Hyperpolarised 13C-MRI identifies the emergence of a glycolytic cell population within intermediate-risk human prostate cancer

Abstract

Hyperpolarised magnetic resonance imaging (HP 13C-MRI) is an emerging clinical technique to detect [1-13C]lactate production in prostate cancer (PCa) following intravenous injection of hyperpolarised [1-13C]pyruvate. Here we differentiate clinically significant PCa from indolent disease in a low/intermediate-risk population by correlating [1-13C]lactate labelling on MRI with the percentage of Gleason pattern 4 (%GP4) disease. Using immunohistochemistry and spatial transcriptomics, we show that HP 13C-MRI predominantly measures metabolism in the epithelial compartment of the tumour, rather than the stroma. MRI-derived tumour [1-13C]lactate labelling correlated with epithelial mRNA expression of the enzyme lactate dehydrogenase (LDHA and LDHB combined), and the ratio of lactate transporter expression between the epithelial and stromal compartments (epithelium-to-stroma MCT4). We observe similar changes in MCT4, LDHA, and LDHB between tumours with primary Gleason patterns 3 and 4 in an independent TCGA cohort. Therefore, HP 13C-MRI can metabolically phenotype clinically significant disease based on underlying metabolic differences in the epithelial and stromal tumour compartments.

The potential of hyperpolarised 13C-MRI to target glycolytic tumour core in prostate cancer

“…This was a retrospective study of data acquired prospectively as part of an institutional review board–approved (National Research Ethics Service Committee East of England, Cambridge South, 16/EE/0205) study that included patients undergoing radical prostatectomy for biopsy-proven PCa [ 4 ]. All patients underwent 3 T pre-biopsy mpMRI and HP- 13 C-MRI, measuring pyruvate SNR, lactate SNR, TC-SNR, and the pyruvate-to-lactate exchange rate constant ( k PL ).…”

“…The slides were analysed using HALO v3.2.1851.266 (Indica Labs) and the FISH v2.2.0 module. Epithelial and stromal cell populations were differentiated using a fluorescent random forest tissue classifier [4].…”

“…Hyperpolarised [1-13 C]pyruvate MRI (HP-13 C-MRI) is a rapidly evolving metabolic imaging technique that can visualise real-time pyruvate-to-lactate conversion, catalysed by the enzyme lactate dehydrogenase (LDH) [1]. Several studies have shown increased [1-13 C]lactate labelling in localised prostate cancer (PCa), highlighting the added value of HP-13 C-MRI to standard-of-care imaging for assessing tumour aggressiveness and monitoring treatment response [1][2][3][4]. However, despite the promise shown by these early studies, there remain several technical challenges that require optimisation to ensure that HP-13 C-MRI assessment is reliable and reproducible prior to clinical translation of the technique.…”

“…The relationship between the area of high [1- 13 C]lactate signal and the underlying tumour area on pathology is unknown, and could vary from representing only the metabolically active tumour core, through detecting labelled lactate beyond the anatomical boundaries of the lesion. In the former case, direct lesion segmentation on HP- 13 C-MRI may be sufficient for accurate tumour metabolic phenotyping based on assessing the glycolytic volume, which is known to harbour more aggressive disease [ 2 , 4 , 6 ]. This is analogous to approaches used in positron emission tomography (PET) to delineate the metabolically active tumour volume based on thresholding the 18 F-fluorodeoxyglucose ( 18 F-FDG) uptake [ 7 ].…”

Multi-nuclear magnetic resonance spectroscopy: state of the art and future directions

“…The first clinical trials of HP-[1- 13 C]-pyruvate MRI and MRSI were successfully used in prostate cancer patients [ 36 ]. Recently, Sushentsev N et al [ 37 ] also found that HP 13C-MRI can noninvasively differentiate indolent from aggressive prostate tumors based on their characteristic metabolic features (Fig. 5 ).…”