T4bN0 colon cancer shows worse survival than T1–2N1 disease: a multicenter real-world study in China Authors: Xin Li, Lei Liang, Zhen-Ya Yang, Jun-Nan Li, Ning Xu, Jun Yang Published: 2026 ABSTRACT BackgroundGrowing evidence suggests that deeply invasive node-negative colon cancer, particularly T4bN0 disease, may have a worse prognosis than early node-positive tumors, challenging the current American Joint Committee on Cancer (AJCC) 8th edition staging system. This study aimed to compare clinicopathological features and survival outcomes between T4N0 and T1–2N1 colon cancer in a multicenter cohort.MethodsWe conducted a multicenter retrospective cohort study of 391 patients with stage IIB (T4aN0), IIC (T4bN0), or IIIA (T1–2N1) colon cancer treated with curative-intent surgery between 2014 and 2024 at three centers in China. Rectal cancer cases were excluded to reduce treatment heterogeneity. Clinicopathological characteristics were compared across groups, and progression-free survival (PFS) and overall survival (OS) were analyzed using Kaplan-Meier and Cox regression methods.ResultsAmong the 391 patients, 254 had stage IIB, 56 had stage IIC, and 81 had stage IIIA disease. Compared with stage IIIA, T4N0 tumors showed a less favorable pathological profile, including poorer differentiation, more frequent perineural invasion, and more frequent carcinoembryonic antigen (CEA) elevation. In pairwise analyses, stage IIC (T4bN0) was associated with significantly worse PFS (log-rank P=0.008) and OS (log-rank P=0.02) than stage IIIA, whereas no statistically significant differences were observed between stage IIB and stage IIIA. When T4N0 tumors were regrouped and compared with T1–2N1 disease, T4N0 was associated with worse OS (log-rank P=0.04), whereas the PFS difference did not reach statistical significance (log-rank P=0.11). CEA-stratified analyses suggested a more unfavorable survival pattern in the elevated-CEA subgroup, although formal interaction testing was not statistically significant.ConclusionsIn colon cancer, the adverse prognostic signal within T4N0 disease appears to be driven primarily by T4bN0. These findings support reconsideration of current staging paradigms, particularly for T4bN0 colon cancer, and warrant further prospective validation of risk-adapted postoperative strategies. Highlight box Key findings • T4bN0 colon cancer showed significantly worse progression-free survival and overall survival than T1–2N1 disease. • Within T4N0 colon cancer, the adverse prognostic signal appeared to be driven primarily by T4bN0 rather than T4aN0. • Elevated preoperative carcinoembryonic antigen (CEA) identified a subgroup with a more unfavorable survival pattern, although formal interaction testing was not statistically significant. What is known and what is new? • The American Joint Committee on Cancer (AJCC) 8th edition classifies T4N0 colon cancer as stage II and T1–2N1 disease as stage IIIA, despite increasing evidence of a survival paradox between deeply invasive node-negative tumors and early node-positive tumors. • In this multicenter colon-only cohort, T4bN0 showed worse survival than T1–2N1 disease, whereas T4aN0 did not consistently differ from stage IIIA. These findings suggest that the adverse prognostic signal within T4N0 colon cancer is heterogeneous and is driven mainly by T4bN0. What is the implication, and what should change now? • Current staging may underestimate the biological risk of T4bN0 colon cancer. • These patients may warrant closer postoperative surveillance and prospective evaluation of stage III-like risk-adapted management strategies. Introduction Background The American Joint Committee on Cancer (AJCC) tumor-node-metastasis (TNM) staging system remains the cornerstone for prognostication and treatment selection in colon cancer (1). Traditionally, stage II disease—defined by tumor penetration without nodal involvement—has been considered to have a more favorable prognosis than stage III disease, which involves regional lymph node metastasis (2). However, accumulating evidence suggests that this paradigm may not fully apply to deeply invasive node-negative tumors, particularly those staged as T4bN0 (3-9). Rationale and knowledge gap Multiple registry-based and retrospective studies have reported inferior survival in T4bN0 colon cancer compared