Deep learning radiomic nomogram can predict the number of lymph node metastasis in locally advanced gastric cancer: an international multi-center study
Menée à partir de clichés de tomographies numériques réalisées sur 730 patients atteints d'un cancer de l'estomac de stade avancé, cette étude internationale (Chine et Italie) évalue la performance d'un nomogramme, basé sur des caractéristiques radiomiques des lésions cancéreuses et développé à l'aide d'un algorithme d'apprentissage automatique, pour prédire le nombre de métastases ganglionnaires
Background : Preoperative evaluation of the number of lymph node metastasis (LNM) is the basis of individual treatment of locally advanced gastric cancer (LAGC). However, the routinely used preoperative determination method is not accurate enough.
Patients and methods : We enrolled 730 LAGC patients from 5 centers in China and 1 center in Italy, and divided them into 1 primary cohort, 3 external validation cohorts, and 1 international validation cohort. A deep learning radiomic nomogram (DLRN) was built based on the images from multi-phase computed tomography (CT) for preoperatively determining the number of LNM in LAGC. We comprehensively tested the DLRN and compared it with three state-of-the-art methods. Moreover, we investigated the value of the DLRN in survival analysis.
Results : The DLRN showed good discrimination of the number of LNM on all cohorts (overall C-indexes: 0.821, 95% CI: 0.785-0.858 in the primary cohort; 0.797, 95% CI: 0.771-0.823 in the external validation cohorts; and 0.822, 95% CI: 0.756-0.887 in the international validation cohort). The nomogram performed significantly better than the routinely used clinical N stages, tumor size, and clinical model (p<0.05). Besides, DLRN is significantly associated with the overall survival of LAGC patients (n=271).
Conclusion : A deep learning-based radiomic nomogram had good predictive value for LNM in LAGC. In staging-oriented treatment of gastric cancer, this preoperative nomogram could provide baseline information for individual treatment of LAGC.
Annals of Oncology , article en libre accès, 2020