A prognostic nomogram to predict clinical outcomes of patients with primary malignant and benign retroperitoneal tumors.

Abstract

Background: Primary retroperitoneal tumor (PRT) is a relatively rare tumor with diverse histological and molecular types. We aim to develop and validate a concise prognostic nomogram for patients with benign and malignant PRT.

Methods: The clinical data of 206 PRT patients who underwent surgical management in authors' institution during January 2016 and December 2021 were analyzed. Logistic regression was used to select independent risk variables of binary outcome, while COX regression was performed for time-to-event. A predictive nomogram was developed based on multivariate analyses.

Results: Of the reported 206 PRTs, 113 patients were benign (54.85%) and 93 (45.15%) were malignant. Radical resection, extended radical resection and cytoreductive resection were performed in 141, 50 and 15 patients, respectively. The comprehensive complication index (CCI) was higher than 26.2% in 14 cases. Postoperative recurrence was experienced in 20 (9.7%) cases during the median 49 months. Nomogram was developed to predict severe complication and parameters included time of surgery, loss of blood and type of surgery with a moderate prediction capability [Area under curve (AUC) = 0.64)] after interval validation. While model for postoperative recurrence prediction includes parameters as chemotherapy, metastasis, combined resection and time of surgery with acceptable performance (AUC = 0.913). Overall survival is related to tumor size, metastasis and pathology as risk factors and the prediction capability was good (AUC = 0.800).

Conclusion: The proposed nomogram showed favorable predictive accuracy for prognosis in patients with PRTs. This has the potential to contribute to clinical decision-making.