with early node-positive subgroups such as T1–2N1, despite the former being classified as stage II under the AJCC 8th edition (2,3,8,10). These observations highlight a mismatch between anatomic stage and biological risk, raising concern that the depth of local invasion may carry prognostic implications comparable to, or even exceeding, limited nodal involvement. Biologically, deeply invasive tumors may reflect a more aggressive phenotype associated with epithelial-mesenchymal transition, occult residual disease, and higher metastatic potential (5,11,12). In this context, the AJCC Colon Cancer Expert Panel has recently proposed, at American Society of Clinical Oncology (ASCO) 2025, reassignment of T4N0 tumors within the forthcoming AJCC Version 9 framework (13). This proposal aims to better align staging with emerging biological and clinical evidence. However, real-world validation remains limited, particularly in multicenter Asian cohorts. In addition, whether the adverse prognostic signal applies uniformly across T4aN0 and T4bN0 disease, or is driven primarily by T4bN0, remains unclear. Another unresolved question is whether systemic biomarkers such as carcinoembryonic antigen (CEA) further refine risk stratification within these stage groups. Objective To address these gaps, we conducted a multicenter retrospective study in a Chinese population restricted to colon cancer cases. We aimed to: (I) compare survival across stage IIB (T4aN0), stage IIC (T4bN0), and stage IIIA (T1–2N1) disease under the AJCC 8th edition framework; (II) characterize the clinicopathological aggressiveness of T4 subgroups; and (III) explore whether preoperative CEA identifies a subgroup with a more unfavorable survival pattern. These findings may help refine risk-adapted staging and postoperative management strategies for biologically high-risk colon cancer. We present this article in accordance with the STROBE reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0545/rc). Methods Study design and cohort selection We conducted a multicenter retrospective cohort study to evaluate the outcomes of T4N0 colon cancer under the AJCC 8th edition framework. Consecutive patients treated between 2014 and 2024 were identified from three academic centers in Yunnan, China: The First Affiliated Hospital of Kunming Medical University, Honghe Prefecture Tumor Hospital, and Yunnan Cancer Hospital. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional review boards of all participating centers [The First Affiliated Hospital of Kunming Medical University (No. 2024-L-124), The Third People’s Hospital of Honghe Prefecture (No. 2024-KYXM-29), and the Yunnan Cancer Hospital (No. KYLX2024-146)], and the requirement for informed consent was waived because of the retrospective nature of the study. Eligible patients had histologically confirmed colon adenocarcinoma staged as IIB (T4aN0), IIC (T4bN0), or IIIA (T1–2N1) according to the AJCC 8th edition and had undergone curative-intent surgery. To reduce treatment heterogeneity related to rectal cancer management, all rectal cancer cases were excluded from the revised analysis. Patients with hereditary colorectal cancer syndromes, synchronous or metachronous malignancies, receipt of neoadjuvant therapy, or incomplete pathology or follow-up data were also excluded. Data collection Clinicopathological data were extracted from institutional electronic medical records and harmonized across centers. Variables collected included age, sex, body mass index (BMI), primary tumor location, histologic subtype, tumor differentiation, pathologic T and N categories, lymph-node yield, perineural invasion, vascular invasion, bowel obstruction, preoperative CEA, adjuvant chemotherapy, and follow-up outcomes. Outcome definitions and follow-up Progression-free survival (PFS) was defined as the time from surgery to documented recurrence, progression, or death, whichever occurred first. Overall survival (OS) was defined as the time from surgery to death from any cause. Patients without an event were censored at the last available follow-up. Follow-up was performed according to institutional routine clinical practice. Survival status and recurrence information were obtained from outpatient records, inpatient records, and follow-up documentation from the participating centers. Statistical