elevated TMB of 126 mut/Mb (57). These cases highlight the interplay between genetic instability and immune-checkpoint blockade. In a case report of OC (OCCC inoperable at stage IIIB), resistant to platinum-based chemotherapy, the same P286R POLE mutation was found, and the third-line treatment attempt with a programmed cell death protein 1 (PD-1) inhibitor showed a tumor with postoperative pathologic complete response. The patient achieved a PFS of 29 months under maintenance with ICI therapy (58). Therefore, even if the ProMisE classification does not find an application in OC, the simple presence of POLE hotspot mutation can, in exceptional cases, guide for treatment with ICIs.

At the present, it is important to acknowledge that the changes in clinical treatment of POLE-mutated ECs as recommended by the ESGO/ESTRO/ESP guideline of 2021 are based on much less data than what is typically used for such profound and nearly dogmatic shifts in clinical care. Evaluation of the benefit of these clinical and therapeutic changes through prospective studies can help provide more information on this issue. Currently, prospective clinical trials, such as PORTEC-4 and TAPER, are ongoing and will shed light and yield more insights whether POLE-mutated ECs have a favorable outcome even without or de-escalated adjuvant treatment. The TAPER trial is an interventional study based on tailored adjuvant therapy in POLE-mutated and p53 wild-type/NSMP early-stage EC. Its primary objective is to determine if women with cancers with the specific molecular characteristics who underwent adequate surgery have a relative low risk (lower than 5%) of pelvic and vaginal recurrence at 3 years with no or de-escalated adjuvant treatment (59). The PORTEC-4a trial is a randomized phase III trial of molecular profile-based versus standard recommendations for adjuvant radiotherapy in stage I EC and is focused on cancers classified as high-intermediate risk according to the ESMO-ESGO-ESTRO consensus of 2015. The primary endpoint is vaginal recurrence, and other oncologic but secondary endpoints are recurrence-free and overall survival, as well as pelvic and distant recurrence. As, in this study, all patients with POLE-mutated cancers are allocated without restrictions to the “favorable molecular risk group” with omission of vaginal brachytherapy and external pelvic beam radiotherapy, this trial is very likely to extend our knowledge on prognosis of POLE-mutated cancers at least in this prespecified subset of patients. First results are expected at the end of 2024 or in the spring 2025, and study completion is set for 2028 with 550 patients enrolled (60). Lastly, the blue arm of the RAINBO trial is also focused at POLE-mutated ECs. The RAINBO program is a platform of four international clinical trials and an overarching research program, including a randomized PHASE III trial with three arms for p53-abn EC (red), MMRd EC (green), and NSMP (orange) ECs. The POLEmut-BLUE trial is a phase II trial in which the safety of de-escalation of adjuvant therapy is investigated for women with stage I–III POLEmut EC. This trial will evaluate no adjuvant therapy for lower-risk disease and no adjuvant therapy or radiotherapy alone with omission of concomitant chemotherapy for higher-risk disease. The primary endpoint of this trial will be pelvic recurrence at 3 years. This study is in the recruiting phase and main trial results are expected in 2028 (61).

A summary of relevant concluded or ongoing trials about ICIs and adjuvant treatments in EC based on molecular classification is shown in Table 5 .

POLE mutational status in EC is of great clinical interest. It determines the prognosis of the patient, and the FIGO classification system 2023 stipulates that its presence should result in a significant de-escalation of adjuvant treatment. In the future, unresolved questions will be better answered by the results of the POLEmut-BLUE arm of the prospective phase II RAINBO trial, where even stage III patients are included. POLE-mut ECs are assigned depending on risk status either to an observational arm with complete omission of adjuvant treatment or to radiotherapy alone. We describe in detail that the sole presence of a POLE variant is not sufficient to classify a tumor as POLEmutated or to classify a POLE mutation as pathogenic. For that, the exact localization of the mutation in the POLE gene needs to be known. It has a higher probability of being pathogenic when it is located in the EDM of Pol ϵ. However, unanswered questions, such as the exact molecular pathways associated with the good prognosis of POLE mutations, remain unclear and, therefore, unmentioned. Due to inconsistencies in POLE mutation testing and its interpretation, the study investigators have advocated a concomitant TMB determination in order to underscore the pathogenicity of the POLE-EDM mutation. Performed as a routine procedure, this appears to be one of the best and easiest approaches to identify new pathogenic POLE mutations. We are hopeful that the increasing knowledge on the exact oncologic driver qualities of POLE mutations together with the outcome of mentioned prospective clinical research will enable reassured avoidance or at least de-escalation of adjuvant treatment in EC without harming patients by under- or overtreatment.