Session Item

EBRT to the prostate and pelvic lymph nodes (PLNRT) is part of the curative treatment of high-risk prostate cancer (PCa). However, we do not know how radiotherapy influences the overall metabolism of patients. This question has become increasingly topical with improvements in radiotherapy accuracy, which have led to rising interest in PLNRT. To better understand the impact of radiotherapy on metabolic profiles, we conducted deep metabonomic analysis in patients undergoing EBRT to the prostate and pelvis for high-risk PCa.

We sampled urine, serum and stools in 32 patients at 6 timepoints: baseline, 2/3 and 4/5 weeks of PLNRT; and 3, 6, and 12 months after PLNRT. We characterised the whole metabolome in urine/serum with nuclear-magnetic resonance (NMR) and mass spectrometry (MS), and in stools (faecal water; FW) with NMR. We used linear mixed-effects modelling to remove subject-specific variance from data. Partial least squares (PLS) discriminant analysis was used to model changes between timepoints for each biofluid and assay. Significant metabolites were determined by cross-validation for valid models (α=0.05).

After removal of individual variation, we found that metabolites in urine, serum and FW changed significantly after PLNRT initiation (figure 1). In urine and serum, some recovery was observed after treatment completion however not to baseline levels, as significant metabolic differences were observed between baseline and 12 weeks post-PLNRT, which subsequently remained stable. In FW, no recovery was observed. When analysing specific metabolites over time we observed that at 12 weeks the metabolic state had not returned to baseline, which was sustained until one year post-treatment completion.


Figure 1: Changes in overall metabolic profiles over time in patients undergoing PLNRT after removal of individual variance. Only assays where robust significant changes were found are shown (Positive Q2, Permutation test value < 0.01). Green cells show significant changes in metabolic profiles. Red cells show no or non-significant changes. MS: mass spectrometry. LPOS and LNEG: lipid (tailored to detect complex-lipid species) positive and negative ionisation modes. HPOS: hilic (tailored to detect small and polar-molecules) positive ionisation mode. RPOS and RNEG: reverse-phase (tailored to detect small molecules of moderate hydrophobicity) positive and negative ionisation modes.  NMR: Nuclear Magnetic Resonance.  standard 1D: standard one dimension pulse program. CPMG: Carr-Purcell-Meiboom-Gill pulse program (to specifically supress macromolecule signals).

We show for the first time that, when individual variation is removed, a lasting overall metabolic impact is observed in patients undergoing EBRT for high-risk prostate cancer. This observation has potential in terms of treatment response biomarker discovery. We are also exploring relationships with our recently published microbiota/toxicity observations in this group of patients.