Key points are not available for this paper at this time.
Background Due to the lack of a comprehensive evaluation of the short-term prognosis of patients undergoing radical resection of esophageal cancer after neoadjuvant therapy, recent clinical strategies have remained subjective and controversial. The recognition of pretreatment risk factors and tailored treatment could improve outcomes of esophagectomy patients. Therefore, we aimed to develop a predictive model that differentiates high-risk conditions after surgery in patients with esophageal cancer receiving neoadjuvant therapy. Methods We conducted a muti-center, retrospective cohort study in Fujian Medical University Union Hospital, Ganzhou Cancer Hospital and Affiliated Hospital of Jinggangshan University. A principal component (PC) analysis was applied for data simplification and the extraction of patient short-term outcome characteristic. We identified risk status on admission and operation, via a logistic regression and then constructed prediction models for worsened short-term outcomes. Results Of 334 patients of train set underwent neoadjuvant therapy and esophagectomy, 142(42.5%) received neoadjuvant chemotherapy (NC), 192(57.5%) received neoadjuvant immunochemotherapy (NIC). After grouping by principal component analysis, patients were divided into high-risk group (83,25%) and low-risk group (251,75%). Twelve features regarding clinical feature, nutrition indicators, laboratory indicators and intraoperative data were identified. The prediction model showed the best performance in predicting high short-term outcome risk, with an area under the receiver operating characteristic curve (AUC) of 0.794 (95%CI, 0.712,0.876) in NC, and 0.781 (95%CI, 0.705,0.858) in NIC. Conclusion A novel short-term outcome prediction model, offers a comprehensive assessment of posttreatment recovery in esophagectomy patients after neoadjuvant immunotherapy combined with chemotherapy and neoadjuvant chemotherapy.
Zhong et al. (Fri,) studied this question.