Marking hypoxia in rat prostate carcinomas with beta-D-[125I]azomycin galactopyranoside and [Tc-99m]HL-91: Correlation with microelectrode measurements

The purpose of this study was to determine, with a rodent tumor model, if m icroelectrode measurements of unmodulated tumor oxygenation predict for the avidity of hypoxic markers to tumor tissue. Methods: The rapidly growing, anaplastic variant of the Dunning rat prostate carcinoma cell line (R3327-A T) was implanted subcutaneously on the upper backs of Fischer X Copenhagen rats. Approximately 100 measurements of Po-2 were obtained from tumors of 5 -10 g in animals that were restrained and then subjected to different anest hetic procedures. Values of median Po-2 (in mm Hg) and percentage of measur ements (5 mm Hg obtained from individual tumors were used to define tumor o xygenation status. The radiodiagnostic hypoxic markets P-D-iodinated azomyc in galactopyranoside (IAZGP) and [Tc-99m]HL-91 were simultaneously administ ered to 26 animals whose tumor oxygen levels had been measured. Six hours a fter marker administration, the animals were killed; tumor, blood, and musc le tissues were sampled; and percentage injected dose per gram (%ID/g*), tu mor/blood ratio (T/B), and tumor/muscle ratio (T/M) parameters were determi ned. Parameters of marker avidity to individual tumors were linearly correl ated with microelectrode measurements of tumor oxygenation to determine the significance of inverse associations. Results: The median Po-2 Of 41 tumor s varied from 2.0 to 20.9 mm Hg, with an average value of 7.5 +/- 1.4 mm Hg . Six tumors had unusually high values; that is, >10 mm Hg, and when these were excluded from the analysis, the average median Po-2 of the remaining 3 5 was 4.3 +/- 0.7 mm Hg. When electrode measurements of tumor oxygenation w ere obtained under conditions of halothane anesthesia with the animals brea thing O-2, carbogen, or air, median Po-2 values increased significantly (P = 0.001). When animals were deeply anesthetized by intraperitoneal injectio n of ketamine-xylazine, median Po-2 values were not significantly different (P = 0.13) from those obtained while the animals were restrained and breat hing air. There was no inverse correlation of significance between the elec trode measurements of median Po-2 and the avidity of beta -D-IAZGP nor [99m Tc]HL-91 in this tumor model. The range of median Po-2 values in these tumo rs was at least 3 mm Hg, and the range of hypoxic marker avidity was less t han twofold. Conclusion: These data demonstrate that microelectrode measure ments of rat tumor oxygenation did not correlate with the avidity of the tw o hypoxic markers, at least in this tumor model. The larger dynamic range o f tumor oxygen measurements obtained with microelectrodes might be biased t o low values by their necrotic fractions, the zones within solid tumors tha t contain dead cells and debris that will not be labeled by bioreducible hy poxic markers. Hypoxic marker avidity to individual tumors will have to be validated by other assays that can predict for their radiosensitivity.