Abstract
Background/Aim: Circumferential resection margin (CRM) has been shown to be a strong predictor of risk of local recurrence (LR) and distant metastases (DM) in patients with rectal cancer treated with total mesorectal excision (TME). However, not all patients with a CRM of zero develop LR or DM. This study aimed to identify factors associated with disease-free survival (DFS) despite non-radical excision (CRM=0 mm).
Patients and Methods: Patients surgically treated for adenocarcinoma of the rectum between 2005 and 2013 were identified through the Swedish ColoRectal Cancer Registry.
Results: Of 8,392 patients included in the study, 159 (1.8%) were found to have CRM 0 mm. The rates of LR (n=27; 17%) and DM (n=62; 39%) were high in the CRM 0 mm group; however, more than half of these patients developed neither (n=84; 52.8%). Stage II disease was associated with a higher DFS [odds ratio (OR)=2.15; 95% confidence interval (CI)=1.11-4.18] compared with stage III disease. Neoadjuvant chemotherapy was a negative prognostic factor of DFS (OR=0.47, 95% CI=0.23-0.97) independent of the stage of disease.
Conclusion: Over half of the patients with CRM 0 mm after rectal cancer surgery did not develop LR or DM. Lower stage of disease and no neoadjuvant chemotherapy were associated with a better DFS. These findings suggest that other biological or molecular factors may influence prognosis after non-radical excision, highlighting the need for further research to improve postoperative management and follow-up.
- Rectal cancer
- circumferential resection margin
- local recurrence
- distant metastases
- disease-free survival
Introduction
Rectal cancer is a significant global health burden, with an estimated 700.000 new cases annually (1). Surgery remains the cornerstone of curative treatment, with total mesorectal excision (TME), i.e., complete removal of the rectum and its surrounding lymphatic and fatty tissue, being the gold standard (2). Since the introduction of TME, rates of local recurrences (LR) and mortality have decreased (3, 4). Improved oncological treatment with radio- and chemotherapy has further increased survival rates. Several factors influence prognosis in patients with rectal cancer. Inadequate circumferential resection margin (CRM) has been shown to be an independent risk factor for LR (5-8) and, more recently, also for distant metastases (DM) (9). The cut-off for when CRM is considered inadequate has been under debate. Some studies have advocated a limit of 2 mm (10, 11) and others have considered 1 mm (5) to be acceptable.
Patients with a CRM of 0 mm are considered to have the highest risk of poor outcome. However, not all patients with non-radical excision develop LR or DM. Existing literature does not explore which specific variables within this group are most predictive of recurrence, emphasizing the need of further investigation into patient-specific risk factors contributing to this outcome. Moreover, current guidelines emphasize the importance of negative margins but provide limited guidance on risk stratification and management of patients with a CRM of 0 mm (12). The lack of risk-adapted strategies represents a critical gap in clinical practice, necessitating further research which may guide more personalized treatment strategies and follow-up practices.
This study aimed to investigate a subgroup of Swedish patients from the national registry-based study by Agger et al., namely those with non-radical excision (CRM 0 mm) to more thoroughly characterize this patient group and to identify patient, tumor, or treatment factors associated with the risk of LR and DM. Identifying these factors could help optimize adjuvant treatment strategies and surgical decision-making. Hopefully, this study will contribute to a more nuanced risk stratification for rectal cancer patients with compromised surgical margins.
Patients and Methods
Patients who had undergone TME surgery [anterior resection (AR), Hartmann’s procedure, or abdominoperineal resection] for a rectal adenocarcinoma (lower border below 15 cm from anal verge measured by rigid sigmoidoscopy) during the time period January 1st, 2005, and December 31st, 2013, were identified through the Swedish ColoRectal Cancer Registry (SCRCR). The SCRCR is a national registry in which data on predefined patient, tumor and treatment factors of patients with colorectal adenocarcinomas is gathered since 1995 for the rectum and 2007 for the colon. The SCRCR has been continuously validated and is considered a reliable source for quality assurance and research (13-16).
Exclusion criteria were lack of data on exact CRM, stage IV disease at diagnosis, non-curative intention at surgery, non-radical excision macroscopically, death within 30 days, LR within 90 days or if they were lost in follow-up.
Statistical analysis. To investigate potential associations between patient, tumor, and treatment factors with risk of LR, DM and disease-free survival (DFS), logistic regression analysis was used. The outcome variable DFS was created to identify the patients who developed neither LR nor DM during the follow-up period. Continuous variables were categorized. Age was dichotomized based on median value. Values established by the World Health Organization implicating under-weight, over-weight, and obesity were used as cut-offs for body mass index (BMI) (17). Tumor height was categorized into low (0-5 cm), mid (5-10 cm) and high (11-15 cm) rectal cancers (18, 19). A complete-case analysis was conducted excluding observations containing missing data. Results are shown as odds ratios (ORs) with 95% confidence intervals (95%CIs). A multivariable analysis was performed including the variables that showed a statistically significant association in the univariable analysis. SPSS® version 30 (IBM, Armonk, NY, USA) was used for statistical analyses.
Results
In all, 12,146 patients were identified as potentially eligible for the study in the SCRCR. After exclusion, 8,392 patients remained. Of these, 159 patients had CRM 0 mm (Figure 1). The characteristics of these patients are shown in Table I.
Study flow chart.
Characteristics of the study population*.
Follow-up time for the study population was median 42 (1-91) months. In all, 84 patients (52.8 %) with CRM 0 mm developed neither LR nor DM. Results of logistic regression analyses for the outcome variables LR, DM, and the combined variable DFS are shown in Table II.
