Endoscopic resection is a standard management option for dysplasia arising in patients with ulcerative colitis (UC). Despite achieving histologically clear resection margins, a subset of patients develop subsequent advanced neoplasia (AN), defined here as high-grade dysplasia and/or colorectal cancer. Conventional risk stratification relies mainly on histopathological assessment and procedural features, which do not capture genomic alterations in morphologically normal mucosa that may reflect underlying field cancerisation.
To determine whether genomic alterations present within histologically non-dysplastic resection margins are associated with later neoplastic progression following endoscopic resection of UC-associated dysplasia.
This is a retrospective cohort study of 51 UC patients treated at a tertiary referral centre who underwent endoscopic resection of dysplastic lesions. The cohort had a median follow-up of 1947 days. Eleven patients experienced local progression to advanced neoplasia during follow-up. The analysis included 56 dysplastic lesions and 91 adjacent histologically non-dysplastic margins sampled for genomic profiling.
Low-coverage whole-genome sequencing (lcWGS) was performed on dysplastic lesions and adjacent histologically non-dysplastic resection margins. The primary genomic metric assessed was somatic copy number alteration (CNA) burden. Phylogenetic analysis was also performed to examine clonal relationships between dysplastic lesions and adjacent margins. CNA burden and phylogenetic relationships were correlated with clinical outcomes, specifically progression to AN.
Margins that were histologically non-dysplastic but adjacent to lesions that later progressed to AN demonstrated a significantly higher CNA burden compared with margins from lesions that did not progress. This indicates the presence of genomic alterations in morphologically normal-appearing mucosa contiguous to dysplastic lesions in cases that later develop advanced neoplasia.
In multivariate analysis, a high CNA burden in the histologically non-dysplastic resection margin was independently associated with local progression to AN. By contrast, traditional measures such as the presence of histologically clear resection margins and whether the lesion was excised en bloc were not independently predictive of progression in this cohort.
These results demonstrate that histological clearance at the resection margin does not necessarily indicate removal of the underlying cancerised field in UC-associated dysplasia. Genomic evidence of field cancerisation can be present in morphologically normal margins and is associated with greater risk of subsequent neoplastic progression. The phylogenetic analyses performed support a clonal relationship between dysplastic lesions and genetically altered adjacent mucosa in progressing cases, consistent with field effects extending beyond the histologically apparent dysplasia.
Detection of a high margin CNA burden in histologically non-dysplastic resection margins could function as a biomarker to stratify patients after endoscopic resection of UC-associated dysplasia. Incorporating genomic assessment of margins may enable biologically informed post-resection surveillance, identifying patients who are at higher risk of developing AN despite histologically clear margins and who might benefit from more intensive follow-up.
This study is retrospective and conducted at a single tertiary referral centre. The sample size was 51 patients with 11 local progressions; further validation in independent and prospective cohorts would be required before clinical implementation. Details such as thresholds for defining high CNA burden or how genomic testing would be operationalised in routine practice were not reported in the source. The authors state that all data produced are available on reasonable request.
Genomic alterations in histologically non-dysplastic resection margins, specifically a high somatic CNA burden, are associated with subsequent local progression to advanced neoplasia after endoscopic resection of UC-associated dysplasia. Histological clearance alone may not exclude the presence of a cancerised field. Margin genomic profiling offers a potential biomarker to inform post-resection surveillance strategies in UC, pending further validation.