AUTHOR=Ke Hu , Jiang Shengming , He Ziqi , Song Qianlin , Yang Dashuai , Song Chao , Dong Caitao , Liu Junwei , Su Xiaozhe , Zhou Jiawei , Xiong Yunhe 

TITLE=Clinical features and prognostic factors in patients diagnosed with lymphovascular invasion of testicular germ-cell tumors: Analysis based on the SEER database

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.1142441

DOI=10.3389/fonc.2023.1142441

ISSN=2234-943X

ABSTRACT=Background: Lympho-vascular invasion (LVI) is a high-risk factor for testicular germ cell tumors (TGCT), while TGCT-LVI patients lacked prognostic model. Our study aimed to develop a nomogram to predict the overall survival (OS) of TGCT-LVI patients.
Methods: A complete cohort of 3288 eligible TGCG-LVI patients (training cohort: 2300 cases, validation cohort: 988 cases) were obtained from the Surveillance, Epidemiology and End Results (SEER) database. Variables screened by multivariate Cox regression analysis were used to construct nomogram, which would be subsequently evaluated by consistency index (C-index), time-dependent ROC and calibration plots. The advantages and disadvantages of the AJCC staging system and the nomogram were assessed by IDI and NRI, as well as the decision analysis curves (DCA) was used to measure the net clinical benefit of nomogram versus the AJCC staging system. Finally, the Kaplan-Meier curves were used to value the ability to identify different risk groups between the traditional AJCC staging system and the new risk stratification system built on nomogram.
Results: Nine variables were screened by multivariate Cox regression analysis to construct nomogram. The C-index (training cohort: 0.821, validation cohort: 0.819) and time-dependent ROC of 3-, 5-, and 9-year between two cohorts suggested that the nomogram had good discriminatory ability. Calibration curves showed good consistency of nomogram. The NRI values of 3-, 5- and 9-year OS were respectively 0.308, 0.274 and 0.295, while the corresponding parts of validation cohort were 0.093, 0.093 and 0.099（P<0.01）. Additionally, the nomogram had more net clinical benefit showed by DCA curves and new risk stratification system provided better differentiation than AJCC staging system.
Conclusions: A prognostic nomogram and new risk stratification system were developed and validated to assist clinicians to assess the TGCT-LVI patients.