Abstract C140: RNA expression patterns in serum microvesicles from patients with glioblastoma multiforme and ovarian cancer.

Abstract

Abstract Background: Recent studies have shown that RNA isolated from exosomes and other microvesicles (exoRNA) contain diagnostically relevant transcripts of tumor origin. In this study we sought to identify specific biomarkers of glioblastoma multiforme (GBM) and malignant ovarian cancer in the ribonucleic acid fraction extracted from microvesicles (exoRNA) isolated from the serum of affected patients. Methods: All samples were collected following informed consent in accordance with the appropriate protocols approved by the MGH Institutional Review Board (GBM) and the Umeå University ethics committee (ovarian tumors). Patients with primary GBM (N=9), normal controls (N=7); and pathologically proven benign (N=16), borderline (N=19), and malignant (N=29, stage I-IV, grade 1–3) neoplastic transformation of the ovaries were included. ExoRNA isolated from patient serum was amplified and analyzed by microarray analysis on Agilent 4×44K arrays (GBM) and Agilent 8×60K arrays (ovarian tumors). Differential expressions between control and GBM samples were validated by qRT-PCR in a separate set of samples (N=10 in both groups). Results: Array analysis of the amplified exoRNA yielded significant signals from at least 10,000 genes on the array for all samples. The expression profiles of the exoRNA from GBM patients were shown to be different from those of normal healthy volunteers. The most significant expression differences observed in the array analysis pertained to down-regulated genes (121 genes >2-fold down) in the GBM patient exoRNA, which was validated by qRT-PCR on several genes. Gene ontology analysis of the down-regulated genes indicated these are primarily mRNAs coding for ribosomal proteins and other genes related to ribosome production. Supervised analysis showed no significant differential gene expression between the benign, borderline, and malignant ovarian tumors. However, unsupervised Consensus Clustering yielded a clear segregation of the patients into two new groups (100% resampling consensus for 61 samples). Gene Set Enrichment Analysis (GSEA) showed a strong association of one of the two groups with genes linked to translation initiation (FDR q-value 0.04), and mRNA processing (FDR q-value 0.032). GSEA also confirmed a strong overlap of the genes separating the ovarian cancer samples into two groups with those discriminating GBM from healthy control individuals. Conclusions: Microvesicle isolated RNA from patients with GBM is significantly different from that of normal control individuals. The same genes support a binary classification of ovarian tumors suggesting that common biological processes may be in effect for these different tumor types. The common signature is hallmarked by messenger RNAs coding for ribosome production, which are significantly downregulated in GBM exoRNA and in one of two novel subgroups of ovarian tumors. While the biological role of this signature remains to be elucidated, we conclude that exosomal RNA expression profiling has the potential to serve as a clinically relevant diagnostic source of information about the biology, malignancy, and state of tumors. This underlines the diagnostic potential of exosomes not only in early diagnosis, but also in directing therapies, evaluating response and in situations where biopsy samples are difficult to access. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr C140.