Factors potentially associated with local recurrence (LR), distant metastases (DM) and disease-free survival (DFS).
Patients with stage II disease demonstrated a statistically significant higher likelihood of DFS compared to those with stage III disease. Conversely, patients who had received neoadjuvant chemotherapy exhibited a reduced likelihood of DFS. These associations withstood in multivariable analysis adjusting for both stage and neoadjuvant chemotherapy.
Discussion
The results of this study showed that despite the high-risk profile associated with a CRM of 0 mm, more than half of the patients did not develop LR or DM. Patients with stage II disease had a higher DFS compared to those with stage III. Neoadjuvant treatment with chemotherapy was associated with worse DFS.
The finding that, despite having a CRM of 0 mm, more than half of the patient cohort (52.8%) did not develop LR or DM, indicates a more favorable prognosis than traditionally expected. This result challenges the widely-held assumption that a CRM of 0 mm always implies poor prognosis. On the contrary, the findings from this study raise important questions about what other biological or molecular factors might modulate the risk of recurrence in this subset of patients, underlining the need for a more personalized risk assessment. The understanding of the heterogeneity of colorectal disease has evolved greatly during the last decades, with analyses of microsatellite instability and oncogenic mutations in B-RAF and RAS being incorporated in the clinical work-up when designing treatment plans. Unfortunately, information about these biomarkers was not accessible for this study and could not be analyzed.
Another notable finding from this study is the confirmation of disease stage as a statistically significant prognostic factor. Patients with stage II disease had a better DFS compared to those with stage III tumors. This suggests that nodal involvement, a hallmark of stage III disease, plays a considerable role in determining prognosis, even in the context of a non-radical circumferential margin. Also, tumor deposits and extramural venous invasion may be more frequent in stage III disease, which has been shown in a study by Lord et al. to be an even better prognostic factor of poor outcome compared to TNM stage (20). The finding that patients with lower stage of disease have a better prognosis aligns with previous studies, for example the systematic review and meta-analysis by Detering et al. (8), in which the significance of tumor stage in predicting survival outcomes even when the CRM is positive was emphasized. This result is valuable because it helps refine risk stratification within a group already considered high-risk based solely on CRM status, supporting the idea that not all CRM 0 mm patients carry equal prognostic risk. This highlights the need for further exploration into how tumor stage and CRM status interact, and whether this allows for subsets of CRM-positive patients to be reclassified into lower-risk categories.
Patients who had undergone neoadjuvant chemotherapy had a poorer DFS. This might reflect a more advanced tumor burden preoperatively. However, the difference withstood when adjusting for stage. Neoadjuvant radiotherapy had a close to statistically significant association with DFS, also as a negative prognostic factor. The difference may be partly caused by selection bias, as neoadjuvant treatment is more often recommended to high-risk patients. If the patients had undergone neoadjuvant treatment and still had locally advanced tumor burden, resulting in non-radical margins, it might reflect a higher level of treatment resistance. It may also reflect a more aggressive tumor biology and a higher likelihood of poor response to adjuvant treatment.
Other clinicopathological factors, such as tumor height and surgical technique (anterior resection/Hartmann’s operation/abdominoperineal resection), were not significantly associated with LR, DM or DFS. The absence of statistically significant associations may reflect the small sample size (n=159), which limits the statistical power to detect subtle effects and risk of type II errors. Additionally, this could be explained by potential confounding by unmeasured variables such as tumor biology.
A strength of this study is its national range by the use of the SCRCR, a validated and reliable data source, which may enhance the applicability of the results to clinical practice. Even though the sample size can be considered small, it is a large number compared to other studies in the field.
The results of the study may contribute to improved risk-adapted surveillance strategies. Despite CRM of 0 mm, patients with stage II disease and no neoadjuvant chemoradiotherapy could possibly be spared an intensive follow-up program which is indicated for higher-risk patients.
The limitations of the study include its retrospective and registry-based design, which carries some risk of reporting bias, missing data and variability in how clinical events are reported across different hospitals. Within the SCRCR, certain variables were incompletely documented, for example, data on adjuvant chemotherapy was missing in nearly 40% of patients. This limits the ability to fully assess its impact on outcomes. However, the effect of adjuvant chemotherapy is questioned in patients with rectal cancer who have received neoadjuvant chemoradiotherapy (21, 22).
It can be noted that during the study time period (2005-2013) the treatment strategies differed somewhat from the current guidelines. Oncological treatment consisted more of short course radiotherapy and chemoradiotherapy in contrast to for example total neoadjuvant therapy. Also, the majority of patients were operated with open technique whereas today, more and more patients are undergoing minimally-invasive surgery.
Conclusion
Overall, this study highlights that a substantial proportion of patients with CRM 0 mm remain disease-free, with disease stage and neoadjuvant chemotherapy being the most important predictors of outcome. Further translational research is warranted to identify biomarkers that explain why some patients with CRM 0 mm remain disease-free. Also, future studies are needed to evaluate new and evolving treatment strategies including total neoadjuvant therapy and proton therapy.
Footnotes
Authors’ Contributions
PB/CD designed the study; CD performed statistical analyses; BE/CD drafted the manuscript; All Authors revised and approved the final manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
Funding
This study received financial support from the ALF and SUS Foundations.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received September 22, 2025.
- Revision received October 6, 2025.
- Accepted October 7, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
Jump to section
Related Articles
Cited By...
- No citing articles found.