AbstractBackgroundTumour DNA circulates in the plasma of cancer patients admixed with DNA from non-cancerous cells. The genomic landscape of plasma tumour DNA has been characterised in metastatic castration-resistant prostate cancer (mCRPC) but the plasma methylome in mCRPC has not been extensively explored.MethodsmCRPC plasma samples collected from three different centres were concurrently subject to targeted genomic and pan-genome methylation profiling. The treatment courses and outcomes were also obtained. Tumour fraction of each plasma sample was estimated based on heterogyzous SNPs located in two truncal genomic deletions (8q21 and 21q22). Targeted methylome was performed using pre-designed capture panel followed by deep sequencing. We integrated genomic information with methylation data to extract methylation signatures associated with genomically-determined tumour fraction or private to individual`s tumour.ResultsPrincipal component analysis on the mCRPC plasma methylome indicated that the main contributor to methylation variance (principal component one, or PC1) was strongly correlated with genomically-determined tumour fraction (r=-0.96; P < 10-9), characterised by hypermethylation of targets of the polycomb repressor complex 2 components. Further deconvolution of the top PC1 correlated segments revealed that these segments comprised of methylation patterns specific to either prostate cancer or prostate normal epithelium. To extract information specific to an individual`s cancer, we then focused on an orthogonal methylation signature which revealed enrichment for androgen receptor (AR) binding sequences and where hypomethylation of these segments associated with AR copy number gain. Individuals harbouring this methylation pattern had a more aggressive clinical course, including a significantly shorter overall survival (HR = 8.18, 95% CI = 1.93-34.76, P = 0.0044).ConclusionsPlasma methylome analysis can accurately quantitate tumour fraction and identify distinct biologically-relevant mCRPC phenotypes associated with worse clinical outcome.Clinical trial identificationIstituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST), Meldola, Italy (REC 2192/2013) Royal Marsden, London, UK (REC 04/Q0801/6) PREMIERE trial (EudraCT: 2014-003192-28, NCT02288936).Legal entity responsible for the studyThe authors.FundingCancer Research UK (CRUK), Prostate Cancer Foundation (PCF).DisclosureV. Conteduca: Honoraria (self), Advisory / Consultancy: Bayer; Honoraria (self), Advisory / Consultancy: Astellas; Honoraria (self), Advisory / Consultancy: Janssen-Cilag; Honoraria (self), Advisory / Consultancy: Sanofi-Aventis. E. Gonzalez-Billalebeita: Honoraria (self): Astellas; Honoraria (self): Janssen-Cilag; Honoraria (self): Sanofi-Aventis. U.D. Giorgi: Honoraria (self), Advisory / Consultancy: Bayer; Honoraria (self), Advisory / Consultancy: Astellas; Honoraria (self), Advisory / Consultancy: Janssen-Cilag,; Honoraria (self), Advisory / Consultancy: Sanofi-Aventis. G. Attard: Honoraria (self): Institute of Cancer Research; Honoraria (self), Advisory / Consultancy: Astellas; Honoraria (self), Advisory / Consultancy: Medivation; Honoraria (self), Advisory / Consultancy: Janssen; Honoraria (self), Advisory / Consultancy: AstraZeneca; Honoraria (self), Advisory / Consultancy: Arno. All other authors have declared no conflicts of interest.
|