Tumour cell proliferation under hypoxic conditions in human head and neck squamous cell carcinomas

SummaryTwo mechanisms of radiotherapy resistance of major importance in head and neck cancer are tumour cell repopulation and hypoxia. Hypoxic tumour cells that retain their clonogenic potential can survive radiation treatment and lead to local recurrences. The aim of this study was to quantify this cellular population in a cohort of human head and neck carcinomas and to investigate the prognostic significance. The proliferation marker iododeoxyuridine (IdUrd) and the hypoxia marker pimonidazole were administered intravenously prior to biopsy taking in patients with stage II–IV squamous cell carcinoma of the head and neck. Triple immunohistochemical staining of blood vessels, IdUrd and pimonidazole was performed and co-localization of IdUrd and pimonidazole was quantitatively assessed by computerized image analysis. The results were related with treatment outcome. Thirty-nine biopsies were analyzed. Tumours exhibited different patterns of proliferation and hypoxia but generally the IdUrd signal was found in proximity to blood vessels whereas pimonidazole binding was predominantly at a distance from vessels. Overall, no correlations were found between proliferative activity and oxygenation status. The fraction of IdUrd-labelled cells positive for pimonidazole ranged from 0% to 16.7% with a mean of 2.4% indicating that proliferative activity was low in hypoxic areas and occurring mainly in the well-oxygenated tumour compartments. IdUrd positive cells in hypoxic areas made up only 0.09% of the total viable tumour cell mass. There were no associations between the magnitude of this cell population and local tumour control or survival. Co-localization between proliferating cells and hypoxia in head and neck carcinomas was quantified using an immunohistochemical triple staining technique combined with a computerized simultaneous analysis of multiple parameters. The proportion of cells proliferating under hypoxic conditions was small and no correlation with treatment outcome could be found.