In vivo imaging of changes in tumor oxygenation during growth and after treatment

Abstract

A novel procedure for in vivo imaging of the oxygen partial pressure (pO2) in implanted tumors is reported. The procedure uses electron paramagnetic resonance imaging (EPRI) of oxygen-sensing nanoprobes embedded in the tumor cells. Unlike existing methods of pO2 quantification, wherein the probes are physically inserted at the time of measurement, the new approach uses cells that are preinternalized (labeled) with the oxygen-sensing probes, which become permanently embedded in the developed tumor. Radiation-induced fibrosarcoma (RIF-1) cells, internalized with nanoprobes of lithium octa-n-butoxy-naphthalocyanine (LiNc-BuO), were allowed to grow as a solid tumor. In vivo imaging of the growing tumor showed a heterogeneous distribution of the spin probe, as well as oxygenation in the tumor volume. The pO2 images obtained after the tumors were exposed to a single dose of 30-Gy X-radiation showed marked redistribution as well as an overall increase in tissue oxygenation, with a maximum increase 6 hr after irradiation. However, larger tumors with a poorly perfused core showed no significant changes in oxygenation. In summary, the use of in vivo EPR technology with embedded oxygen-sensitive nanoprobes enabled noninvasive visualization of dynamic changes in the intracellular pO2 of growing and irradiated tumors.

FIG. 1

Internalization of the oxygen-sensing probe particulates in RIF-1 cells, and their effect on cell viability and tumor growth. a: Phase-contrast microscopic image (1500× magnification) of RIF-1 cells internalized with probes. The nanoprobes are seen as black aggregates in the cell. The arrow indicates a large (micron-size) uninternalized single crystal of the probe. b: Tumor growth curves are shown for tumors grown with cells untreated (control; N = 5) or internalized with LiNc-BuO probe (N = 7). The tumor volume was calculated using measurements of three orthogonal dimensions. The tumors, in general, showed similar growth patterns. However, the data from the internalized group showed an overall reduced growth rate as compared to the control group. Inset in b: Effect of internalized particulates on cell viability. RIF-1 cells were cultured in the presence of extracellular LiNc-BuO crystals (size > 5 μm) for 24 h or intracellular LiNc-BuO crystals (size 270 ± 120 nm) and assayed at the end of a single or four passages. The results indicate that the particulates had no significant effect on the viability of RIF-1 cells in culture. c: Histology image under a high color threshold showing the distribution of the oxygen-sensing probes (indicated by arrows) in nonirradiated tumor tissue. The histology was obtained 12 days after implantation of cells internalized with LiNc-BuO nanoparticulates.

FIG. 2

Representative images (10 mm × 10 mm) of probe and oxygen distribution in a growing RIF-1 tumor. The images were obtained on day 5 (volume = 86 mm³) and day 9 (volume = 113 mm³) after the mouse was inoculated with RIF-1 cells internalized with the nanoparticulate spin probes. The probe is seen to be distributed in the core of the tumor, which is about 62% ± 9% of the tumor volume on day 5, and 42% ± 6% on day 9. Note that the oxygen information is obtained only from the regions where the particulates are present. The images show the presence of significantly hypoxic pockets in the region of the tumor under examination.

FIG. 3

Images of tumor oxygenation (pO₂) in an RIF-1 tumor taken when the mice were breathing room air or carbogen. The images (10 mm × 10 mm) were obtained on day 12 after the mice were inoculated (tumor volume = 125 mm³) with RIF-1 cells internalized with the nanoparticulate spin probes. The left panels show the distribution of the probe in the core of the tumor. Oxygen information is obtained only from regions where the particulates are present. The pO₂ image of the air-breathing mouse shows the presence of pockets of hypoxic regions in the tumor, which become more oxygenated during carbogen breathing. The histograms reveal a significant increase in tumor oxygenation upon carbogen treatment.

FIG. 4

Effect of X-ray irradiation on tumor oxygenation. Change and redistribution of tumor oxygen levels are shown for a small (volume = 113 mm³) RIF-1 tumor before (pre-) and 1 h after (post-) X-ray irradiation. The tumor was implanted with RIF-1 cells internalized with nanoparticulates of LiNc-BuO probe. A dose of 30 Gy was delivered with 6 MeV electrons at a dose rate of 3 Gy/min. The images were obtained on the day of irradiation. Data show a redistribution of pO₂ with a modest increase in the pO₂ values.

FIG. 5

Effect of X-ray irradiation on the oxygenation of a large tumor. Redistribution of tumor oxygen concentration is shown for a large (volume = 327 mm³) RIF-1 tumor before (pre-) and 1.5 h and 7.2 h after X-ray irradiation. The tumor was implanted with RIF-1 cells internalized with nanoparticulates of LiNc-BuO probe. A dose of 30 Gy was delivered with 6 MeV electrons at a dose rate of 3 Gy/min. The images were obtained on the day of irradiation. The data show a redistribution of pO₂ in the central core of the tumor. The data also show a significant decrease in the first 1.0 –1.5 h followed by an increase 7.2 hr after irradiation.

FIG. 6

Changes in tumor oxygenation levels following irradiation. Two groups of tumors were studied: (●) small tumors with an average volume of 114 ± 14 mm³ (N = 4), and (○) large tumors with an average volume of 250 ± 28 mm³ (N = 5). The tumors were obtained by inoculating the mice with RIF-1 cells internalized with nanoparticulates of LiNc-BuO spin probe. The right hind limbs containing the tumor were irradiated with a single dose of 30 Gy at a dose rate of 3 Gy/min. Insets: Blood-perfusion images of a small and a large tumor. The mice were infused with a solution of Patent blue to visualize blood perfusion in the tumor. a: Perfusion image of a small tumor (volume = 133 mm³). b: Perfusion image of a large tumor (volume = 413 mm³). The intensity of the blue color in the image represents the extent of blood perfusion in the tissue. The dark blue color in the images corresponds to the highly perfused muscle tissues surrounding the tumor (indicated by contour). The core of the large tumor is mostly pink, suggesting that the region is poorly perfused.