Tumor Hypoxia Assessment: In Vivo 3D Oxygen Imaging Through Electron Paramagnetic Resonance

We are looking into tumor microenvironment and role of hypoxia in tumor development in response to therapy. Hypoxia is crucial not only in cancer but also in many other pathologies, such as cardiac or neurological disorders. Our goal is to quantitatively determine oxygen levels in tissues of living animals in the context of different anti-cancer treatments and several preclinical tumor models.

Several techniques in APR are applied in oximetry, including continuous wave, pulse, or rapid scan. However, preclinical pulse imager, JIVA-25, enables precise and high spatial and oxygen resolution 3D whole-body oxygen mapping. The images are inquired in minutes and could be acquired over several days, corresponding to physiological changes in tissue, PO2, and also enables following changes in time.

One of the biggest challenges of electron paramagnetic resonance imaging is lack of the anatomical reference. In many laboratories either MRI or CT is used to acquire key anatomical features. However, the disadvantage is high cost and necessity of moving the animal.

To solve these problems we propose using ultrasound as easy to use, portable, and cost-efficient modalities. Our results show the feasibility of the ultrasound and EPR image fusion, which opens up the possibility of comparing functional and atomic images at different time points. For example, EPR oxygen maps could be registered with US tissue anatomy, vascular network, or tissue perfusing using microbubbles as a contrast agent.

This technology helps us to discover hypoxia regions within tumors and design dose painting in radiotherapy. It also allows assessment of the efficiency of anti-hypoxia treatments, such as oxygen microbubbles. Oxygen mapping may also be used as a marker for therapy efficiency at the early stage of treatment, and appropriate patient classification for treatment.

To begin, secure the tumor-inoculated anesthetized mouse on the animal bed. Administer one milliliter of physiological saline subcutaneously to the mouse to maintain hydration. Use a respiratory pillow sensor and a surface thermometer using vinyl polysiloxane dental clay to monitor the animal's heart rate and temperature.

Insert polytetrafluoroethylene tubing intraperitoneally for spin probe administration and secure the cannula using vinyl polysiloxane dental clay to prevent it from dislodging from the abdominal cavity. Then insert a urine catheter to collect the excreted spin probe. For anatomical ultrasound imaging, rotate the mouse inside the animal bed and secure its position using vinyl polysiloxane dental clay.

Transfer the animal to a 3D controlled table prior to electron paramagnetic resonance measurement. Confirm that the animal is secured and not in contact with the heating platform. Fix the bed holder position in three dimensions to prevent rotation, especially in the XZ or sagittal plane, which is essential for accurate registration.

Next, using a 0.35 millimeter fishing wire on the tape, place a position marker on the bed holder to mark the beginning of the resonator. Conduct B-mode imaging manually with a one millimeter step in both the axial and sagittal directions towards the Y-axis. Image the tumor structure using parameters specific to the transducer frequency.

After the anatomical ultrasound imaging, transfer the immobilized mouse in the animal bed promptly to the electron paramagnetic resonance or EPR imager. Carefully maintain the position of the animal bed to minimize rotations within the resonator. Reconnect the temperature probe to ensure continuous monitoring of the animal's temperature, aligning it with the resonator's internal temperature.

In the EPR spectrometer software, adjust the resonator with the tuning wheel to center the frequency around 725 megahertz. To optimize the microwave power, adjust the attenuation from 20 to three decibels until a peak at 60 nanoseconds is achieved. Tune the instrument to center the free induction decay of fiducials placed within the animal bed within the magnetic field and phase it accordingly.

Then administer 100 microliters of OXO 71 intraperitoneally via the previously inserted cannula, followed by flushing with 50 to 100 microliters of saline solution. Use a programmed Q sequence to acquire measurements with specific parameters, including T1 and T2 relaxation times, 3D electron spin echo for fiducial imaging, and 4D inversion recovery electron spin echo for imaging the animal. After imaging, perform data analysis and visualize the tumor mask within the oxygen partial pressure map of the mouse and export the values for each voxel.

The ultrasound cross-section showed the LN 229 tumor growing in the intrascapular fat pad with vasculature both within and outside the tumor boundary. Tumor ultrasound B-mode images provided an anatomical reference, showing tumor surface and markers for aligning with EPR-OI. Cross-sectional images of the 3D oxygen partial pressure map displayed notable oxygen level heterogeneity, with more hypoxic areas within the tumor and a highly hypoxic fiducial marker at the image's bottom.

The images of tumors in varying sizes demonstrated that larger tumors contained more voxels, while the smallest tumor showed a suboptimal oxygen partial pressure map due to insufficient spin probe concentration.