Single Cell Transcriptomic Analysis Reveals Signature Cell Types and Gene Sets As Severity Prediction Indicators in Acute Pancreatitis

Severe acute pancreatitis (SAP) is a life-threatening inflammatory disorder with a high mortality rate. Early SAP diagnosis is challenging as the mechanisms underlying its progression remains unclear. Here, we performed single-cell RNA-seq sequencing on peripheral blood mononuclear cells (PBMC) drawn on days 1, 3, and 7 from patients with mild and severe AP and from healthy controls. We found that the population expansion of plasma cells is significantly associated with the prognosis of SAP. We used a machine learning-based algorithm to design and build an early diagnosis model for SAP. It could forecast the onset of SAP as early as day 1 after diagnosis, required only ten genes as its input, and was based on the B cell expression profile. Furthermore, the 10-gene signature was further validated using machine learning model that based on qRT-PCR results from 114 PBMC samples of the patients with AP, with an area under the receiver operating characteristic curve (AUROC) score of 0.92.Funding Information: This research was financially supported by the Natural Science Foundation of China (numbers 81670581, 82100678, 81470941, 81270801 and 82100678), Program for Outstanding Medical Academic Leader and Shanghai Municipal Science and Technology Commission (number 18411966400).Declaration of Interests: The authors declare that they have no conflict of interest.Ethics Approval Statement: This study was approved by the Ruijin Hospital Ethics Committee affiliated with the Shanghai Jiao Tong University School of Medicine. All participants enrolled in this trial provided informed written consent (2019-90).