Risk Factors for Microscopically Non-radical Surgical Resection in Patients With Rectal Cancer Undergoing Neoadjuvant Radiotherapy

Abstract

Background/Aim

Achieving a microscopically radical (R0) resection is essential in rectal cancer surgery to reduce local recurrence, distant metastases, and improve long-term survival. This retrospective cohort study aimed to identify factors associated with microscopically non-radical (R1) excision in patients undergoing surgery for rectal adenocarcinoma at a Swedish tertiary referral center.

Patients and Methods

Patients with rectal adenocarcinoma operated on between January 2015 and December 2020 at Skåne University Hospital, Malmö, were identified through the Swedish Colorectal Cancer Registry and complementary chart review. Clinical, pathological, and perioperative variables were compared between patients with R0 and R1 resections. R1 resection was defined as a circumferential resection margin (CRM) ≤0.1 mm or an involved longitudinal resection margin. Factors associated with R1 status were analyzed using univariable and multivariable logistic regression.

Results

Among 348 included patients, 17 of 309 (5.5%) with available resection status had a microscopically non-radical resection (R1), while resection status was missing for 39 patients. Distal tumor location was strongly associated with R1 status: tumors ≤7 cm from the anal verge had an increased risk of non-radical excision [adjusted odds ratio (aOR)=3.77, 95% confidence interval (CI)=1.23-14.2]. Longer operating time was related to R1 resection (median 531 vs. 415 min, p=0.045); however, this association was not retained after adjustment (OR=1.21, 95%CI=0.42-3.58). Advanced pathological T stage (pT3-4) was associated with higher R1 risk (aOR=4.15, 95%CI=1.06-27.7). In contrast, examination of more than 22 lymph nodes was associated with a lower risk of R1 resection (aOR=0.17, 95%CI=0.04-0.56).

Conclusion

Distal tumor location and advanced pathological tumor stage are key risk factors for microscopically non-radical resection in rectal cancer, whereas an extensive lymph node yield appears protective. These findings support meticulous preoperative planning and tailored surgical strategies for low rectal tumors to minimize the risk of R1 excision.

Introduction

Surgical resection remains the cornerstone of curative treatment in rectal cancer, despite major advancements in multimodal therapy. Each year, approximately 700,000 individuals are diagnosed with rectal cancer worldwide, and although neoadjuvant radiotherapy, chemotherapy, and organ-preserving strategies are increasingly utilized, long-term survival continues to depend on successful oncological surgery (1, 2). The introduction of total mesorectal excision (TME) has led to substantial improvements in local control, enabling en bloc removal of the rectum and its mesorectum within an intact fascial envelope (3, 4). This technical refinement has significantly reduced locoregional recurrence and has emphasized the pivotal role of achieving tumor-free margins.

Achieving tumor-free resection margins remains one of the strongest prognostic markers for patient outcomes in rectal cancer (5, 6). Both circumferential resection margins (CRM) and longitudinal resection margins (LRM) must be clear of tumor infiltration to minimize the risk of recurrence. A compromised CRM is particularly important, as it has been strongly associated with increased local recurrence, distant metastases, and decreased overall survival (7-9). Although neoadjuvant radiochemotherapy has reduced the frequency of involved margins, R1 resections continue to occur even within specialized colorectal centers, especially for distal tumors, locally advanced disease, and cases with difficult pelvic anatomy, where surgical planes are more difficult to maintain (10-13).

Understanding which patients are most at risk of non-radical excisions is clinically important for several reasons. First, it may guide decision-making regarding the optimal surgical approach, including the need for extended resections, extralevator abdominoperineal resection (ELAPE), or referral to high-volume centers. Second, it may inform individualized treatment planning, improve preoperative patient counseling, and justify intensified neoadjuvant therapy. Furthermore, identifying objective predictors of microscopically non-radical (R1) resection can support quality improvement initiatives within colorectal surgery.

The aim of this study was therefore to investigate the clinical, pathological, and perioperative risk factors associated with R1 excision in patients undergoing neoadjuvant radiotherapy followed by surgery for rectal adenocarcinoma at a Swedish tertiary referral center.

