LAUSR

We much appreciated for the comment on our manuscript [1]. However, there are some misunderstanding in this comment. Firstly, the authors concluded that circPVT1 played a cancer suppressor role in vivo (from the clinical correlation), whilewas proved to act an oncogene in vitro (from the functional assays). However, the clinical correlation (biomarker) may have no functional implication. We were also very curious about the contradiction between clinical implication and cellular function of circPVT1 in GC. Indeed, we observed similar results in liver cancer (unpublished data). Secondly, the authors claimed that only one study reported the effect of circPVT1 in the past [2]. But this paper published in 2006 did not mention anything about circPVT1 or circRNA. In fact, a most recent paper published in Nucleic Acids Research and reported that circPVT1, elevated in dividing cells and reduced in senescent cells, sequesters let-7 to enable a proliferative phenotype [3]. This function is consistent with our result in GC cells. Thirdly, although we first characterized the circular RNA profile of GC from three paried GC tissues, the expression of circPVT1 was further validated from about 200 patients (Fig. 3A). We used ROC curves to compare