analysis Categorical variables were compared using the chi-square test or Fisher’s exact test, as appropriate. Continuous variables were summarized as medians with interquartile ranges (IQRs) and compared using the Kruskal-Wallis test. Survival curves for PFS and OS were estimated using the Kaplan-Meier method and compared using the log-rank test. Prespecified comparisons included overall comparison among stage IIB, stage IIC, and stage IIIA; pairwise comparisons between stage IIC and stage IIIA, stage IIB and stage IIIA, and stage IIB and stage IIC; and regrouped comparisons between T4N0 and T1–2N1 disease. Multivariable Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for PFS and OS. Covariates included stage group, sex, age, preoperative CEA, adjuvant chemotherapy, perineural invasion, vascular invasion, bowel obstruction, and tumor sidedness. Additional regrouped Cox models were fitted to compare T4N0 versus T1–2N1 disease. CEA-stratified analyses were performed exploratorily. Formal stage-by-CEA interaction testing was conducted by adding a multiplicative interaction term between stage subgroup and preoperative CEA status to the Cox proportional hazards model. All statistical analyses were performed using SPSS version 26.0 and R version 4.3.0. A two-sided P value <0.05 was considered statistically significant. Results Clinicopathological characteristics A total of 391 patients with colon cancer were included in the final analysis, including 254 with AJCC 8th edition stage IIB (T4aN0), 56 with stage IIC (T4bN0), and 81 with stage IIIA (T1–2N1) disease. Baseline clinicopathological characteristics are summarized in Table 1. Age, sex, BMI, adjuvant chemotherapy exposure, and primary tumor sidedness were generally balanced across the three groups. Table 1Clinicopathological characteristics by AJCC 8th edition stage groups in the colon-only cohortVariableSubgroupStage IIB (T4aN0)Stage IIC (T4bN0)Stage IIIA (T1–2N1)P valueGenderFemale112 (44.09)27 (48.21)45 (55.56)0.20 Male142 (55.91)29 (51.79)36 (44.44)HistologyAdenocarcinoma204 (80.31)46 (82.14)73 (90.12)0.13 Mixed adenocarcinoma26 (10.24)4 (7.14)7 (8.64)Mucinous adenocarcinoma24 (9.45)6 (10.71)1 (1.23)DifferentiationHigh70 (27.56)15 (26.79)35 (43.21)0.01 Low/undifferentiated34 (13.39)6 (10.71)1 (1.23)Medium149 (58.66)34 (60.71)44 (54.32)PNINegative142 (55.91)33 (58.93)57 (70.37)0.04 Positive82 (32.28)19 (33.93)15 (18.52)VINegative189 (74.41)37 (66.07)57 (70.37)0.15 Positive36 (14.17)14 (25.00)15 (18.52)ObstructionNegative195 (76.77)46 (82.14)72 (88.89)0.054Positive59 (23.23)10 (17.86)9 (11.11)CEA>5 ng/mL118 (46.46)26 (46.43)23 (28.40)0.01 ≤5 ng/mL136 (53.54)30 (53.57)58 (71.60)Tumor locationLeft-sided colon126 (49.61)29 (51.79)50 (61.73)0.16 Right-sided colon128 (50.39)27 (48.21)31 (38.27)Age<64 years142 (55.91)29 (51.79)45 (55.56)0.85 ≥64 years112 (44.09)27 (48.21)36 (44.44)ChemotherapyNo78 (30.71)20 (35.71)21 (25.93)0.47 Yes176 (69.29)36 (64.29)60 (74.07)BMI18.5–24 kg/m2147 (57.87)32 (57.14)48 (59.26)0.29 >24 kg/m273 (28.74)18 (32.14)29 (35.80)≤18.5 kg/m234 (13.39)6 (10.71)4 (4.94)Data are presented as n (%). P values were calculated using the chi-square test or Fisher’s exact test, as appropriate. AJCC, American Joint Committee on Cancer; BMI, body mass index; CEA, carcinoembryonic antigen; PNI, perineural invasion; VI, vascular invasion. Clinicopathological characteristics by AJCC 8th edition stage groups in the colon-only cohort Data are presented as n (%). P values were calculated using the chi-square test or Fisher’s exact test, as appropriate. AJCC, American Joint Committee on Cancer; BMI, body mass index; CEA, carcinoembryonic antigen; PNI, perineural invasion; VI, vascular invasion. By contrast, T4N0 tumors showed a less favorable pathological profile than stage IIIA disease. Compared with stage IIIA, T4N0 tumors were associated with poorer differentiation, more frequent perineural invasion, and more frequent elevation of preoperative CEA. These adverse features were particularly evident in the T4bN0 subgroup. Overall, these findings suggest that deeply invasive node-negative colon cancer carries a heavier burden of adverse clinicopathological characteristics than early node-positive disease. Survival disparities under AJCC 8th edition framework Kaplan-Meier analysis showed a borderline difference in OS among