Abstract 1962: Visualizing the efficacy of PI3K inhibitor combination treatment on tumor metabolism

Purpose: Many types of cancer exhibit an upregulation in glycolysis. Cancer therapies that target tumor metabolism include PI3K inhibitor combinations, which are currently in clinical trials. Real-time metabolic imaging can be possible using hyperpolarization, which is a signal enhancement technique for MRI that increases the signal by a factor of 10,000 or more. The purpose of this work was to determine whether the anti-metabolic effect of such therapies on preclinical breast cancer models could be visualized non-invasively by MRI of hyperpolarized 13C-pyruvate to lactate conversion. Methods: Experiments were carried out on a 9.4 T animal MRI scanner (Biospec 94/20, Bruker, Billerica MA), and all animal studies were done with IACUC approval. Two models of triple negative BRCA1 mutant breast cancer, either resistant (T1R1, n=4) or sensitive (T2S, n=4) to PARP inhibition, were implanted in the inguinal fat pad of mice. Animals with tumor sizes of approximately 10x10x10 mm3 were then imaged prior to and after four consecutive days of combination treatment with a PI3K inhibitor and PARP inhibitor. In order to visualize the interconversion of pyruvate to lactate in vivo, a bolus of hyperpolarized 13C pyruvate (Hypersense, Oxford Instruments, Oxfordshire UK) with concentration=100 mM and volume=200uL was administered to the mouse via tail vein. 13C MR images were acquired using echo-planar spectroscopic imaging (Flip angle=4 degrees; matrix size=16x16; resolution = 2.5mm; slice thickness=5mm; spectral points=512; spectral width=4kHz; repetitions=32) with a 13C radiofrequency transmit/receive surface coil (Bruker, MA) and overlaid on proton anatomical images obtained using a proton volume coil (78 mm diameter, Bruker, MA). Data were reconstructed (Mathematica, Wolfram) to obtain pixel by pixel time integrated pyruvate and lactate spectra for an axial slice. The area under the curve for both metabolites was calculated for all pixels and summed within the tumor to obtain a lactate to pyruvate ratio for each case. Statistical analysis was carried out using a paired t-test. Results: For the animals with T2S tumors, the lactate to pyruvate ratio reduced from 1.65 ± 0.44 to 0.84 ± 0.32 (mean ± std. dev.) after treatment with significance p = 0.0378. For animals with T1R1 tumors, the ratio before and after treatment was 0.91 ± 0.44 and 0.93 ± 0.20 (p = 0.9553). Conclusions: Noninvasive visualization of pyruvate to lactate interconversion can be achieved using 13C MRI,