Interlaboratory variation in oxygen tension measurement by Eppendorf "Histograph" and comparison with hypoxic marker

The median of pO2 values in tumor measured by Eppendorf "Histograph" with a needle-type electrode has been used as a prognostic indicator in cancer patients. However, it is not established that a pretreatment measured pO2 value can be used as a universal predictor of local control probability, because the variation in pO2 values, especially in hypoxic tissue, among institutes may not allow comparison of measured "absolute pO2 values." The purpose of this study was to examine the variation in oxygen tension measurement by Eppendorf "Histograph" among six laboratories using a single batch of mice and tumors and the same detailed protocol. These results were also compared to the immunohistochemical staining of 2-nitromidazole adducts.

C3H mice bearing FSaII murine fibrosarcoma subcutaneously were shipped to all laboratories, and the oxygen status in tumors and in normal subcutis was examined using Eppendorf "Histograph" and immunohistochemical hypoxic marker.

All laboratories showed that the FSaII tumor was hypoxic with at least 77% of measured points under 10 mmHg in pO2 and with a median pO2 value less than that of normal subcutis. These results were further confirmed immunohistochemically. These findings are interpreted as evidence that the pO2 values measured by Eppendorf "Histograph" can be useful. However, the median values of tumor pO2 varied from 1.5 mmHg to 5.6 mmHg among the laboratories, and pO2 of normal subcutis also varied from 28 mmHg to 38 mmHg. There were also significant differences in hypoxic fraction, defined as the fraction under a given oxygen partial pressure (i.e., under 2.5, 5, or 10 mmHg), among institutes.

Caution needs to be exercised in using the absolute, median, or distribution of pO2 values measured by the Eppendorf "Histograph" to compare the data between laboratories or to predict the radiation response in an individual subject.