To explore the risk factors of lymphovascular invasion in patients with upper tract urothelial carcinoma and construct a prediction model.

Background and objective: To explore the risk factors and construct a prediction model of the lymphovascular invasion (LVI) in patients with upper tract urothelial carcinoma (UTUC).

Methods: Clinical data of 143 UTUC patients treated in our hospital during Jan. 2010 and Dec. 2022 were retrospectively analyzed. The patients were divided into LVI positive group and LVI negative group according to the postoperative lymphovascular conditions. Kaplan-Meier method was used to evaluate the overall survival (OS) and cancer-specific survival (CSS) of the two groups, and the survival curve was drawn. The correlation between LVI and inclusion indexes was analyzed using univariate and ultivariate. A prediction model was established and receiver operating characteristic (ROC) curve was drawn to analyze the diagnostic value.

Results: The median survival time of LVI positive patients was 78 months (95%CI 44.47-111.53), lower than the 90months (95%CI 72.77-107.23) for LVI negative patients, and the 5-year OS of LVI positive patients was 53.0%, lower than that of LVI negative patients (79.6%). The difference was statistically significant (P=0.005). The 5-year CSS of LVI positive patients was 57.0%, lower than that of LVI negative patients (85.7%, P=0.009). The results of univariate analysis showed that there were statistically significant differences between the two groups (P < 0.05) in exfoliation cytology (P=0.044), hydronephrosis (P=0.015), preoperative fibrinogen level (P=0.003), lymph node status (P=0.014), pathological stage (P=0.001) and grade (P=0.047). Multivariate Logistic regression analysis showed that hydronephrosis (P=0.022), pathological stage (P < 0.001), lymph node status (P=0.025) and fibrinogen level (P=0.019) were independent factors influencing the occurrence of lymphovascular invasion, and the combination of four indexes above was better than any single index. the ROC curve showed that the area under the curve (AUC) of postoperative LVI was the largest when combined with the four predictors, and the AUC was 0.833 (95%CI 0.759-0.907). When the Youden index was 0.594, the sensitivity was 81.1%, and the specificity was 78.3%.

Conclusion: Lymphovascular invasion is related to hydronephrosis, pathological stage, lymph node condition and fibrinogen level. Patients with preoperative hydronephrosis, high pathological stage, lymph node metastasis and high fibrinogen level were at higher risk of lymphovascular invasion.