Letter to the Editor: From colorectal to pan-gastrointestinal cancer: Unlocking the therapeutic potential of DNA polymerase epsilon mutations

Abstract

A recent investigation by Taskiran et al published in the World Journal of Gastroenterology on DNA polymerase epsilon (POLE) mutations in colorectal cancer prompted a broader discussion on the role of pathogenic variants as pan-gastrointestinal therapeutic biomarkers. Pathogenic exonuclease domain mutations (EDMs) in POLE have been identified in other digestive system malignancies including gastric, pancreatic, biliary, and hepatocellular carcinomas. Although rare, POLE EDMs consistently confer a hypermutator phenotype and high tumor mutational burden, irrespective of the tissue of origin. This shared biological consequence serves as a powerful predictor of response to immune checkpoint inhibitors, offering a vital, tissue-agnostic therapeutic option for patients with advanced or refractory diseases with established tumor mutational burden-high approvals. Therefore, to ensure that these rare but highly actionable opportunities are not missed, broader implementation of comprehensive genomic profiling, including POLE analysis, is warranted for all patients with advanced pan-gastrointestinal cancers.

Key Words: DNA polymerase epsilon; Pan-gastrointestinal cancer; Immune checkpoint inhibitors; Tumor mutational burden; Biomarker; Mismatch repair deficiency

Core Tip: To maximize the identification of immunotherapy candidates in gastrointestinal cancer, we advocate for a paradigm of concurrent mismatch repair and DNA polymerase epsilon testing. This dual strategy provides a crucial second pathway to therapy for patients with proficient mismatch repair tumors who harbor pathogenic DNA polymerase epsilon mutations, ensuring these “hidden hot tumors” are not overlooked by conventional screening.

TO THE EDITOR

We read with interest the article by Taskiran et al[1] published in the World Journal of Gastroenterology. Their work advanced the discourse from simple variant annotation to meaningful clinical stratification by linking different polymerase epsilon (POLE) mutation types to specific molecular profiles. Although the reported POLE prevalence of 53.65% is striking, it is largely driven by a single variant (p.V1446fs*3) with conflicting pathogenicity classifications. This crucial detail encourages a broader perspective positioning pathogenic POLE mutations as rare but clinically actionable biomarkers across the gastrointestinal (GI) landscape. Building on this, our letter highlights the varying frequencies of these pathogenic variants across GI cancers, summarizes compelling clinical evidence for their use as predictive biomarkers, and proposes a necessary evolution in clinical testing.

Frequency and biological rationale of POLE mutations across GI cancers

The therapeutic rationale for targeting POLE is based on its fundamental role in maintaining genomic fidelity. As the catalytic subunit of DNA POLE, the exonuclease domain is critical for proofreading during DNA replication. Pathogenic exonuclease domain mutations (EDMs) abrogate this function, leading to profound accumulation of somatic mutations and an ultra-mutated phenotype. This is characterized by an exceptionally high tumor mutational burden (TMB), which in turn generates a diverse neoantigen landscape, heightening tumor immunogenicity and fostering an inflamed, or “hot”, tumor microenvironment permissive to robust T-cell-mediated anti-tumor responses[2,3]. Notably, the reported frequencies of pathogenic POLE mutations vary across studies, which may reflect differences in the sequencing platforms, gene panel composition, depth of coverage, variant classification criteria, and population-specific factors. Despite this variability, pathogenic EDMs have consistently been identified in a biologically distinct hypermutated subgroup of GI cancers.

Figure 1 summarizes the frequency and distribution of POLE mutations in patients with major GI malignancies. Although POLE alterations are detected across tumor types, pathogenic EDMs constitute only a small subset of all identified POLE variants. In most GI cancers, most POLE alterations are located outside the exonuclease domain or are classified as non-pathogenic, with pathogenic mutations occurring at consistently low frequencies. The prevalence of these pathogenic variants in advanced colorectal cancer (CRC) is well characterized, albeit low. One study reported a hotspot mutation frequency of 0.65% (2/307)[4]. Similarly, a Korean cohort found pathogenic POLE mutations in 1.1% (2/181) of patients, which were linked to a TMB > 200 mut/Mb, distinguishing them from the more common non-pathogenic variants found in the same study[5].

Figure 1 Pan-gastrointestinal landscape of somatic DNA polymerase epsilon mutations. Schematic overview of the frequency and classification of somatic polymerase epsilon gene mutations across major digestive system cancers. For each organ, the figure displays the total number of samples analyzed (n). Mutations are classified based on their location relative to the proofreading region of the gene: Those located within this region are termed exonuclease domain mutations (EDMs), while those located elsewhere are termed non-EDMs. The esophagus data specifically represents esophageal adenocarcinoma. Specific variants are color-coded by their classification: Non-EDMs (green), EDMs (red), and driver EDMs (blue). Data were summarized from a comprehensive cancer genomics database. EDMs: Exonuclease domain mutations.

