Faculty, Staff and Student Publications
Language
English
Publication Date
11-18-2025
Journal
The Journal for ImmunoTherapy of Cancer
DOI
10.1136/jitc-2025-012190
PMID
41253491
PMCID
PMC12636876
PubMedCentral® Posted Date
11-18-2025
PubMedCentral® Full Text Version
Post-print
Abstract
Background: Only a subset of polymerase epsilon (POLE) mutations is associated with hypermutant phenotype; we hypothesized that only loss-of-proofreading (LOP) POLE mutations are associated with favorable immunotherapy response.
Methods: This retrospective cohort study included a pan-cancer cohort of 69,223 patients from cBioPortal and a cohort of patients with 41 POLE mutant metastatic colorectal (CRC) treated with immunotherapy at the MD Anderson Cancer Center between January 2017 and May 2023. We evaluated prognosis according to POLE mutation functionality.
Results: In the pan-cancer cBioPortal cohort (n=69,223) POLE was mutated in 2.8% (1,965) of tumors; of these, only 7.5% (n=148) had LOP POLE mutations (0.0% of entire cohort). Endometrial cancer (6.6%) and CRC (1.2%) were the only tumor types with greater than 1% incidence of LOP POLE mutation. Overall survival was similar between non-LOP POLE and POLE wildtype patients (HR=0.94, 95% CI 0.82 to 1.07, p=0.34); conversely, LOP POLE patients had significantly better outcomes (HR=0.23, 95% CI 0.16 to 0.32, p< 0.0001). In the clinical cohort, all nine patients with LOP POLE mutations achieved durable clinical benefit (objective response rate (ORR) 88.9%, complete response rate (CR) 33.3%, disease control rate (DCR) 100%) and median progression-free survival (PFS) was not reached at the time of analysis, after median follow-up of 32 months. None of the nine patients with microsatellite stable (MSS), non-LOP POLE tumors achieved an ORR of 0%, with median PFS of 3.7 months (HR 0.05 LOP relative to non-LOP POLE, 95% CI 0.01 to 0.19, p< 0.0001). Interestingly, median PFS on first-line cytotoxic agents was significantly shorter for patients with LOP POLE compared with patients with MSS non-LOP POLE mutant tumors (2.1 vs 9.7 months, HR 3.33, 95% CI 0.87 to 12.74, p=0.012).
Conclusions: Identifying the subset of POLE mutations that cause LOP is critical to distinguish patients likely to respond to immunotherapy. Patients with CRC with LOP POLE mutant tumors experienced deep, sustained response to immunotherapy but were resistant to standard cytotoxic chemotherapy, in stark contrast to those with non-LOP POLE mutations.
Keywords
Humans, Female, Male, Prognosis, Retrospective Studies, Middle Aged, DNA Polymerase II, Aged, Poly-ADP-Ribose Binding Proteins, Immunotherapy, Adult, Neoplasms, Mutation, Aged, 80 and over, Immune Checkpoint Inhibitor, Biomarker, Colon Cancer
Published Open-Access
yes
Recommended Citation
Maddalena, Giulia; Zeineddine, Fadl A; Chowdhury, Saikat; et al., "Prognosis and Treatment Response Stratification According to Loss of Proofreading (LOP)" (2025). Faculty, Staff and Student Publications. 6009.
https://digitalcommons.library.tmc.edu/uthgsbs_docs/6009
Included in
Bioinformatics Commons, Biomedical Informatics Commons, Genetic Phenomena Commons, Medical Genetics Commons, Oncology Commons