Patients and Methods

Patients with rectal adenocarcinoma, who received neoadjuvant radiotherapy and underwent surgical resection at Skåne University Hospital, Malmö, between January 2015 and December 2020 were eligible. Tumor level was defined as ≤15 cm from the anal verge, measured using rigid proctoscopy. Patients were identified through the Swedish ColoRectal Cancer Registry (SCRCR), a nationwide registry that prospectively collects data on all patients with colorectal adenocarcinoma in Sweden. The SCRCR has shown high validity and coverage, with >95% completeness and high interobserver agreement (14, 15).

Data on patient characteristics, tumor stage and location, perioperative details, and pathological outcome were retrieved from the registry and supplemented through individual chart review. A microscopically non-radical resection (R1) was defined as tumour, tumour deposits, extranodal extension, or vascular/perineural invasion involving or ≤0.1 mm from any inked resection margin, including circumferential, proximal/distal, or perforation-site margins.

Several measures were used to minimize bias. Selection bias was reduced by including consecutive patients with the use of a population-based registry limiting the risk of selective inclusion. To reduce information and misclassification bias, exposure and outcome variables were defined a priori using standardized criteria. R1 resection was defined according to a predefined pathological margin definition (≤0.1 mm from any inked margin). Registry data were cross-validated through individual chart review to ensure data accuracy. Tumor height was uniformly measured using rigid proctoscopy. Potential confounding was addressed through multivariable regression modelling. Clinically relevant covariates were selected a priori based on known prognostic factors in rectal cancer, including tumor stage, tumor height, neoadjuvant treatment characteristics, and operative factors.

Ethics. The study protocol was approved by the Swedish Ethical Review Authority, Sweden (Dnr 2023-00199-01). Patient data were pseudonymized before analysis to ensure confidentiality and compliance with data protection regulations. Since this was a strictly register-based study, individual informed consent was not required.

Statistics. Categorical variables were analyzed in dichotomized form, reported as n (%). Continuous variables are reported as median (IQR) given their skewed distributions. Comparisons between groups were performed using the χ² test or Fisher’s exact test for categorical variables and the Mann-Whitney U-test for continuous variables. A p-value <0.05 was considered statistically significant.

To evaluate factors associated with R1 resection, univariable and multivariable logistic regression models were applied. Variables were dichotomized using clinical cut-offs or median values where appropriate. The multivariable model was adjusted for known potential confounders, including age, sex, tumor stage (pT), tumor level, number of lymph nodes examined, and operating time.

The proportion and distribution of missing data were examined for each variable. No systematic patterns suggesting informative missingness were identified. Analyses were performed using complete cases, and no formal imputation procedures were applied. All statistical analyses were performed using R software for statistical programming (version 4.5.0; R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 348 patients diagnosed with rectal adenocarcinoma and treated surgically between 2015 and 2020 were included. Baseline characteristics are presented in Table I. The median age was 71 years [interquartile range (IQR)=65-76], 210 (60.3%) were male, and most patients were classified as ASA II or III. Neoadjuvant chemotherapy had been administered in 112 (32.2%) patients, and almost everyone had been preoperatively discussed at a multidisciplinary tumor board.

Table I. Characteristics of the study population.

Demographic, clinical, and tumor-related characteristics of all included patients (n=348). Continuous variables are presented as median [interquartile range (IQR)] and categorical variables as number (%). Missing values are reported separately. BMI: Body mass index; IQR: interquartile range; ASA: American Society of Anesthesiologists physical status classification; MDT: multidisciplinary team; cT: clinical tumor stage; cN: clinical nodal stage; cM: clinical metastatic stage; cTX: primary tumor cannot be assessed; cNX: regional lymph nodes cannot be assessed.

A microscopically radical excision (R0) was achieved in 292 (94.5%) patients, whereas 17 (5.5%) patients had a non-radical resection (R1; CRM ≤0.1 mm or involved LRM). Data on resection status were missing for 39 patients (11.2%). Comparisons between the two groups are shown in Table II. There were no differences regarding age, BMI, sex, ASA class, clinical tumor stage, or metastatic status. However, a higher proportion of patients in the R1 group had received neoadjuvant chemotherapy (58.8% vs. 30.5%, p=0.029).

