Introduction

Prostate cancer is the third most common cause of cancer-related death in U.S. men[1]. Radical prostatectomy or external beam radiation therapy for patients with high-risk prostate cancer can lead to serious adverse effects including urinary and rectal toxicities [1]. ⁹⁰Yttrium (⁹⁰Y) radioembolization offers the potential to deliver high-dose radiotherapy for prostate cancer with minimal non-target radiation[2]. While this novel approach is currently being carried out, evaluation of ⁹⁰Y radiation effects upon the prostate based on the biodistribution of ⁹⁰Y following radioembolization is imperative. This preliminary study aimed to assess the biodistribution of ⁹⁰Y and the relationship to acute response of prostate tissue to ⁹⁰Y prostate arterial radioembolization using PET/MRI in a dog prostate model.

Methods

Canine model: All the experiments are approved by the Institutional Animal Care and Use Committee. Six male 10-14 kg castrated beagles were treated with hormones for 3 months to generate benign prostatic hypertrophy. Each dog was catheterized through femoral artery under fluoroscopy and ⁹⁰Y catheterization of the prostatic artery was performed and confirmed by cone beam CT. ⁹⁰Y microspheres (~100 Gy) were administered.

PET/MRI: Beagles were anesthetized with isoflurane and scanned on the second day (~20 hours) post ⁹⁰Y arterial administration with a 3T PET/MRI (Biograph mMR, Siemens, Germany). MRI-based attenuation correction was applied using DIXON-VIBE sequences. PET of prostate comprised a 60-min list mode acquisition, during which MRI protocols were performed. Multiparametric MRIs were acquired with diffusion-weighted and dynamic contrast-enhanced (DCE) MRI. Diffusion imaging was performed using a diffusion-weighted SE-EPI sequence with b-values of 400, 600, 800 s/mm². Gadopentetate dimeglumine was injected intravenously as a rapid bolus (0.1 mmol/kg) at a rate less than 10 mL per 15 seconds. Contrast uptake was followed for 5 minutes with DCE-MRI. The DCE-MRI acquisition consisted of a 3D VIBE dynamic acquisition with TR/TE = 4.47/1.81 ms, FOV=159 x 159 mm², matrix size = 192 x192, flip angle = 12°, slice thickness = 3 mm, and a temporal resolution of 2.8 sec. PET data was aligned with MR images on Siemens workstation. Image post-processing was performed in MATLAB (MathWorks). Apparent diffusion coefficient (ADC) map was generated with a mono-exponential fitting. Initial area under the curve (IAUC) for Gd was calculated with integration times of 30 and 180 second post–contrast infusion (IAUC₃₀ and IAUC₁₈₀)[3].

Results

PET/MRI images represent areas of increased radioactivity reflecting the biodistribution of ⁹⁰Y microsphere (Fig. 1c, and Fig.2a), where ⁹⁰Y microspheres were found distributed in the treated prostate lobe. Both representative ADC (Fig. 2b) and IAUC₃₀ (Fig. 2c) maps showed a similar pattern as the fused PET/MRI images (Fig. 2a) with enhancement of the treated lobes. However, slight differences were observed as expected. Relative Gd concentration curve (Fig. 2d) showed the ⁹⁰Y treated regions higher signal enhancement compared to the normal regions with region of interest drawn from fused PET/MRI image. The calculated IAUC values showed changes in treated and control regions of the prostate while those of neighboring muscle remained stable (Fig. 3).

Discussion

The clinical ability to visualize and quantify ⁹⁰Y microsphere deposition in prostate cancer would benefits dose optimization to maximize tumor kill while limiting adverse effects on normal tissues. Acute phase response of cancer tissue could predict long-term effects of treatment in patients. PET/MRI offers the potential to detect in vivo ⁹⁰Y biodistribution and simultaneously evaluate tissue response following ⁹⁰Y arterial radioembolization of the prostate.

Conclusions

With the current study we have demonstrated the potential to build the relationship between tumor ⁹⁰Y biodistribution and therapeutic response prediction for ⁹⁰Y radioembolization therapy. Quantitative PET/MRI in future studies intend to quantify localized microsphere concentrations in vivo and correlate with long-term therapeutic effects in order to provide a significant improvement of patient-specific dosimetry associated with ⁹⁰Y microspheres.

Acknowledgements

ACKNOWLEDGEMENTS: This research was supported by NCI grant Number RO1CA181658 and grant from BTG.