TY - JOUR
T1 - Toward an Optimized Staging System for Pancreatic Ductal Adenocarcinoma
T2 - A Clinically Interpretable, Artificial Intelligence-Based Model
AU - Bertsimas, Dimitris
AU - Margonis, Georgios Antonios
AU - Huang, Yifei
AU - Andreatos, Nikolaos
AU - Wiberg, Holly
AU - Ma, Yu
AU - McIntyre, Caitlin
AU - Pulvirenti, Alessandra
AU - Wagner, Doris
AU - Van Dam, J. L.
AU - Gavazzi, Francesca
AU - Buettner, Stefan
AU - Imai, Katsunori
AU - Stasinos, Georgios
AU - He, Jin
AU - Kamphues, Carsten
AU - Beyer, Katharina
AU - Seeliger, Hendrik
AU - Weiss, Matthew J.
AU - Kreis, Martin
AU - Cameron, John L.
AU - Wei, Alice C.
AU - Kornprat, Peter
AU - Baba, Hideo
AU - Koerkamp, Bas Groot
AU - Zerbi, Alessandro
AU - D'Angelica, Michael
AU - Wolfgang, Christopher L.
N1 - Publisher Copyright:
© 2022 American Society of Clinical Oncology.
PY - 2021
Y1 - 2021
N2 - PURPOSE: The American Joint Committee on Cancer (AJCC) eighth edition schema for pancreatic ductal adenocarcinoma treats T and N stage as independent factors and uses positive lymph nodes (PLNs) to define N stage, despite data favoring lymph node ratio (LNR). We used artificial intelligence-based techniques to compare PLN with LNR and investigate interactions between tumor size and nodal status. METHODS: Patients who underwent pancreatic ductal adenocarcinoma resection between 2000 and 2017 at six institutions were identified. LNR and PLN were compared through shapley additive explanations (SHAP) analysis, with the best predictor used to define nodal status. We trained optimal classification trees (OCTs) to predict 1-year and 3-year risk of death, incorporating only tumor size and nodal status as variables. The OCTs were compared with the AJCC schema and similarly trained XGBoost models. Variable interactions were explored via SHAP. RESULTS: Two thousand eight hundred seventy-four patients comprised the derivation and 1,231 the validation cohort. SHAP identified LNR as a superior predictor. The OCTs outperformed the AJCC schema in the derivation and validation cohorts (1-year area under the curve: 0.681 v 0.603; 0.638 v 0.586, 3-year area under the curve: 0.682 v 0.639; 0.675 v 0.647, respectively) and performed comparably with the XGBoost models. We identified interactions between LNR and tumor size, suggesting that a negative prognostic factor partially overrides the effect of a concurrent favorable factor. CONCLUSION: Our findings highlight the superiority of LNR and the importance of interactions between tumor size and nodal status. These results and the potential of the OCT methodology to combine them into a powerful, visually interpretable model can help inform future staging systems.
AB - PURPOSE: The American Joint Committee on Cancer (AJCC) eighth edition schema for pancreatic ductal adenocarcinoma treats T and N stage as independent factors and uses positive lymph nodes (PLNs) to define N stage, despite data favoring lymph node ratio (LNR). We used artificial intelligence-based techniques to compare PLN with LNR and investigate interactions between tumor size and nodal status. METHODS: Patients who underwent pancreatic ductal adenocarcinoma resection between 2000 and 2017 at six institutions were identified. LNR and PLN were compared through shapley additive explanations (SHAP) analysis, with the best predictor used to define nodal status. We trained optimal classification trees (OCTs) to predict 1-year and 3-year risk of death, incorporating only tumor size and nodal status as variables. The OCTs were compared with the AJCC schema and similarly trained XGBoost models. Variable interactions were explored via SHAP. RESULTS: Two thousand eight hundred seventy-four patients comprised the derivation and 1,231 the validation cohort. SHAP identified LNR as a superior predictor. The OCTs outperformed the AJCC schema in the derivation and validation cohorts (1-year area under the curve: 0.681 v 0.603; 0.638 v 0.586, 3-year area under the curve: 0.682 v 0.639; 0.675 v 0.647, respectively) and performed comparably with the XGBoost models. We identified interactions between LNR and tumor size, suggesting that a negative prognostic factor partially overrides the effect of a concurrent favorable factor. CONCLUSION: Our findings highlight the superiority of LNR and the importance of interactions between tumor size and nodal status. These results and the potential of the OCT methodology to combine them into a powerful, visually interpretable model can help inform future staging systems.
UR - http://www.scopus.com/inward/record.url?scp=85122904450&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85122904450&partnerID=8YFLogxK
U2 - 10.1200/CCI.21.00001
DO - 10.1200/CCI.21.00001
M3 - Article
C2 - 34936469
AN - SCOPUS:85122904450
SN - 2473-4276
VL - 5
SP - 1220
EP - 1231
JO - JCO clinical cancer informatics
JF - JCO clinical cancer informatics
ER -