Table II. Comparison between patients with microscopically radical (R0) and microscopically non-radical (R1) resections.

BMI: Body mass index; ASA: American Society of Anesthesiologists physical status classification; cT: clinical tumor stage; cN: clinical nodal stage; cM: clinical metastatic stage; cTX: primary tumor cannot be assessed; cNX: regional lymph nodes cannot be assessed; R0: microscopically radical resection; R1: microscopically non-radical resection. Continuous variables are presented as median [interquartile range (IQR)] and categorical variables as number (%). Missing values are reported separately. Significant values are shown in bold.

Table III shows perioperative and pathological associations with R1 resection. Patients in the R1 group had longer operating time (median 531 vs. 415 min, p=0.045) and lower tumor level measured from the anal verge (median 5 vs. 7 cm, p=0.029). Additionally, they had fewer examined lymph nodes (median 17 vs. 23, p=0.048) and more frequent node positivity (pN1-2: 64.7% vs. 46.2%, p=0.026). No associations were observed for surgical approach, intraoperative perforation, postoperative complications, or 30-day mortality.

Table III. Comparison of operative, pathological, and postoperative outcomes between patients with microscopically radical (R0) and microscopically non-radical (R1) resections.

Continuous variables are presented as median [interquartile range (IQR)] and categorical variables as number (%). Missing values are reported separately. APR: Abdominoperineal resection; ICU: intensive care unit; BMI: body mass index; IQR: interquartile range; pT: pathological tumor stage; N0-N2: pathological nodal stage; NX: regional lymph nodes cannot be assessed; R0: microscopically radical resection; R1: microscopically non-radical resection. Significant values are shown in bold.

Table IV presents the results of regression analyses. Independent predictors of R1 resection were distal tumor location ≤7 cm from the anal verge [adjusted odds ratio (aOR)=3.77; 95% confidence interval (CI)=1.23-14.2] and advanced pathological tumor stage pT3-4 (aOR=4.15; 95%CI=1.06-27.7). Examination of more than 22 lymph nodes was associated with lower risk of R1 resection (OR=0.17; 95%CI=0.04-0.56). Operating time >415 min was associated with R1 status in comparative analysis but did not remain significant after adjustment.

Table IV. Uni- and multivariable logistic regression analysis of factors associated with microscopically non-radical (R1) resection.

Odds ratios (ORs) with 95% confidence intervals (CIs) are presented. Reference categories are shown in parentheses. pT: Pathological tumor stage; pN: pathological nodal stage; APR: abdominoperineal resection; R1: microscopically non-radical resection.

Discussion

Achieving a microscopically radical resection remains a critical determinant of long-term oncological outcome in rectal cancer. In this single-center cohort, distal tumor location and advanced pathological tumor stage (pT3-4) were independently associated with an increased risk of R1 resection, whereas a higher lymph node yield was protective. These findings highlight the interplay between pelvic anatomy, tumor biology, and technical difficulty in rectal cancer surgery.

The importance of tumor height from the anal verge is well established. Low rectal tumors challenge the integrity of the dissection planes, particularly where the mesorectum tapers and fascial envelopes become narrow. In the MERCURY II study, tumors <4 cm from the anal verge were strongly associated with positive CRM, despite standardized magnetic resonance imaging (MRI) staging and high-quality TME surgery (16). Our findings are consistent with these results, and the strong association between tumor size ≤7 cm and R1 resection suggests that low tumors remain at particular risk of margin compromise, even in a specialized center.

Pathological tumor stage pT3-4 was also an independent predictor of R1 resection in our material. Deeper invasion into or beyond the mesorectal fascia increases the likelihood of threatened or involved CRM, and several studies have demonstrated that CRM involvement in advanced T-stage disease is associated with higher rates of local recurrence and poorer survival (17, 18). It remains debated whether positive margins primarily reflect aggressive tumor biology or technical limitations. In our cohort, longer operative time was associated with R1 status in comparative analysis but not after adjustment, which may indicate that extended operating time is more a marker of surgical complexity than of inadequate technique.

