Correction: Tumor-immune microenvironment and nutritional status biomarkers predict recurrence after total colectomy for locally advanced colon cancer: development and internal validation of a nomogram

Abstract

Query no.Author's response as the author, to ensure you have permission to use the images in the article.Please check this link for information on author responsibilities and the publication images. �Please provide the names of the remaining authors in becht et al., 2012. Frontiers guidelines require to list at least 6 authors, + et al. for more than 6 authors.The reference " Becht et al., 2012, Guo et al., 2026", is missing a DOI, or the DOI provided is invalid or links to an incorrect article. Please check this reference and provide a working DOI, if applicable. Invalid DOIs will not be added to the references. References without DOIs will still be included. � Q17 Cite the reference " André et al., 2024;Becht et al., 2012;Dienstmann et al., 2017;Fridman et al., 2012;Galon et al., 2006;Guo et al., 2023;Guo et al., 2026;Khare et al., 2025;Le et al., 2015;Liu et al., 2024;Ogino et al., 2011;Overman et al., 2018;Sinicrope et al., 2015;Venderbosch et al., 2014;Xiu et al., 2024;Xu et al., 2024" inside the text. �If you decide to use previously published and/or copyrighted figures in your article, please keep in mind that it is your responsibility as the author to obtain the appropriate permissions and licenses to reproduce them, and to follow any citation instructions requested by third-party rights holders. We require attribution for any elements where you are not the original copyright holder. If obtaining the reproduction rights involves the payment of a fee, these charges are to be paid by the authors. Please provide us with the following attribution details in each case: [Citation], [License], and [Source], and confirm whether you have reproduced or adapted the material. Third-party material includes any proprietary text, illustrations, figures, tables, images, artworks, datasets, databases, as well as trademarked logos or brand names, and any material taken from websites, such as: screenshots, photographs, illustrations, maps, icons, clipart, tables, or social media posts*. *Note: Social media posts require permission from the original user/poster before they can be reproduced. � Q19 For Figure(s) 1-4, is of low quality/distorted. Therefore, please provide higher resolution/corrected images. If your figure is mainly text and lines, you can upload a vector file for the best quality (EPS or PDF, which can be uploaded as Related Article 1)Please confirm that the below mandatory Frontiers AI generated Alt-Text is an accurate visual description of your Figure (s). As part of our open science commitment, Alt-Text will enhance the accessibility of your manuscript. It is a short visual description that allows people using screen reading technology to clearly understand the contents of an image. These Figure Alt-text proposals won't replace your figure captions and will not be visible on your article. If you wish to make any changes, kindly provide the exact revised Alt-Text you would like to use, ensuring that the word-count remains at approximately 100 words for best accessibility results. Further information on Alt-Text can be found here. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.Frontiers in Genetics frontiersin.org 01 Colorectal cancer ranks third in incidence and second in cancer mortality globally, with colon cancer accounting for the predominant burden of new cases (Roshandel et al., 2024). For patients with locally advanced disease requiring total colectomyincluding those with multi-segment involvement, synchronous lesions, or extensive nodal disease-the risk of post-surgical recurrence remains substantial, with 5-year recurrence rates of 25%-45% in stage III disease despite adjuvant chemotherapy (Benson et al., 2021). The emergence of immune checkpoint inhibitors (ICIs) targeting the PD-1/PD-L1 axis has transformed the therapeutic landscape: pembrolizumab demonstrated superior progression-free survival versus chemotherapy in deficient mismatch repair (dMMR)/microsatellite instability-high (MSI-H) metastatic colorectal cancer (KEYNOTE-177), and adjuvant nivolumab demonstrated significant DFS benefit in dMMR/MSI-H stage II-III resected colon cancer (CheckMate 89Q) (André et al., 2020;Cercek et al., 2022). However, the approximately 80% of patients with proficient MMR (pMMR)/microsatellite stable (MSS) disease derive minimal benefit from single-agent anti-PD-1 