Comprehensive genomic and transcriptomic profiling analysis of the recurrent gastrointestinal stromal tumor (GIST).

person Lu Sun Department of Pathology, The First Medicine Center of PLA General Hospital, Beijing, China info_outline Lu Sun, Xiaohui Wang, Xin Wu, Fengwei Zhu, Liu Hong, Wen Chen, Yingshi Piao, Hong Luo, Yiqun Zhang, Ling Liu, Yuehua Zang, Wenjing Li, Zhihua Pei, Dongliang Wang, Qiming Zhou, Huaiyin Shi

Authors person Lu Sun Department of Pathology, The First Medicine Center of PLA General Hospital, Beijing, China info_outline Lu Sun, Xiaohui Wang, Xin Wu, Fengwei Zhu, Liu Hong, Wen Chen, Yingshi Piao, Hong Luo, Yiqun Zhang, Ling Liu, Yuehua Zang, Wenjing Li, Zhihua Pei, Dongliang Wang, Qiming Zhou, Huaiyin Shi Organizations Department of Pathology, The First Medicine Center of PLA General Hospital, Beijing, China, ChosenMed Technology (Zhejiang) Co., Ltd., Beijing, China, Chinese People’s Liberation Army General Hospital, Beijing, China, Department of Pathology, The Sixth Medicine Center of PLA General Hospital, Beijing, China, Department of Pathology, The Eighth Medicine Center of PLA General Hospital, Beijing, China, Department of Pathology, Beijing TongRen Hospital, Capital Medical University, Beijing, China, Beijing, China, Department of Pathology, Siping Central People’s Hospital, Jilin, China, Beijing, China, ChosenMed Technology (Zhejiang) Co., Ltd, Beijing, China Abstract Disclosures Research Funding No funding received None. Background: The gastrointestinal stromal tumor (GIST) is a potentially malignant mesenchymal tumor (sarcoma), whose management has been transformed by approvals of TKIs. However, recurrence or metastasis is a formidable challenge in managing high-risk GIST. Comprehensive investigation of the transcriptome and genome in GIST is helpful to determine the therapy strategy for patients. This study aims to explore the molecular features of high-risk recurrence GIST. Methods: The tumor tissue samples of 42 patients with primary GIST were collected, in which all patients were divided into low-risk (n = 12), intermediate-risk (n = 11), and high-risk (n = 19) groups according to the NIH 2017 th , in which 8 developed recurrence within three years and 11 without recurrence over three years in the high-risk group. Tumor-normal whole-exome and tumor mRNA sequencing was performed on baseline tissue DNA isolated from all patients. Results: Consistent with prior knowledge, KIT was the predominant mutation. The top mutated genes in GISTs were KIT (82%), PHLDA1 (39%), MUC6 (22%), ARID1B(20%), SURF2(20%), MGA(16%), MUC17(16%), MPRIP(16%) and PDGFRA(16%), which MUC6 was a significantly different mutation gene (DMG) associated with disease progression in terms of either high-risk(p = 0.04), recurrence(p = 0.02) or progression(p = 0.05). Interestingly, PDGFRA mutation co-occurs with CTNND2(p < 0.01) and its correlated genes. Additionally, we further explored the somatic copy number variations (CNVs). amp_RAD51B(49%), amp_MLH3(41%), amp_CHEK2(37%), and amp_RB1(12%) seemed to be more frequent in the recurrence group, while these observations were not significant (p > 0.05). Notably, del_NCOR1 or amp_NCOR1 was uniquely observed in the recurrence group. Compared with non-recurrence GIST, gene expression profiles revealed that mast cell activate (MCA)，M1-macrophages infiltration were higher in recurrence group, and mRNA was significantly upregulated in Systemic lupus erythematosus、Neutrophil extracellular trap formation、Viral carcinogenesis、Endocytosis. These pathways were indentified as being associated with recurrence and progression in other cancer studies. For example, the endocytosis pathway expression is positively associated with oncogenic pathway signaling in CRC, and neutrophil extracellular trap on cancer cells promotes liver metastasis in the tumor microenvironment. In the other, NCOR1 is a crucial regulator of T cell and B cell development. Conclusions: Here, we demonstrated the DNA/RNA profiles of GISTs. We observed higher CNVs in high-risk GIST. Especially, the high-risk recurred tumor presented different CNVs frequency and microenvironment. These results may contribute to understanding the mechanism of recurrence and improving recurrence detection.

Pre-clinical