stage IIB, stage IIC, and stage IIIA disease (log-rank P=0.07) (Figure 1A), whereas the corresponding difference in PFS was statistically significant (log-rank P=0.03) (Figure 1B). Kaplan-Meier curves for OS and PFS among patients with stage IIB (T4aN0), stage IIC (T4bN0), and stage IIIA (T1–2N1) colon cancer after exclusion of all rectal cancer cases. (A) OS across the three stage groups. (B) PFS across the three stage groups. (C) OS for stage IIC versus stage IIIA. (D) PFS for stage IIC versus stage IIIA. (E) OS for stage IIB versus stage IIIA. (F) PFS for stage IIB versus stage IIIA. (G) OS for stage IIB versus stage IIC. (H) PFS for stage IIB versus stage IIC. OS, overall survival; PFS, progression-free survival. In pairwise comparisons, patients with stage IIC (T4bN0) had significantly worse OS than those with stage IIIA (T1–2N1) disease (log-rank P=0.02) (Figure 1C). A similar pattern was observed for PFS, with stage IIC showing significantly poorer outcome than stage IIIA (log-rank P=0.008) (Figure 1D). In contrast, no statistically significant differences were observed between stage IIB (T4aN0) and stage IIIA for either PFS (log-rank P=0.26) or OS (log-rank P=0.057) (Figure 1E,1F). Direct comparison between stage IIB and stage IIC further showed significantly worse PFS in stage IIC (log-rank P=0.03), whereas the OS difference did not reach statistical significance (log-rank P=0.38) (Figure 1G,1H). Taken together, these results indicate that the adverse prognostic signal within T4N0 colon cancer is driven primarily by the T4bN0 subgroup. Multivariable survival analysis Multivariable Cox proportional hazards models for PFS and OS under the AJCC 8th edition framework are presented in Table 2. For PFS, stage IIC remained independently associated with poorer outcome compared with stage IIIA, whereas stage IIB did not. For OS, stage IIC showed an adverse direction of effect relative to stage IIIA, although the association was attenuated after adjustment. Across models, bowel obstruction was consistently associated with worse PFS and OS, while elevated CEA was independently associated with poorer OS. Right-sided tumor location was associated with better PFS. Table 2Multivariable Cox proportional hazards analysis for survival under the AJCC 8th edition framework in the colon-only cohortVariablePFSOSHR (95% CI)P valueHR (95% CI)P valueStage IIB vs. IIIA1.349 (0.685–2.656)0.39 1.636 (0.762–3.512)0.21 Stage IIC vs. IIIA2.471 (1.151–5.301)0.02 2.137 (0.906–5.039)0.08 Male sex1.372 (0.852–2.211)0.19 1.338 (0.838–2.138)0.22 Age ≥64 years0.647 (0.391–1.070)0.09 2.277 (1.386–3.743)0.001 CEA >5 ng/mL1.244 (0.782–1.978)0.36 1.816 (1.136–2.902)0.01 Adjuvant chemotherapy1.332 (0.761–2.330)0.32 1.109 (0.669–1.836)0.69 PNI1.321 (0.796–2.195)0.28 0.982 (0.576–1.675)0.95 VI1.081 (0.562–2.081)0.82 1.437 (0.774–2.668)0.25 Obstruction1.951 (1.173–3.244)0.01 2.182 (1.325–3.595)0.002 Right-sided colon0.525 (0.322–0.858)0.01 0.960 (0.599–1.540) 0.87 AJCC, American Joint Committee on Cancer; CEA, carcinoembryonic antigen; CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; PNI, perineural invasion; VI, vascular invasion. Multivariable Cox proportional hazards analysis for survival under the AJCC 8th edition framework in the colon-only cohort AJCC, American Joint Committee on Cancer; CEA, carcinoembryonic antigen; CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; PNI, perineural invasion; VI, vascular invasion. Exploratory analyses stratified by CEA CEA-stratified Kaplan-Meier analyses suggested that the adverse survival pattern of stage IIC relative to stage IIIA was more pronounced in patients with elevated preoperative CEA (Figure 2). In patients with elevated CEA, stage IIC showed worse OS than stage IIIA, whereas this difference was less evident in the normal-CEA subgroup. A similar directional pattern was observed for PFS. However, formal stage-by-CEA interaction testing did not reach statistical significance (Table 3). Therefore, the CEA findings should be interpreted as exploratory and hypothesis-generating rather than definitive evidence of interaction. Kaplan-Meier curves for OS and PFS comparing stage IIC (T4bN0) and stage IIIA (T1–2N1) colon cancer after stratification by preoperative CEA level. (A) OS in patients with elevated CEA (>5 ng/mL). (B) OS in patients with normal