Defining absolute postoperative desmoid risk in familial adenomatous polyposis according to APC genotype and family history: retrospective cohort study.

APC

Pathogenic APC variants located 3' of codon 1399 (high risk) or 5' of or at this codon (low risk), with the risk of desmoid tumour after risk-reducing colectomy modulated by family history of desmoid disease

Desmoid disease is a major cause of morbidity and mortality in familial adenomatous polyposis, particularly after prophylactic colorectal surgery; APC genotype and family history are recognised risk factors, but previous sizeable studies report relative rather than absolute risks. This retrospective observational study, using records from a prospectively maintained registry, quantified the absolute risk of desmoid after risk-reducing colectomy by classifying patients according to the location of the pathogenic APC variant (3' of codon 1399: high risk; 5' of or at codon 1399: low risk), with at least 5 years of postoperative follow-up and no desmoid before or at the time of surgery. Among 48 high-risk patients, 30 (63%) developed a desmoid, with no significant difference between colectomy and proctocolectomy (54% and 70% respectively; p = 0.233), family history (present in 35 of 48 patients, 73%) being associated with a higher risk (80% versus 15% without family history; p < 0.01). Among 1,213 low-risk patients, 154 (12.7%) developed a desmoid, with no significant effect of the type of surgery (13.9% after proctocolectomy versus 12.1% after total or partial colectomy; p = 0.380), and family history was again associated with a higher risk (30% versus 10%; p < 0.01). The authors conclude that a family history of desmoid disease is an important determinant of postoperative risk, that the high risk associated with variants 3' of codon 1399 is largely confined to individuals with an affected first-degree relative, and that these data support an individualised approach to perioperative counselling and surgical decision-making.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

Moving from relative to absolute risks is the concrete contribution: without a family history of desmoid, postoperative risk appears relatively low whatever the genotype, which qualifies the idea of a uniformly high risk in carriers of variants 3' of codon 1399. This conclusion nonetheless rests on a small subgroup: 35 of the 48 high-risk patients had a family history, so those without one are few, and the comparisons (80% versus 15%) are crude tests, with no confidence interval or adjustment in the abstract. The absence of a difference between colectomy and proctocolectomy, like the particularly high-risk subgroup proposed for future chemoprevention trials, therefore needs confirmation in larger samples.

Analysis by Dr Thibaut Benquey