Guidelines discourage pre-resection biopsy because of concerns about inducing submucosal fibrosis, impairing mucosal lifting, and increasing technical difficulty. However, biopsies remain common in routine practice, often performed before referral, creating a need to quantify their real-world impact. To date, no single cohort has comprehensively evaluated these outcomes stratified by biopsy status.

Consecutive patients undergoing endoscopic resection (ER) for LNPCPs at St. Michael’s Hospital, Toronto, Canada between November 2024 and October 2025 were prospectively enrolled in an observational cohort. Lesions with a history of previously attempted endoscopic resection or prior tattooing were excluded. This study evaluates whether prior biopsy adversely affects the technical feasibility and safety of ER for advanced colorectal lesions, and examines how these effects vary according to biopsy timing. Time-to-event analysis was performed using a Cox proportional hazards model with the biopsy-to-procedure interval as the underlying time axis to identify variables associated with the hazard of intermediate/high technical difficulty, including lesion size, submucosal fibrosis, and lifting adequacy as covariates. 

A total of 150 lesions were included (69 naïve and 81 previously biopsied). The mean biopsy-to-procedure interval was 112 ± 81 days in biopsied lesions (p < 0.001). Biopsied lesions were larger (4.41 ± 3.19 cm vs. 3.57 ± 1.69 cm; p = 0.04). Lesion location distribution was similar between groups across the colorectum (p = 0.4). Resection type differed (EMR: 81% vs. 72%; ESD: 19% vs. 28%; p = 0.2), respectively. Among EMR cases, biopsied lesions had lower en-bloc resection (4.9% vs. 19%; p = 0.007), higher hot piecemeal use (63% vs. 43%; p = 0.02), and lower cold piecemeal use (4.9% vs. 19%; p = 0.007). Planned hybrid EMR occurred only in biopsied lesions (4.9%, p = 0.12). Use of adjunctive techniques was similar between groups: hot avulsion was used in 2.9% naïve vs 2.5% biopsied lesions (p = 0.9), CAST in 5.8% vs 7.4% (p = 0.8), and submucosal release in 2.9% vs 6.2% lesions (p = 0.5). Lifting adequacy was similar between groups (p = 0.9), good lifting was achieved in 93% naïve vs 90% biopsied lesions, partial lifting in 4.3% vs 6.2%, and no lifting in 2.9% vs 3.7%. Submucosal fibrosis was more common in biopsied lesions (F0: 65% vs. 84%; F1: 23% vs. 12%; F2: 11% vs. 4%; p = 0.03). Technical difficulty differed significantly (p = 0.002), with high-difficulty resections more frequent in biopsied lesions (26% vs. 4%). Technical success rates were 97% in naïve lesions and 91% in biopsied lesions (p = 0.2). Total procedure time was longer in biopsied lesions (105 ± 125 min vs. 74 ± 67 min; p = 0.02). Adverse events were infrequent and similar between groups: major early bleeding (0% vs. 0%), major delayed bleeding (0% vs. 0%), and deep mural injury without perforation (7.4% vs. 4.3%; p = 0.5), and perforation (3.7% vs. 5.8%; p = 0.7). Sydney DMI classification distributions were comparable (p = 0.2). Most lesions were Type 0 (69% vs. 83%). Length of stay did not differ significantly between groups (p = 0.3); hospitalizations of 3 or 7 days occurred only in biopsied lesions (3 days: 1/81 [1.2%]; 7 days: 1/81 [1.2%]), whereas no naïve lesions required admission beyond 2 days. Using the biopsy-to-procedure interval as a proxy for healing time, severe submucosal fibrosis was associated with a higher hazard of intermediate/high technical difficulty (HR 2.41, 95% CI 1.15–5.06; p = 0.020), supporting that biopsied lesions, especially with shorter intervals from biopsy to resection, are more fibrotic and technically more challenging to resect, while mild fibrosis showed a non-significant trend in the same direction (HR 1.62, 95% CI 0.92–2.87; p = 0.094). Kaplan–Meier curves indicated that higher technical difficulty was clustered among procedures performed shortly after biopsy, while lesions with longer biopsy-to-procedure intervals were less likely to reach intermediate or high technical difficulty (p < 0.001).

Biopsied lesions demonstrated significantly more fibrosis and higher technical difficulty, especially when the biopsy-to-resection interval was short. These findings reinforce current guidelines that large colorectal lesions should not be biopsied prior to referral, as biopsy adversely impacts resection feasibility without improving diagnostic or therapeutic outcomes.