Intratumoral heterogeneity in colorectal carcinoma: trucut sampling for DNA ploidy analysis

Solid tumors, such as colorectal carcinomas, consist of cell subpopulations that differ both genetically and in their clinical behavior. Many authors have examined cell kinetics and DNA content in colorectal tumors in correlation to clinical and pathological variables with different results. The interpretation of those results present some difficulties related to tumor heterogeneity that to date are unsolved. Our study is based on a new method of colon cancer sampling for DNA content determination. The aim of this work was to reduce the risk of incorrect DNA evaluation due to tumor heterogeneity.

Our study was based on eleven selected cases of T3 colorectal carcinoma. Fresh surgical specimens from the primary tumor site were taken during surgery. For each case at least four samples were taken using a 23 gauge trucut from the outside of the serosa through the tumor to the lumen of the colon. The specimens were stained according to a modified Feulgen method and DNA content was measured by image analysis. Three parameters were evaluated: DNA index, ploidy and proliferation level (considered as the sum of elements corresponding to the S and G2 phases).

One of the eleven (9.1%) tumors showed a diploid pattern; four out of eleven (36.4%) cases showed a tetra/polyploid pattern and six out of eleven (54.5%) cases showed an aneuploid pattern. Three tumors were monoclonal (27.3%), one diploid and two aneuploid. Eight were polyclonal (72.7%). Considering the single specimen, seven out of sixty-eight specimens (10.3%) were inadequate because of scanty material. Twenty-five out of the sixty-one adequate specimens (41%) were monoclonal and thirty-six (59%) were polyclonal. Five tumors (three monoclonal and two polyclonal) showed the same cell clones on all the examined samples. The remaining six tumors showed interregional variability. The six of the eight polyclonal cases (75%) multiple stem lines were evident, analyzing only one sample taken close to colon serosa, while in one case (25%) it was necessary to examine two samples in order to see the polyclonality of the lesion. When samples taken close to mucosa where analyzed, however, one sample was not enough to show tumor polyclonality in five of the eight polyclonal examined cases. Proliferation level varied greatly in different parts of the same carcinoma and did not correlate to the site from which the sample was taken.

In the present study, we demonstrated that DNA ploidy differences may exist between the superficial and the deep part of the same neoplasia and that tumor samples show a greater variability in the deeper layers. Using trucut samplings, it was possible to point out the majority of aneuploid cell populations close to the serosa. In conclusion, trucut biopsy permits full thickness sampling of the tumoral mass and allows, from few samples, to evaluate the multiple DNA stemlines present in different parts of a colorectal tumor.