Association between extent of colonic mucosal sialomucin change and subsequent local recurrence after curative excision of primary colorectal cancer.

Two interrelated studies were carried out to determine whether extent of sialomucin change adjacent to a primary colorectal carcinoma predicted local tumour invasiveness and risk of local recurrence. In the first, depth of tumour penetration was correlated with the length of the sialomucin band adjacent to 72 primary colorectal cancers. There was a significant (P less than 0.05) increase in sialomucin band length adjacent to tumours invading adjacent structures compared with those which had not (Mann-Whitney U test), although there was no overall correlation between depth of penetration, Duke's classification or degree of differentiation (Kruskal-Wallis test). A sialomucin band of greater than 3 cm was associated with a 70 per cent probability of adjacent structure (T4) invasion. These observations were then tested prospectively in a second study involving 256 patients to determine whether the presence of a greater than 3 cm sialomucin band could predict local recurrence. Presence of a greater than 3 cm sialomucin band was a significant (x2 = 7.12, d.f. = 1, P less than 0.001) and independent predictor of local but not distant recurrence. In addition both the interval to local recurrence and survival were significantly shorter if a greater than 3 cm sialomucin band was present. However the accuracy of greater than 3 cm sialomucin band as a predictive test for local recurrence was only 70 per cent. The extent of sialomucin adjacent to a primary colorectal cancer does provide a crude assessment of tumour invasiveness and risk of local recurrence.