Japan Digestive Disease Week 2021
International Poster Session 4 (JDDW)
November 5, 10:10–10:50, Room 16 (Kobe International Exhibition Hall No.3 Building Digital Poster Venue)
IP-19_S
Prediction of early recurrence of hepatocellular carcinoma after resection: an international validation of the ERASL risk models
- 1
- Department of Gastroenterological Surgery, Okayama University
- 2
- Department of Surgery, Erasmus MC
- 3
- Department of Biomedical Data Science, Leiden University MC
Background: Prediction of early recurrence of hepatocellular carcinoma (HCC) after resection is an important issue for surgical strategy. Recently, the pre- and post-operative Early Recurrence After Surgery for Liver tumour (ERASL) models have been proposed. This study aimed to investigate the performance of the pre- and post-operative ERASL models at external validation.
Methods: Our international cohort study included 279 patients undergoing primary resection for HCC from the Netherlands and 392 patients from Japan. Performance was assessed according to discrimination (concordance (C) statistic) and calibration (correspondence between observed and predicted risk), with recalibration in a Weibull model.
Results: The discriminatory power of both models was lower in the Netherlands compared to Japan (C 0.57 [95%CI 0.52-0.62] vs. 0.69 [95%CI 0.65-0.73] for the ERASL-pre model and 0.62 [95%CI 0.57-0.67] vs. 0.70 [95% CI 0.66-0.74] for the ERASL-post model). The predictions of the ERASL models were systematically too optimistic for both cohorts. Recalibrated ERASL models improved local applicability for both cohorts.
Conclusions: The discrimination of ERASL models was limited in Western patients, in contrast to Japan with good performance. Recalibration models improved the accuracy of predictions. However, in general a model that explains the East-West difference or one that is tailored to Western patients still needs to be developed.
Methods: Our international cohort study included 279 patients undergoing primary resection for HCC from the Netherlands and 392 patients from Japan. Performance was assessed according to discrimination (concordance (C) statistic) and calibration (correspondence between observed and predicted risk), with recalibration in a Weibull model.
Results: The discriminatory power of both models was lower in the Netherlands compared to Japan (C 0.57 [95%CI 0.52-0.62] vs. 0.69 [95%CI 0.65-0.73] for the ERASL-pre model and 0.62 [95%CI 0.57-0.67] vs. 0.70 [95% CI 0.66-0.74] for the ERASL-post model). The predictions of the ERASL models were systematically too optimistic for both cohorts. Recalibrated ERASL models improved local applicability for both cohorts.
Conclusions: The discrimination of ERASL models was limited in Western patients, in contrast to Japan with good performance. Recalibration models improved the accuracy of predictions. However, in general a model that explains the East-West difference or one that is tailored to Western patients still needs to be developed.
- Index Term 1: hepatocellular carcinoma
- Index Term 2: external validation