Background: Two recent studies conducted in one cohort (the ATBC Study) identified serum metabolites related to aggressive prostate cancer up to 20 years prior to diagnosis. Lipid and energy metabolites were significantly lower in cases than in controls, including glycerophospholipids, fatty acids, inositol-1-phosphate, alpha-ketoglutarate, and citrate. The present metabolomics investigation aimed to re-examine those associations in another cohort.

Methods: A nested case-control study of prostate cancer in the Prostate, Lung, Colorectal and Ovarian Cancer Screening (PLCO) Trial cohort was conducted based on 380 cases (298 stage 3-4 or Gleason sum 8+; “aggressive disease”), and 380 controls matched on age, race, study center, and date of baseline blood collection (samples were collected between 7am and 4pm). Median time from serum collection to prostate cancer diagnosis was 10.0 years (range, 4.4-17.0), and the majority of cases included in this study were diagnosed after the end of the PLCO screenings. Sera were analyzed on a high resolution accurate mass (HRAM) platform of ultrahigh performance liquid chromatography/mass spectroscopy (LC-MS) and gas chromatography/mass spectroscopy (GC-MS) (Metabolon, Inc.) that identified 722 metabolites. Conditional logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) of risk associated with a one standard deviation (1-SD) increment in metabolite concentration.

Results: Of the 27 metabolites associated with prostate cancer at p<0.05, 12 were amino acids or dipeptides, and amino acids as a metabolite class were somewhat overrepresented among the top signals for stage 3-4 but not overall prostate cancer (p = 0.003 and p = 0.08, respectively). Pyroglutamine, gamma-glutamylphenylalanine, phenylpyruvate, N-acetylcitrulline, and stearoylcarnitine yielded the strongest metabolite prostate cancer risk signals (1-SD increment ORs 0.78, 0.76, 0.73, 0.80, and 1.25, respectively; 0.001 < p < 0.006). Findings were similar for aggressive disease. We did not replicate our earlier findings of inverse associations with aggressive prostate cancer for the energy metabolites alpha-ketoglutarate or citrate in the present study (ORs 1.17 and 1.16 (p = 0.10 and 0.03), respectively), or the positive associations for trimethylamine-N-oxide (TMAO) or thyroxin (ORs 0.88 and 0.84 (p = 0.13 and 0.04), respectively).

Conclusions: In this prospective case-control analysis of serum collected throughout the day without regard to fasting status, we did not find strong prostate cancer associations for lipid or energy metabolites. Rather, amino acids appeared overrepresented in relation to prostate cancer risk. Whether the baseline and subsequent prostate cancer screening with PSA and digital rectal exams or the non-fasting status of the parent cohort contributed to the findings should be examined in other similar studies.

Citation Format: Alison M. Mondul, Stella Koutros, Stephanie J. Weinstein, Sonja I. Berndt, Demetrius Albanes. Serum metabolomic profiling of prostate cancer risk in the PLCO Cancer Screening Trial. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4287.