This is the first report of repeated tissue pO2
measurements of orthotopic 9L and C6 gliomas after a single radiation dose of 9.3 Gy and during carbogen breathing. The results have shown that the orthotopic 9L gliomas are well oxygenated, with a tissue pO2
of approximately 30–32 mm Hg. The tissue pO2
of subcutaneously grown 9L tumors has been reported to be <10 mm Hg. Cerniglia et al.
) reported a pO2
of <8 mm Hg in subcutaneously grown 9L tumors in rats using a phosphorescence quenching method. A median pO2
of 2 mm Hg was reported by Teicher et al.
) using Eppendorf in subcutaneously grown 100-mm3
9L tumors in rats. The tumor volumes of these experiments were similar to those of the orthotopic tumors in the present study. Experiments performed in our laboratory also indicated fairly hypoxic subcutaneous 9L tumors, with a pO2
of 4 ± 1 mm Hg (n
= 6). Wallen et al.
) suggested a difference in vasculature between intracerebral and subcutaneous 9L tumors as a likely reason for the absence of hypoxia in intracerebral 9L tumors. Our results provide evidence of the possible influence of such vascular differences on the tumor pO2
. The tissue pO2
of intracerebral 9L gliomas reported in the present study agrees with reports of a small radiobiologic hypoxic fraction (0–3%) in 9L gliomas (24
). These results indicate that the pO2
of the 9L tumors is site specific, and, therefore, care must be exercised in evaluating strategies whose outcome is dependent on the oxygen levels in tumors such as that of radiotherapy.
A single dose of 9.3 Gy resulted in significant oxygenation of intracerebral 9L tumors on Days 1 and 2 after irradiation. The magnitude and time of the oxygenation of individual 9L gliomas varied, highlighting the importance of pO2
measurement in individual tumors for the purpose of therapeutic optimization. No such increase in tissue pO2
of 9L tumors in the control group was observed. In contrast, the intracerebral C6 gliomas were relatively hypoxic, with a tissue pO2
of approximately 12–14 mm Hg. Guerin et al.
) have reported a significantly lower vascular density of C6 gliomas compared with 9L tumors. The vascular density of these tumors was also significantly lower than that of the normal brain. On the basis of that report, we speculate that the differences in the tumor pO2
of 9L and C6 gliomas are likely due to differences in their vasculature. In addition to the tumor vasculature, the tumor growth characteristics and oxygen metabolism of these tumors might play an important role.
Furthermore, C6 tumors irradiated with 9.3 Gy did not show any significant changes in tumor pO2
after irradiation. A similar tissue pO2
in the contralateral brain in the control and irradiated groups with either 9L or C6 gliomas confirmed that our procedure for hemisphere irradiation did not influence the contralateral brain of the rats. A significant decrease in the tissue pO2
of the tumors over time was likely a result of the compromised tumor vasculature with an increase in tumor volume. In addition, an increase in intracranial pressure has been reported with increases in the tumor volume of intracerebral tumors (28
). We anticipate that this might have resulted in a significant decrease in the contralateral brain pO2
on Day 5 in our experiments. The doubling time of intracerebral 9L and C6 tumors is reported to be 67 and 23 h, respectively (29
). In view of these findings, we suggest that perhaps the hypoxic and rapidly proliferating orthotopic gliomas might not show increased oxygenation after irradiation.
Our results indicated a significant increase in intracerebral tumor and contralateral brain pO2
during 60 min of carbogen breathing in both the 9L and the C6 groups. The baseline tissue pO2
of the contralateral brain of the 9L and C6 groups and their response to carbogen breathing was similar on the 5 days of experiments. However, a decrease in the baseline 9L tumor pO2
and its response to carbogen breathing was observed, with an increase in tumor size during the 5 days of measurements. In contrast, the baseline tissue pO2
of the C6 gliomas and the response to carbogen breathing was similar on all 5 days of the experiments. The use of carbogen as a potential radiosensitizer has been investigated in animal and human tumors with mixed results. In some studies, carbogen resulted in an increase in tumor blood flow and oxygenation, but in other experiments, it did not. It has been suggested that the effect of carbogen to increase tumor oxygenation depends on several factors such as tissue type and breathing time (8
). Our results agree with these observations and indicate that the tumor response to carbogen breathing is likely to vary among different tumor types. These results further highlight the need to monitor the tumor pO2
during such strategies to determine its effect in individual tumors and tumor types. This information could then be potentially used to individualize radiotherapy for enhanced efficacy.
The clinical implications of these results are several, although their value needs to be tested under clinically pertinent conditions. The overall message is that the effects of treatments on tumor pO2 can be profound but cannot be readily predicted. Therefore, although it seems logical to use approaches to enhance tumor pO2 to enhance therapy, this should be done on the basis of explicit knowledge of the effects of the treatment on the pO2 of that particular tumor type. Even better, it would be desirable to make the measurements in the particular patient under the conditions being used, rather than to rely on the expected results. Inasmuch as methods such as in vivo EPR oximetry are being developed that can monitor the tumor pO2 directly in patients, these should be used. This should be especially useful in clinical trials of new methods, so that the evaluation of the effectiveness of the new approach can be done with knowledge of the effectiveness of the approach in modifying the tumor pO2 in preclinical studies.