CEA (≤5 ng/mL). (C) PFS in patients with elevated CEA (>5 ng/mL). (D) PFS in patients with normal CEA (≤5 ng/mL). CEA-stratified analyses suggested a stronger adverse survival pattern for stage IIC disease in the elevated-CEA subgroup; however, formal stage-by-CEA interaction testing was not statistically significant. CEA, carcinoembryonic antigen; OS, overall survival; PFS, progression-free survival. Table 3Exploratory CEA-stratified survival analyses and formal stage-by-CEA interaction testing in the colon-only cohortOutcomeCEA stratumIIC (T4bN0), nIIIA (T1–2N1), nLog-rank P valueP for interactionOS>5 ng/mL26230.02 0.31≤5 ng/mL30580.66 PFS>5 ng/mL26230.02 0.36≤5 ng/mL30580.10 CEA, carcinoembryonic antigen; OS, overall survival; PFS, progression-free survival. Exploratory CEA-stratified survival analyses and formal stage-by-CEA interaction testing in the colon-only cohort CEA, carcinoembryonic antigen; OS, overall survival; PFS, progression-free survival. Survival under regrouping of T4N0 versus T1–2N1 To evaluate the prognostic direction of the proposed stage reassignment framework, we regrouped stage IIB and stage IIC tumors as T4N0 and compared them with stage IIIA (T1–2N1) disease. In this analysis, regrouped T4N0 disease was associated with significantly worse OS than T1–2N1 disease (log-rank P=0.04) (Figure 3A), whereas the difference in PFS did not reach statistical significance (log-rank P=0.11) (Figure 3B). Multivariable Cox models for this regrouped comparison are presented in Table 4. These findings suggest that the adverse prognostic signal of T4N0 colon cancer is most strongly supported by OS and is largely driven by the unfavorable behavior of T4bN0 tumors. Kaplan-Meier curves comparing regrouped T4N0 disease [stage IIB (T4aN0) plus stage IIC (T4bN0)] versus T1–2N1 disease (stage IIIA) in the colon-only cohort. (A) OS. (B) PFS. This regrouping analysis was performed to evaluate the prognostic direction of the proposed stage reassignment framework for T4N0 colon cancer. OS, overall survival; PFS, progression-free survival. Table 4Multivariable Cox proportional hazards analysis for regrouped T4N0 versus T1–2N1 in the colon-only cohortVariablePFSOSHR (95% CI)P valueHR (95% CI)P valueT4N0 vs. T1–2N11.562 (0.810–3.013)0.18 1.746 (0.826–3.690)0.14 Male sex1.391 (0.864–2.240)0.18 1.338 (0.837–2.137)0.22 Age ≥64 years0.654 (0.395–1.085)0.10 2.287 (1.391–3.759)0.00 CEA >5 ng/mL1.207 (0.759–1.920)0.43 1.786 (1.119–2.851)0.02 Adjuvant chemotherapy1.314 (0.748–2.307)0.34 1.119 (0.675–1.855)0.66 PNI1.305 (0.784–2.173)0.31 0.968 (0.569–1.647)0.91 VI1.154 (0.600–2.217)0.67 1.484 (0.803–2.742)0.21 Obstruction1.914 (1.149–3.188)0.01 2.165 (1.315–3.566)0.00 Right-sided colon0.523 (0.320–0.853)0.01 0.953 (0.595–1.527) 0.84 T4N0 disease comprised stage IIB (T4aN0) plus stage IIC (T4bN0). CEA, carcinoembryonic antigen; CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; PNI, perineural invasion; VI, vascular invasion. Multivariable Cox proportional hazards analysis for regrouped T4N0 versus T1–2N1 in the colon-only cohort T4N0 disease comprised stage IIB (T4aN0) plus stage IIC (T4bN0). CEA, carcinoembryonic antigen; CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival; PNI, perineural invasion; VI, vascular invasion. Discussion Key findings In this multicenter colon-only cohort, the adverse prognostic signal within T4N0 disease was not uniform but appeared to be driven primarily by T4bN0. Patients with stage IIC (T4bN0) colon cancer had significantly worse progression-free survival and overall survival than those with stage IIIA (T1–2N1) disease, whereas stage IIB (T4aN0) did not consistently differ from stage IIIA. In addition, T4N0 tumors carried a heavier burden of adverse clinicopathological features, including poorer differentiation, more frequent perineural invasion, and more frequent elevation of preoperative CEA. Together, these findings suggest that deeply invasive node-negative colon cancer, particularly T4bN0 disease, may carry biological risk comparable to or greater than that of early node-positive tumors. Strengths and limitations There are several strengths in this study. First, it was conducted across multiple centers in a real-world clinical setting. Second, by restricting the analysis to colon cancer and excluding rectal cancer, we reduced an important