The relationship between proliferative and oxygenation status in spontaneous canine tumors

Immunocytochemical markers have been applied to biopsy specimens from spontaneous canine tumors to assess the prevalence and spatial distribution of proliferating and hypoxic cells, and their "geographic" relationship to each other. Both types of cells have been implicated in the failure to locally control human tumors treated with radiation and chemotherapy.

For the detection of hypoxic cells, a rabbit polyclonal antibody raised against a protein-bound, hexafluorinated, 2-nitroimidazole, designated CCI-103F, was used. The unmetabolized drug must first be injected into the dog to allow time for hypoxic metabolism and cellular binding to occur. For the detection of proliferating cells, a mouse monoclonal antibody raised against an endogenous nuclear protein, the "proliferating cell nuclear antigen," or PCNA, was used. This protein is expressed in most actively proliferating cells, but not in quiescent ones. An indirect immunostaining technique was used to visualize these markers in the tissue sections, and image analysis was used to estimate the area fraction of positive staining in representative, low magnification microscope fields.

Tumors with both high and low hypoxic and proliferative area fractions have been identified. No systematic relationship between the prevalence of the two markers, nor of the relationship between tumor grade and proliferative fraction, could be established. Staining with the proliferation marker was more commonly found near blood vessels, but some "nests" of tumor cells apparently distant from vasculature contained many proliferating cells. Staining with the hypoxia marker tended to be distant from the vasculature and/or bordering regions of tumor necrosis, but some labeled cells appeared near blood vessels, and in the absence of necrosis. Staining of sequential sections, one with the proliferation marker and one with the hypoxia marker, indicated that the two cell populations overlapped to varying extents. Some incidental staining of canine normal tissues with both the proliferative and hypoxia markers was observed as well.

The immunochemical marker approach promises to be a useful tool to increase both our basic understanding of tumor physiology and the complex nature of tumor heterogeneity.