Objective: We aimed to investigate the differences of transcriptome profile between 2 groups of high-grade serous ovarian cancer (HGSOC) patients with distinct outcomes and identify potential biomarkers for recurrence. Methods: RNA sequencing was performed in 2 groups of HGSOC patients with similar demographic characteristics but exhibiting distinct progression-free survival (PFS). Transcriptome data of poor response (PR; PFS ≤6 months) and good response (GR; PFS ≥12 months) group were compared. We employed xCell to evaluate the abundance of 63 cells in tumor microenvironment. The predictive value of recurrence-related tumor infiltration cells was validated in cohort data from Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) dataset. The weighted correlation network analysis was performed to identify the genes related to cell infiltration. Results: PR patients exhibited a distinct tumor infiltration immune cells-related transcriptional profile compared to GR patients, such as lower signatures of leukocyte differentiation, activation and chemotaxis. The fraction of T-helper 2 (Th2) cells infiltration was significantly higher in PR group than in GR group. High infiltration of Th2 was significantly associated with unfavorable prognosis in the GEO cohort (area under the curve=0.84 at 6 months recurrence) and TCGA cohort (p=0.008). Genes enriched to extracellular matrix organization and integrin binding were relevant to Th2 infiltration. Conclusion Patients with HGSOC having shorter PFS exhibited a distinct gene signature that related to tumor-infiltrating immune cells. The level of Th2 infiltration could facilitate patient recurrence risk stratification and may be a promising biomarker for prognosis prediction and immune-related treatment.

Sinapic acid (SA) is a phenolic acid that is widely distributed in fruits and vegetables, which has various bioactivities, such as antidiabetic, anticancer and anti-inflammatory functions. Over-activated microglial is involved in the development progress of neurodegenerative diseases, such as Parkinson’s disease and Alzheimer’s disease. The objective of this study was to investigate the effect of SA in microglia neuroinflammation models. Our results demonstrated that SA inhibited secretion of the nitric oxide (NO) and interleukin (IL)-6, reduced the expression of inducible nitric oxide synthase (iNOS) and enhanced the release of IL-10 in a dose-dependent manner. Besides, our further investigation revealed that SA attenuated the phosphorylation of AKT and MAPK cascades in LPS-induced microglia. Consistently, oral administration of SA in mouse regulated the production of inflammationrelated cytokines and also suppressed the phosphorylation of MAPK cascades and AKT in the mouse cerebral cortex. These results suggested that SA may be a possible therapy candidate for anti-inflammatory activity by targeting the AKT/MAPK signaling pathway.