The association between lower lymph node yield and R1 resection is noteworthy. Low node counts have been linked to understaging and impaired survival after colorectal cancer surgery (19, 20). Although all patients in this cohort had received neoadjuvant radiotherapy – which is known to reduce nodal size and yield – the finding remained significant. In our study, examination of more than 22 lymph nodes was associated with a significantly lower risk of R1 resection. This may reflect more complete mesorectal excision and optimal specimen handling, in line with TME-quality concepts where intact mesorectal envelopes and generous lymph node harvest are markers of high-quality surgery.

We observed a trend towards more frequent abdominal perineal resection (APR) in the R1 group, although this did not reach statistical significance. Several studies have reported higher CRM involvement after conventional APR compared with anterior resection, leading to the development of ELAPE to improve radial clearance (21-23). While ELAPE may reduce CRM positivity, it is also associated with increased morbidity and more extensive perineal defects. In our unit, ELAPE was used selectively, which limits our ability to evaluate its independent effect. Our findings support careful case selection for extended procedures, particularly in patients with low, bulky tumors or threatened CRM on preoperative MRI.

Although detailed restaging data were not available for all individuals, virtually all patients underwent some form of clinical and radiological reassessment following neoadjuvant treatment. The higher proportion of neoadjuvant chemotherapy in the R1 group therefore likely reflects greater initial tumor burden rather than inadequate response. While post-treatment MRI has recognized limitations – particularly in low rectal tumors where fibrosis and altered anatomy may affect assessment of the mesorectal fascia – its role remains central in guiding surgical strategy. Previous studies, including that by Yuval et al., show that MRI after chemoradiation may occasionally overestimate CRM risk in anterior low tumors (24), yet this should be interpreted within the broader context of multidisciplinary evaluation.

Rather than serving as a stand-alone determinant of resectability, restaging imaging should function as a road mapping tool that integrates radiological findings with clinical examination and intraoperative assessment. When used in this way, restaging contributes meaningfully to operative planning and may support more predictable and precise achievement of radical resection. This approach is consistent with the 2025 European Society for Medical Oncology (ESMO) Clinical Practice Guideline, which emphasizes the role of high-quality restaging in guiding individualized treatment planning rather than determining resectability in isolation (25).

The definition of non-radical resection and CRM involvement varies widely in the literature. Proposed CRM cut-offs range from 0.1 to 2 mm, and several studies have shown that the prognostic impact of CRM distance may differ depending on the use of neoadjuvant therapy (26-29). In the present study, we used a strict definition of R1 as CRM ≤0.1 mm or involved LRM, aiming to capture patients at the highest risk for recurrence and distant metastases. This stringent definition may explain the relatively low proportion of R1 resections but strengthens the clinical relevance of the identified risk factors. The strengths of this study include: (i) prospectively registered cohort with near-complete national registry capture, (ii) Strict pathological definition of R1 (≤0.1 mm) and (iii) combined registry and chart review that improved the accuracy of pathological and perioperative data.

Study limitations. First, the small number of R1 cases limits statistical power, risking type II error for borderline variables. Second, retrospective MRI review was not feasible, restricting analysis of radiological-surgical correlations. Third, ELAPE was selectively performed, limiting evaluation of its protective effect on CRM. Finally, Neoadjuvant regimens varied, which could influence tumor response and resectability.

Clinical implications and Conclusion

In conclusion, distal tumor location and advanced pathological tumor stage increase the risk of R1 resection in rectal cancer surgery, whereas extensive lymph node retrieval is protective. Enhanced preoperative planning and tailored surgical strategies for low rectal tumor patients undergoing neoadjuvant radiotherapy are essential to minimize the risk of R1 resection.

Conflicts of Interest

The Authors report no conflicts of interest related to this work.

Authors’ Contributions

Pamela Buchwald conceived and designed the study, collected and interpreted the data, and did a critical revision of the manuscript. Valentinus Valdimarsson contributed to data interpretation, performed the statistical analyses, and drafted the manuscript. Cecilia Dahlbäck contributed to interpretation of the results, and critical revision of the manuscript. All Authors reviewed, approved the final version of the manuscript, and agree to be accountable for all aspects of the work.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Artificial Intelligence (AI) Disclosure

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning-based image enhancement tools.