The role of multiploidy as unfavorable prognostic variable in colorectal cancer

To analyze the prognostic value of DNA multiploidy in a prospective study on frozen surgical tissue samples from primary colorectal cancer.

Survival data from eleven prospective studies collectively comprising about thirteen hundred patients showed that aneuploidy correlated with a 5-year disease-free survival (DFS) significantly poorer than diploidy, and showed the limited prognostic value of results from retrospective studies employing paraffin-embedded material.

Multiple tumor samples of fresh/frozen surgical tissues from 120 colorectal cancer patients who had undergone radical surgery were taken for flow cytometric analysis of DNA content, and proliferative activity, shown as percentage of cells in S-phase (%S). The minimum follow-up of this series was 30 months. Univariate and multivariate analyses determined the independent significance of both clinical and biological variable on DFS.

Values of %S equal to or higher than 17.3 correlated with a 5-year DFS poorer than values lower than 17.3 (44.5% vs 85.2% respectively; p = .03), even if only in patients younger than 64. The subgroup with multiploid tumors showed a significantly poorer 5-year DFS (44.5% vs. 62.6% in the non multiploid patients; p = .02). Subgrouping the Dukes'B stage alone by multiploidy, the difference in DFS was much more evident (31.2% vs. 68% respectively; p = .0004) and multivariate analysis showed multiploidy as the only significant variable. Above all, adjuvant therapy did not absolutely modify the unfavorable outcome of the multiploid Dukes'B patients.

The prospective evaluation of ploidy allowed us to identify a very high-risk subgroup of patients with multiploid tumors. This biological characterization was easy to demonstrate and, above all in node-negative patients, reliable and very effective in terms of prognosis. The presence of multiploidy should result in a more aggressive therapeutic approach in the adjuvant setting.