source of treatment heterogeneity related to neoadjuvant strategies and local treatment pathways. Third, we distinguished between T4aN0 and T4bN0 rather than treating T4N0 as a homogeneous category, which allowed a more nuanced interpretation of the survival paradox. Several limitations should also be acknowledged. The study was retrospective in nature. The sample size of the T4bN0 subgroup was relatively modest, which may have reduced statistical power in adjusted and interaction analyses. Molecular data were unavailable, precluding evaluation of biomarkers such as MSI, RAS/BRAF status, and consensus molecular subtype. In addition, the analysis did not include more advanced stage III subgroups, and the cohort was derived from three centers in China, which may limit generalizability to other populations. Future prospective studies incorporating molecular profiling are warranted to further refine risk stratification in T4N0 colon cancer. Comparison with similar research Our findings are directionally consistent with previous registry-based and retrospective studies reporting a survival paradox between stage IIB/IIC and stage IIIA colon cancer (6-9,14). However, the present study adds an important nuance: in our colon-only cohort, the adverse prognostic signal was driven mainly by T4bN0 rather than being uniformly present across all T4N0 tumors. This distinction is clinically relevant because it suggests that the biological and prognostic heterogeneity within T4N0 disease should be acknowledged when considering stage reassignment strategies. Explanations of findings The inferior survival of T4bN0 colon cancer compared to T1–2N1 disease reflects its intrinsically aggressive biology. T4b tumors directly invade adjacent organs or structures, indicating a more advanced local phenotype and increasing the likelihood of occult residual disease despite apparently curative resection (5,15,16). Such tumors may require more extensive surgery and may be associated with more challenging postoperative management (17). In our cohort, T4N0 tumors—particularly T4bN0—were also enriched for adverse pathological features such as poorer differentiation, more frequent perineural invasion, and more frequent CEA elevation (18). These findings support the biological plausibility that some deeply invasive node-negative tumors may behave more aggressively than limited node-positive disease. CEA has been associated with tumor burden and prognosis in colorectal cancer in prior studies (19-21). In the present study, CEA-stratified analyses suggested a more unfavorable survival pattern among patients with elevated preoperative CEA, particularly in the T4bN0 subgroup. However, because formal stage-by-CEA interaction testing was not statistically significant, CEA is better interpreted here as a clinically relevant risk-stratification variable rather than a definitively established effect modifier. Even so, the observed pattern suggests that preoperative CEA may help identify a subgroup warranting closer postoperative attention. Implications and actions needed Clinically, these data suggest that the AJCC 8th edition framework may underestimate postoperative risk in a subset of T4N0 colon cancer, especially T4bN0 disease. Our regrouped analysis of T4N0 versus T1–2N1 further supports this interpretation, with the adverse signal being most clearly reflected in overall survival. While our findings do not by themselves mandate immediate treatment reclassification, they support reconsideration of current staging paradigms and justify further prospective evaluation of risk-adapted postoperative strategies, including closer surveillance and refinement of adjuvant treatment decision-making for selected patients. Conclusions In conclusion, T4N0 colon cancer is biologically and prognostically heterogeneous, and the adverse signal within this category appears to be driven primarily by T4bN0. In this multicenter colon-only cohort, T4bN0 showed worse survival than T1–2N1 disease, whereas T4aN0 did not consistently differ from stage IIIA. These findings support reconsideration of current staging paradigms and warrant further prospective validation of risk-adapted postoperative strategies for T4N0 colon cancer. Supplementary The article’s supplementary files as Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-0545/prf