therapy, constituting the dominant unmet clinical need in adjuvant colon cancer management (Overman et al., 2017).The tumor-immune microenvironment (TIME) is a highly dynamic ecosystem in which tumor-intrinsic gene mutationsparticularly in the MMR pathway-directly shape the immunological composition and functional state of the surrounding immune infiltrate. dMMR/MSI-H tumors generate abundant frameshift-derived neoantigens that drive dense TIL infiltration, high PD-L1 co-expression, and an ICI-responsive phenotype (Galon et al., 2014). In contrast, pMMR/MSS tumors typically exhibit low TIL density, an immune-excluded or immunedesert phenotype, and constitutive resistance to PD-1/PD-L1 blockade (Havel et al., 2019). Beyond these tumor-intrinsic immune gene mutations, systemic indices of immune activation and suppression-including the neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR)-reflect the host immune landscape at the tumor-systemic interface and have been associated with outcomes across multiple solid tumors (Templeton et al., 2014). The integration of these multi-dimensional tumorimmune biomarkers into a unified, clinically validated prediction model remains an unmet need for post-colectomy recurrence stratification and adjuvant ICI patient selection.Existing risk stratification tools for colon cancer-including TNM staging and IDEA/ACCENT consortium models-are based exclusively on clinicopathological variables and do not incorporate tumor-immune gene mutation status, TIME characterization, or perioperative nutritional assessment (Taieb et al., 2018). This limitation is clinically consequential: two patients with identical stage III pathology may have radically different tumor-immune interfaces-one with an ICI-responsive dMMR phenotype and dense TIL infiltration, the other with a pMMR, NLR-high, TIL-low, immunologically cold phenotypeyet receive identical adjuvant regimens. Perioperative malnutrition further compounds this heterogeneity: hypoalbuminaemia and elevated NRS 2002 scores reflect both systemic inflammatory burden and impaired immune effector capacity, independently contributing to compromised anti-tumor immunity and attenuated responses to adjuvant therapy. The ability to integrate MMR status, TIL density, PD-L1 CPS, NLR, nutritional status indices, and conventional clinicopathological factors into a validated nomogram would enable biomarker-guided stratification that identifies both recurrence risk and immunotherapy response likelihood, directly addressing the Research Topic's goals of discovering novel biomarkers at the evolving tumor-immune interface that predict response, resistance, and guide immunotherapeutic strategies.This retrospective clinical cohort study was designed to: (1) characterize the tumor-immune biomarker landscape and perioperative nutritional status in 341 patients who underwent total colectomy for locally advanced colon cancer; (2) identify independent tumor-immune, nutritional, and clinicopathological predictors of 3-year DFS; (3) develop and internally validate a nomogram integrating these predictors; and (4) evaluate the model's utility in identifying pMMR/MSS patients who benefit from adjuvant ICI therapy-the most pressing immunotherapy resistance challenge in colon cancer management. MMR status was determined by immunohistochemistry (IHC) for MLH1, MSH2, MSH6, and PMS2 on FFPE tumor sections; dMMR was defined as loss of nuclear expression in ≥1 protein.Equivocal cases were confirmed by PCR-based MSI testing (NovaDx MSI panel; microsatellite instability at ≥2 of five loci = MSI-H). PD-L1 expression was quantified using the combined positive score (CPS) with anti-PD-L1 22C3 pharmDx (Dako Agilent); CPS ≥10 was defined as PD-L1-positive based on trial eligibility criteria (Cercek et al., 2022). Stromal TIL density was evaluated on hematoxylin-eosin-stained slides by two independent pathologists following the International TILs Working Group (ITLR) scoring guidelines; high TIL density was defined as ≥20% stromal TILs. Interobserver reproducibility was assessed in a 20% random subsample (n = 68): kappa for TIL density categorisation (≥20% vs. <20%) was κ = 0.76 (95% CI: 0.62-0.90) and for PD-L1 CPS categorisation (CPS ≥10 vs. < 10) was κ = 0.71 (95% CI: 0.56-0.86). Discordant cases were resolved by consensus review. The NLR was calculated from complete blood count obtained within 2 weeks preoperatively; NLR ≥3.0 was defined as elevated based on published meta-analytic cutoffs (Templeton et al., 2014). PLR ≥150 was defined as