Colorectal cancer (CRC) remains one of the leading causes of cancer-related morbidity and mortality worldwide. The widespread implementation of screening programs led to an increased detection of T1-CRC, defined as CCR with submucosal invasion, and accounting for approximately 40% of cancers identified during screening colonoscopy (1). These lesions are potentially curable through endoscopic resection (ER). ER offers a less invasive alternative to surgery, with lower morbidity and shorter hospital stays (2,3). According to current international guidelines (4-6), when histopathological analysis reveals high-risk features [lymphovascular invasion (LVI), poor differentiation, high-grade budding, deep submucosal invasion, positive resection margins (R1 or Rx)], surgical intervention with lymphadenectomy is recommended. However, only 10% of patients (pts) with T1-CRC exhibit lymph node metastasis (LNM) following surgery (7-9), indicating a high percentage of overtreatment. On the other hand, the definition of a "free resection margin" (FRM) is highly variable in literature: most guidelines suggest that an R0 resection requires a distance from the nearest tumor cell to the deep resection margin of >1 mm, some studies have perhaps indicated that a margin of 0.1 mm may suffice (10), while a large meta-analysis assessing pathological factors associated with LNM in early CCR found no increased risk associated with R1 (11).

On this basis, there is a need for a more precise strategy for oncological risk assessment (local residual, LNM, distant metastasis (M) and recurrence) following ER of early CCR in order to minimize unnecessary surgeries. This study aims to evaluate the risk of local residual cancer, LNM, M, and recurrence in pts who underwent ER for T1-CRC with a positive resection margin (R1) and without other risk factors except deep submucosal invasion.

Single-center retrospective study conducted in our academic tertiary referral center evaluating pts diagnosed with T1-CRC with positive resection margin irrespective of subsequent surgery (2013-2025). Inclusion criteria were: ER of T1-CRC; R1 (FRM ≥ 0 mm); deep submucosal invasion (sm2-sm3) as the unique risk factor; at least 1 year follow-up; evaluability of the margins in case of piece-meal resection of T1 CRC (no Rx). Exclusion criteria were: hereditary CRC syndrome; inflammatory bowel disease; preoperative radio-chemotherapy; presence of synchronous CRC; diagnosis of CRC > T2 in the 5 years before; Rx margins.

FU comprehended endoscopic evaluation (at 6 and 12 months for piece-meal resections and at 12 months for en-bloc resections) and a CT scan at 12 months.

A total of 39 pts were included. These were divided in 2 groups: 1) R1 margin of 0 mm (17pts; 43.6%); 2) R1 margin between 0.1 and 1 mm (22 pts; 56.4%). At diagnosis, none of the pts had LNM, while local residual cancer was found in 2 pts of group 1 (5.13% [95% CI: 1.07% - 17.72%]) and in none of group 2. At 1 year, none of the pts developed LNM, distant metastasis, nor local recurrence.

Our study shows that in endoscopically resected T1-CRC the risk of LNM at diagnosis is extremely low (0 pts), while local residual cancer was observed in a small percentage of pts, irrespective of the entity of R1 (no statistical difference between 0 mm-R1 and 0.1-1mm-R1). The absence of LNM, M, and recurrence at 1 year FU further strengthens the notion that close monitoring may be sufficient for these pts, avoiding overtreating surgery. To support and to refine risk assessment strategies in early-stage CRC, larger, multi-center studies are warranted.