Abstract LB-240: Immuno-genomic detection and prognostication of aggressive prostate cancer phenotypes by next-generation RNA sequencing

Prostate cancer (PC) is the most common non-cutaneous malignancy in men. The prediction of outcome based on multicore biopsy specimens is problematic, mainly due to tumor multifocality compounded by intratumoral heterogeneity. Since 1) cancer assays compare a patient’s profile to that of individuals unknown to have PC, and 2) baseline signatures of the diseased man are specific to his genome while those of the controls are specific to theirs, such intrinsic inter-individual genomic differences have impeded valid disease signature identification. Thus, methods that filter out inter-individual “noise” not related to the disease should enhance the identification of a robust prognostic PC signature. OBJECTIVE: Assess the ability of a novel immuno-genomics blood-based RNA expression assay for PC prognosis. The method, which uses a proprietary algorithm to interrogate CD2 and CD14 cells, is a real-time surveillance of gene expression changes consequent to PC that filters out intrinsic inter-individual genomic signatures not related to PC. We expect this Subtraction-Normalized Expression of Phagocytes (SNEP) approach to be valuable in making apt clinical decisions and stratifying patients with aggressive PC (need life-saving treatments) from those with indolent disease (safe for active surveillance). PATIENTS: Men were eligible for enrollment if they 1) were determined by their physician to have a risk profile that warrants a prostate biopsy (Pre-Biopsy blood draw), 2) had a biopsy >90 d prior to but METHODS: Peripheral blood samples (n = 713) were collected in purple top Vacutainer tubes. CD2 and CD14 cells were isolated 4 h later using Miltenyi’s MACS Microbeads and mRNA was extracted using the Qiagen’s miRNEasy Mini Kit. The raw (FASTQ) RNA-Seq data (Illumina) was trimmed (Trimmomatic), aligned (Bowtie2), and quantified (Express). Transcripts with quantifiable expression changes between samples from the two cell types were identified following differential expression analysis using a linear model and cell type as endpoints. This resulted in the final set of 10,643 transcripts with 1.5 absolute fold change. RESULTS: We identified genomic signatures in biopsy-positive patients that are predictive of Gleason grade, Cores Positive (CP), Maximum Involvement (MI), and an Aggressiveness Index (AI - an aggregate of the 3 endpoints). The signature scores were significantly associated with GG (tau 0.427, p 1.3x10-25), CP (tau 0.275, p3.3x10-11), MI (tau 0.564, p 8.5x10-44), and AI (tau 0.517, p 7.2x10-37). Interestingly, 1) certain transcripts were found to be specific to each endpoint (108 for GG, 88 for CP, 93 for MI, and 102 for AI), 2) some pairwise overlap was seen - highlighting the complementary of the three endpoints, and 3) no overall overlap was detected between all four endpoints. CONCLUSION: The multiple genomic signatures identified from CD2/CD14 RNA expression ratios - per SNEP assay - gives a prognostic summary that is comparable to prostate biopsy information including tumor grade, size/volume, and heterogeneity, and leads to the development of an aggressiveness signature that if validated will affect patient care decisions. Citation Format: Amin I. Kassis, Ricardo Henao, Kirk J. Wojno, Geffrey Erickson, Harry Stylli, Philip W. Kantoff. Immuno-genomic detection and prognostication of aggressive prostate cancer phenotypes by next-generation RNA sequencing [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr LB-240.