elevated. Serum CEA and CA19-9 were measured within 2 weeks preoperatively using electrochemiluminescence immunoassay. All biomarker cutpoints (TIL ≥20%, NLR ≥3.0, CPS ≥10, LNR ≥0.25) were pre-specified from published metaanalyses. Sensitivity analyses using within-cohort ROC/Youdenderived optimal cutpoints (NLR = 3.2, LNR = 0.24, TIL = 18%, CPS = 9) yielded ΔAUC <0.005 and no predictor reclassification, supporting threshold robustness.The primary endpoint was 3-year disease-free survival (DFS), defined as time from surgery to first documented recurrence (local, regional, or distant), contralateral tumor development, or death from any cause. Patients without events were censored at the last follow-up contact. Surveillance was conducted per institutional protocol: clinical assessment and CEA measurement every 3 months for 2 years, then every 6 months; CT chest/abdomen/ pelvis every 6 months for 3 years; colonoscopy at 1 year then every 3 years. Adjuvant therapy (FOLFOX, CAPOX, or ICI-based regimens) was administered per multidisciplinary tumor board recommendation incorporating MMR status, stage, performance status, and patient preference. S1).Of 408 patients screened, 67 were excluded (31 stage IV, 18 incomplete biomarker data, 12 R1 resection, six prior systemic therapy). The final analytical cohort comprised 341 patients. Baseline clinicopathological characteristics stratified by recurrence status are presented in Table 1. The median follow-up was 36.0 months (IQR: 24-50) for the full cohort; 94 patients (27.6%) experienced disease recurrence during follow-up, with a median time to recurrence of 20.3 months (IQR: 12-32). Three-year DFS was 72.4% in the full cohort. Patients with recurrence were more likely to have stage IIIB-IIIC disease (53.1% vs. 21.1%, p < 0.001), G3-G4 histology (37.2% vs. 21.9%, p = 0.001), and a higher number of positive lymph nodes (median 3 vs. 1, p < 0.001). Adjuvant ICI receipt was significantly less frequent in the recurrence group (14.9% vs. 27.5%, p = 0.027), consistent with ICI-associated DFS benefit. Tumor-immune biomarker data stratified by recurrence status are presented in Table 2 dMMR/MSI-H was identified in 68 patients (19.9%); pMMR/MSS in 273 patients (80.1%). The recurrence group had a significantly higher proportion of pMMR/MSS tumors (90.4% vs. 76.1%, p < 0.001) and lower rates of PD-L1 CPS ≥10 (19.1% vs. 34.0%, p = 0.003). High TIL density (≥20% stromal TILs) was markedly less frequent in recurrent patients (20.2% vs. 43.7%, p < 0.001), confirming that a low-TIL, immune-excluded TIME phenotype strongly associates with post-colectomy recurrence. High NLR (≥3.0) was significantly more common in recurrent patients (56.4% vs. 38.5%, p = 0.002), consistent with systemic immunosuppression as a determinant of recurrence risk. CEA >5 ng/mL was present in 54.3% of recurrent versus 32.4% of non-recurrent patients (p < 0.001).Univariate Cox regression results are presented in Table 3. Tumor-immune biomarkers demonstrating significant univariate associations with inferior DFS included: pMMR/MSS status (HR = 3.28, 95% CI: 1.71-6.29, p < 0.001), low TIL density (HR = 2.94, 95% CI: 1.79-4.84, p < 0.001), PD-L1 CPS <10 (HR = 2.14, 95% CI: 1.28-3.58, p = 0.004), high NLR (HR = 2.08, 95% CI: 1.38-3.13, p < 0.001), and elevated CEA (HR = 2.34, 95% CI: 1.56-3.50, p < 0.001). Among clinicopathological variables, advanced tumor stage, G3-G4 histology, high lymph node ratio, and positive lymph node count were significant predictors. Adjuvant ICI receipt was the sole independent protective factor on univariate analysis (HR = 0.52, 95% CI: 0.32-0.84, p = 0.008). Adjuvant chemotherapy alone did not reach statistical significance (p = 0.282).On multivariate Cox regression (Table 4), the following variables were identified as independent predictors of inferior 3year DFS: pMMR/MSS status (aHR = 2.84, 95% CI: 1.44-5.60, p = 0.003), tumor stage IIIB-IIIC (aHR = 2.52, 95% CI: 1.57-4.05, p < 0.001), lymph node ratio ≥0.25 (aHR = 2.13, 95% CI: 1.28-3.54, p = 0.004), low TIL density (aHR = 2.21, 95% CI: 1.31-3.71, p = 0.003), CEA >5 ng/mL (aHR = 1.88, 95% CI: 1.22-2.88, p = 0.004), and high NLR (aHR = 1.78, 95% CI: 1.16-2.73, p = 0.008). Adjuvant ICI receipt remained an independent protective factor (aHR = 0.48, 95% CI: 0.29-0.80, p = 0.005). Additionally, preoperative albumin (per 1 g/L increase: aHR = 0.94, 95% CI: 0.90-0.98, p = 0.003) and NRS 2002 ≥ 4 (vs. < 4: aHR = 1.58, 95% CI: 1.03-2.43, p = 0.035) emerged as independent nutritional predictors of inferior DFS after