Abstract 281: Evaluation of circular RNA profiling in metastatic melanoma patients treated with immune checkpoint blockade

Melanoma comprises a variety of malignant cell types from the skin, the mucous membranes, and the pigmented cells of the eye. Its incidence and mortality rates continue to rise worldwide, mainly driven by increased ultraviolet light exposure. Although screening improvements have led to a better diagnosis in the early stages, the outcome for patients with Metastatic Melanoma (MM) remains poor and the survival rate is considerably low. Recently, Immune Checkpoint Inhibitors (ICB) have revolutionized the treatment of MM, although only 30% of patients enjoy clinical benefit and a plethora of different molecular mechanisms is postulated to cause resistance. Several biomarkers are currently on the spotlight as putative predictors of response; however, they are not sufficiently informative and robust. Circular RNAs (circRNAs) are a class of single-stranded stable RNAs produced by 5′-to-3′ transcription of coding gene exons or long non-coding RNAs that act as translation templates, RNA-binding protein regulators, or miRNA-binding sponges. Here, we present an exploratory study to evaluate the potential use of circRNA as ICB response predictors. We recruited 23 MM patients treated with Nivolumab and categorized them as good responders (10 patients) and bad responders (11 patients). All RNA-seq with ribosomal depletion was performed on RNA isolated from pre-treatment FFPE tumor biopsies. Two bioinformatic tools that use de novo assembly for the identification of circRNAs (Starchip and Ciri) were employed independently to characterize the MM circRNAs profile and to identify a specific pattern for responders. Overall, a total of 731 circRNAs were detected. Interestingly, the number of circRNAs were comparatively higher in bad responders (11.917 reads, 321 circRNAs) versus good responders (7368 reads, 176 circRNAs), with 234 common circRNAs. Additionally, we also explored the biological function of the identified circRNAs in MM, by interrogating KEGG, GO and Reactome databases to delineate the gene regulatory networks specifically associated with the response-related phenotypes. The circRNAs exclusively expressed in patients presenting resistance to Nivolumab were associated with Transcriptional Misregulation in Cancer and Viral Infections processes. In contrast, circRNAs expressed uniquely in good responders were enriched in Generic Transcription pathways. Importantly, amongst the identified circRNAs, CDR1-AS was differentially downregulated in good responders (536 vs 1580 reads, p-value 0.002). Remarkably, CDR1-AS acts as miRNA-binding sponge of miRNA-7 in the context of invasion and migration in melanoma. In conclusion, our exploratory study paves the way to the utilization of novel biomarkers associated to immunotherapy failure in MM patients. Moreover, our results suggest that CDR1-AS is a novel putative predictor of ICB response. Citation Format: Juan Luis Onieva, Javier Oliver, Aurora Laborda-Illanes, Maria Rosario Chica-Parrado, Alicia Garrido-Aranda, Cynthia Robles-Podadera, Daniel Prieto, Elena Gallego, Alfonso Sanchez, Inaki Comino, Vanessa De Luque, Martina Alvarez, Maria Jose Lozano, Pedro Jimenez, Miguel Angel Berciano-Guerrero, Emilio Alba, Manuel Cobo, Isabel Barragan. Evaluation of circular RNA profiling in metastatic melanoma patients treated with immune checkpoint blockade [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 281.