DNA flow-cytometric analysis in colorectal cancer: a comparison of metastasizing and non-metastasizing tumours.

The most common cause of death in patients with colorectal cancer is metastatic liver disease. In order to identify patients at a high risk of developing hepatic secondaries from colorectal cancers, DNA content was measured in metastasizing colorectal primaries (Group I, n = 32) as well as in their subsequently resected liver secondaries and in sections of non-metastasizing colorectal cancers (Group II, n = 25). A modified interpretation system involving both a DNA index and percentage of cycling cells (those in S and G2 + M phases) was developed. DNA content was measured in paraffin-embedded sections by flow cytometry using internal controls (human peripheral blood mononuclear cells) and non-malignant tissue controls (19 patients with diverticular disease). In Group I there were significantly more tumours with both abnormal ploidy (aneuploid or abnormal tetraploid peak) and > 15% cycling cells compared with Group II (Chi-squared; P = 0.034). The combination of abnormal ploidy and > 15% cycling cells was superior to Dukes' classification for identifying metastasizing tumours (Logistic Regression; P = 0.047). However, it was not possible to discriminate between the two groups using either DNA ploidy or the percentage of cycling cells alone. The metastasizing colorectal cancers exhibited similar DNA ploidy characteristics and had a similar percentage of cycling cells compared with their liver metastases. These results suggest that tumour DNA ploidy plus the percentage of cycling cells may predict the development of liver metastases and thus survival in patients with colorectal cancer.