Abstract Epithelial ovarian cancer (EOC) is still one of the deadliest malignancies in women frequently involving peritoneal tumor spread and ascites. Despite the heterogeneity of EOC, patients are mostly treated with standard therapy (cytoreductive surgery and platinum based chemotherapy). Understanding molecular mechanisms of EOC and its peritoneal metastasis is essential to develop new targeted therapies. In this study, we extended the transcriptome analysis of large RNA (Auer et al. Oncotarget 2015, 6:17261) to small RNA including micro RNAs (miRNAs) and piwi protein interacting RNAs (piRNAs). The transcriptome was analyzed concerning peritoneal tumor spread as introduced in our previous study: miliary – i.e. numerous small millet-like tumor seedings - versus non-miliary – i.e. big, bulky implants. Small RNA-sequencing (sRNA-seq) was performed with a total of 43 tumor cell enriched samples isolated from ascites and solid tumors from 23 chemo-naïve high grade serous EOC (HGSOC) patients and analyzed together with corresponding RNA-seq data from our previous study. qPCR was applied for validation of sRNA-seq data (n=48). Expression differences were determined (I) between miliary and non-miliary tumor spread in (Ia) ascitic and (Ib) solid tumor cells and (II) between ascitic and solid tumor cells in (IIa) miliary and (IIb) non-miliary on three levels: single miRNAs, gene-miR sets, and general types of RNAs (especially those of the competing endogenous RNA network, ceRNA). sRNA-seq yielded evidence for expression of 649 known and 2,306 newly predicted miRNAs and 22,987 known and 783 newly predicted piRNAs. qPCR validation rate was 90% for 48 selected sRNAs. (Ia) 401 miRNAs, but no gene-miR set and (Ib) one miRNA, but >19,000 gene-miR sets were differentially expressed between miliary and non-miliary in ascitic and solid tumor cells, respectively. 72% of these deregulated gene-miR sets were characterized by down-regulated coding genes and up-regulated miRNAs. (IIa) 109 miRNAs and 49 gene-miR sets and (IIb) 55 miRNAs and no gene-miR set were differentially expressed between ascitic and solid tumor cells in miliary and non-miliary, respectively. These numbers indicate varying strategies for differential expression between miliary and non-miliary, one the one hand, and between e. g. ascitic and solid tumor cells, on the other hand: thousands of small changes in miRNA and gene expression versus few, but greater changes in individual miRNA and gene expression. Global analyses of coding, long non-coding (lncRNAs), circular (circRNAs) RNAs, and miRNAs revealed a down-regulated ceRNA network (i.e. less circRNAs and lncRNAs compared to coding RNAs), but an unchanged total amount of miRNAs. Given the sponge function of the ceRNA network, this indicates more free miRNAs available for regulation in miliary compared to non-miliary. In addition, a 13 sRNA signature was developed to predict the peritoneal tumor spread type from FFPE tissues. It was used for validation of the impact on overall survival (OS) in an independent cohort of 32 HGSOC patients. In accordance with our previous study, miliary tumor spread had a negative impact on OS independent of age and FIGO stage. We are the first to perform an integrative analysis of large and small RNA-seq data of ascitic and solid tumor cells from 23 HGSOC patients. We found evidence for hundreds of known mi- and piRNAs and thousands of newly predicted mi- and piRNAs. We could show a negative prognostic impact of the miliary tumor spread in a validation cohort. Solid tumors of the miliary type seem to be more influenced by miRNA regulation than solid tumors of the non-miliary type which implies different spread strategies. This fundamentally different mechanism of gene regulation and the different prognosis of these two spread types underline the need for alternative treatment regiments. Citation Format: Anna Bachmayr-Heyda, Katharina Auer, Nyamdelger Sukhbaatar, Stefanie Aust, Simon Deycmar, Agnes T. Reiner, Stephan Polterauer, Sabine Dekan, Dietmar Pils. Peritoneal tumor spread in high grade serous ovarian cancer: An effect of the competing endogenous RNA network? [abstract]. In: Proceedings of the AACR Special Conference on Advances in Ovarian Cancer Research: Exploiting Vulnerabilities; Oct 17-20, 2015; Orlando, FL. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(2 Suppl):Abstract nr B56.