AUTHOR=Yang Dashuai , Zhu Mingqiang , Xiong Xiangyun , Su Yang , Zhao Fangrui , Hu Yong , Zhang Guo , Pei Junpeng , Ding Youming TITLE=Clinical features and prognostic factors in patients with microvascular infiltration of hepatocellular carcinoma: Development and validation of a nomogram and risk stratification based on the SEER database JOURNAL=Frontiers in Oncology VOLUME=Volume 12 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.987603 DOI=10.3389/fonc.2022.987603 ISSN=2234-943X ABSTRACT=Background: This study aimed to establish and validate a prognostic risk stratification and nomogram in hepatocellular carcinoma (HCC) patients with microvascular invasion (MVI) for predicting the cancer-specific survival (CSS). Methods: A total of 1487 HCC patients confirmed to have MVI between 2010 and 2015 were selected from the Surveillance, Epidemiology and End Results (SEER) database, were randomized into training cohort and validation cohort at a ratio of 7:3. The discrimination and calibration of the nomogram were evaluated using concordance index (C-index), area under curve (AUC), calibration plots. Decision curve analysis (DCA) was used to quantify the net benefit of the nomogram at different threshold probabilities and compare it to the American Joint Committee on Cancer (AJCC) tumor staging system. The clinical utility of the nomogram were compared with AJCC tumor staging system using Net reclassification index (NRI) and integrated discrimination improvement (IDI). The prognostic risk stratifications of the nomogram and the AJCC tumor staging system were compared. Results: Eight variables were selected to construct the nomogram for HCC patients with MVI. The C-index (training cohort: 0.785, 95%CI: 0.762-0.796; validation cohort: 0.776, 95%CI: 0.773-0.795) and the AUC values (> 0.7) showed satisfactory discrimination. The calibration plots showed good consistency between the actual observation and the nomogram prediction. The NRI values (training cohort: 1-, 3-, 5-year CSS: 0.75, 0.68, 0.67; validation cohort: 1-, 3-, 5-year CSS: 0.66, 0.71, 0.70) and IDI (training cohort: 1-, 3-, 5-year CSS: 0.17, 0.16, 0.15; validation cohort: 1-, 3-, 5-year CSS: 0.17, 0.17, 0.17) indicated that the established nomogram significantly outperformed AJCC staging system (P<0.05). The DCA curves showed that the nomogram better predicted 1-, 3-, and 5-year CSS in the training and validation cohorts because it added more net benefit than the AJCC staging system. Furthermore, the risk stratification system showed the CSS in different groups had a good regional division. Conclusions: The nomogram for predicting CSS for HCC patients with MVI was established and validated, which provided a new system of risk stratification as a practical tool for individualized treatment and management.