Prognostic value of nuclear morphometry in colorectal cancer

In many types of cancer, certain morphometric characteristic of tumor cells correlate with patient survival. Our observations suggested that the survival of patients with colorectal carcinomas is negatively correlated with tumor-cell nucleus size.

We investigated relationships between postsurgery survival and nucleus morphometrics in 90 patients who had undergone resection for a colorectal tumor. The nucleus-size variables considered were maximum diameter, minimum diameter, perimeter, area, and form factor (means for 100 nuclei from each patient were used in all cases).

Our results confirmed that patients with large maximum nucleus diameter (where large = greater than the first quartile) have significantly worse survival than patients with smaller maximum nucleus diameter (mean survival, 28 vs. 43 months). Similar results were obtained for the other nucleus-size variables. Stepwise Cox regression analysis was then performed, with postsurgery survival time as the dependent variable and the following candidate independent variables: age, gender, Dukes class, degree of histologic differentiation, the various nucleus-size variables, and relative frequencies of different nucleus shapes (spherical, oval, cylindrical, fusiform, and irregular). The variables selected for the prognostic model were Dukes class, relative frequency of irregular nuclei, and maximum nucleus diameter.

These findings indicated that nucleus size and shape are useful predictors of survival. Even if Dukes class is known, consideration of nucleus size and shape significantly improves prediction of survival.