In gastric cancer, the prevalence of POLE mutations that confer the ultramutator phenotype is notably low. This is exemplified by a mixed cohort where, although 6 of 47 POLE-mutant tumors were gastric, none exhibited the characteristic hypermutation, suggesting that functionally significant variants are particularly infrequent in this cancer type[6]. This trend of diminishing frequency continues for hepatobiliary malignancies. Pathogenic POLE mutations are exceptionally rare in advanced hepatocellular carcinoma (HCC), with a reported prevalence of less than 1% (1/755). While a broader 4% of HCC cases harbor alterations in POLE/POLD1, the vast majority are classified as variants of unknown significance[7]. Their presence has also been noted, albeit rarely, in specific subtypes such as Epstein-Barr virus-associated cholangiocarcinoma[8]. The nadir of this frequency was reached in pancreatic ductal adenocarcinoma, where the prevalence is exceedingly low. This was clearly illustrated in a dedicated study of 115 advanced pancreatic ductal adenocarcinoma samples that failed to identify hotspot mutations, reporting a frequency of 0%[9]. However, it is crucial to note that these events are not entirely absent from the pancreas; a pathogenic p.Val411 Leu mutation has been identified in a rare case of medullary pancreatic cancer, a distinct histopathological subtype, confirming that these mutations can occur and may be clinically relevant[10].

A unified therapeutic strategy: Immune checkpoint inhibition in pole-mutant cancers

The immunological consequences of pathogenic POLE mutations provide a compelling rationale for a unified therapeutic strategy, namely, immune checkpoint inhibition. The resultant hypermutated, immunologically “hot” tumor microenvironment is rendered highly susceptible to immune checkpoint inhibitors (ICIs), which function by releasing key immune checkpoints such as programmed death-1/cytotoxic T-lymphocyte antigen-4 to potentiate an anti-tumor T-cell response[2,3]. This principle is clearly illustrated in gastric adenocarcinoma, where a comprehensive analysis by Zhu et al[11] demonstrated that POLE/POLD1 mutations are associated with a constellation of positive predictive biomarkers, including significantly higher TMB, elevated programmed death ligand-1 expression, and an inflamed type I tumor microenvironment, thereby identifying them as prime candidates for therapy with ICIs.

The clinical actionability of this biomarker is validated by the United States Food and Drug Administration’s pan-cancer approval of pembrolizumab for TMB-high (≥ 10 mut/Mb) solid tumors, a molecular signature that pathogenic POLE-mutant cancers exemplify[12]. Clinical data from POLE-mutant cohorts further substantiated this approach. For instance, a phase II trial of toripalimab demonstrated a disease control rate exceeding 50% in a CRC subgroup of patients with POLE/POLD1-mutant advanced solid tumors[6]. This therapeutic benefit is not exclusively confined to classic EDMs. A remarkable complete response to pembrolizumab has also been reported in a patient with recurrent HCC harboring a non-exonuclease domain POLE mutation and low TMB, highlighting the broad potential of this biomarker[13].

Furthermore, a large pan-cancer analysis quantified the profound survival advantage, revealing that patients with pathogenic POLE/POLD1 mutations achieved a significantly longer median overall survival with ICIs than their wild-type counterparts (34 months vs 18 months)[14]. Consequently, this robust body of evidence has been integrated into clinical practice, with established guidelines recommending POLE/POLD1 mutation testing for CRC patients who have progressed to standard therapies[15].

Conclusions

In conclusion, pathogenic POLE mutations define a distinct molecular subset of GI cancers, characterized by a hypermutated, immunologically active tumor microenvironment. Conventional deficient mismatch repair (MMR)/microsatellite instability (MSI) testing does not identify this population, which may result in missed opportunities for ICI therapy. To address this gap in patient selection, we propose implementing a concurrent MMR/MSI and POLE testing paradigm. This integrated approach established two complementary pathways for ICI eligibility: The established deficient mismatch repair/MSI-H route and an additional pathway for proficient MMR/microsatellite stable tumors harboring pathogenic POLE mutations. Although comprehensive genomic profiling may raise concerns regarding cost and accessibility, POLE assessment is increasingly being incorporated into routine next-generation sequencing panels used in advanced GI cancers. In this context, concurrent POLE and MMR/MSI evaluations represent an extension of existing diagnostic workflows, rather than an additional standalone test. The adoption of this dual-testing strategy may facilitate the identification of immunotherapy-sensitive “hidden hot tumors” within the proficient MMR/microsatellite stable population and expand access to precision immunotherapy for a broader group of patients.