International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning

Mariagiulia Dal Cero; Joan Gibert; Luis Grande; on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group; Marta Gimeno; Javier Osorio; Maria Bencivenga; Uberto Fumagalli Romario; Riccardo Rosati; Paolo Morgagni; Suzanne Gisbertz; Wojciech P. Polkowski; Lucio Lara Santos; Piotr Kołodziejczyk; Wojciech Kielan; Rossella Reddavid; Johanna W. van Sandick; Gian Luca Baiocchi; Ines Gockel; Andrew Davies; Bas P.L. Wijnhoven; Daniel Reim; Paulo Costa; William H. Allum; Guillaume Piessen; John V. Reynolds; Stefan P. Mönig; Paul M. Schneider; Elisenda Garsot; Emma Eizaguirre; Mònica Miró; Sandra Castro; Coro Miranda; Xavier Monzonis-Hernández; Manuel Pera

doi:10.3390/cancers16132463

International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning

Mariagiulia Dal Cero, Joan Gibert, Luis Grande, on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group, Marta Gimeno, Javier Osorio, Maria Bencivenga, Uberto Fumagalli Romario, Riccardo Rosati, Paolo Morgagni, Suzanne Gisbertz, Wojciech P. Polkowski, Lucio Lara Santos, Piotr Kołodziejczyk, Wojciech Kielan, Rossella Reddavid, Johanna W. van Sandick, Gian Luca Baiocchi, Ines Gockel, Andrew DaviesBas P.L. Wijnhoven, Daniel Reim, Paulo Costa, William H. Allum, Guillaume Piessen, John V. Reynolds, Stefan P. Mönig, Paul M. Schneider, Elisenda Garsot, Emma Eizaguirre, Mònica Miró, Sandra Castro, Coro Miranda, Xavier Monzonis-Hernández, Manuel Pera^*

^*Corresponding author for this work

Surgery

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

Background: Radical gastrectomy remains the main treatment for gastric cancer, despite its high mortality. A clinical predictive model of 90-day mortality (90DM) risk after gastric cancer surgery based on the Spanish EURECCA registry database was developed using a matching learning algorithm. We performed an external validation of this model based on data from an international multicenter cohort of patients. Methods: A cohort of patients from the European GASTRODATA database was selected. Demographic, clinical, and treatment variables in the original and validation cohorts were compared. The performance of the model was evaluated using the area under the curve (AUC) for a random forest model. Results: The validation cohort included 2546 patients from 24 European hospitals. The advanced clinical T- and N-category, neoadjuvant therapy, open procedures, total gastrectomy rates, and mean volume of the centers were significantly higher in the validation cohort. The 90DM rate was also higher in the validation cohort (5.6%) vs. the original cohort (3.7%). The AUC in the validation model was 0.716. Conclusion: The externally validated model for predicting the 90DM risk in gastric cancer patients undergoing gastrectomy with curative intent continues to be as useful as the original model in clinical practice.

Original language	English
Article number	2463
Journal	Cancers
Volume	16
Issue number	13
DOIs	https://doi.org/10.3390/cancers16132463
Publication status	Published - 5 Jul 2024

Bibliographical note

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.3390/cancers16132463Licence: CC BY

cancers-16-02463-v2Final published version, 987 KBLicence: CC BY

Cite this

Dal Cero, M., Gibert, J., Grande, L., on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group, Gimeno, M., Osorio, J., Bencivenga, M., Fumagalli Romario, U., Rosati, R., Morgagni, P., Gisbertz, S., Polkowski, W. P., Lara Santos, L., Kołodziejczyk, P., Kielan, W., Reddavid, R., van Sandick, J. W., Baiocchi, G. L., Gockel, I., ... Pera, M. (2024). International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning. Cancers, 16(13), Article 2463. https://doi.org/10.3390/cancers16132463

@article{c5279800bea342faa3db99a717839f08,

title = "International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning",

abstract = "Background: Radical gastrectomy remains the main treatment for gastric cancer, despite its high mortality. A clinical predictive model of 90-day mortality (90DM) risk after gastric cancer surgery based on the Spanish EURECCA registry database was developed using a matching learning algorithm. We performed an external validation of this model based on data from an international multicenter cohort of patients. Methods: A cohort of patients from the European GASTRODATA database was selected. Demographic, clinical, and treatment variables in the original and validation cohorts were compared. The performance of the model was evaluated using the area under the curve (AUC) for a random forest model. Results: The validation cohort included 2546 patients from 24 European hospitals. The advanced clinical T- and N-category, neoadjuvant therapy, open procedures, total gastrectomy rates, and mean volume of the centers were significantly higher in the validation cohort. The 90DM rate was also higher in the validation cohort (5.6%) vs. the original cohort (3.7%). The AUC in the validation model was 0.716. Conclusion: The externally validated model for predicting the 90DM risk in gastric cancer patients undergoing gastrectomy with curative intent continues to be as useful as the original model in clinical practice.",

author = "{Dal Cero}, Mariagiulia and Joan Gibert and Luis Grande and {on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group} and Marta Gimeno and Javier Osorio and Maria Bencivenga and {Fumagalli Romario}, Uberto and Riccardo Rosati and Paolo Morgagni and Suzanne Gisbertz and Polkowski, {Wojciech P.} and {Lara Santos}, Lucio and Piotr Ko{\l}odziejczyk and Wojciech Kielan and Rossella Reddavid and {van Sandick}, {Johanna W.} and Baiocchi, {Gian Luca} and Ines Gockel and Andrew Davies and Wijnhoven, {Bas P.L.} and Daniel Reim and Paulo Costa and Allum, {William H.} and Guillaume Piessen and Reynolds, {John V.} and M{\"o}nig, {Stefan P.} and Schneider, {Paul M.} and Elisenda Garsot and Emma Eizaguirre and M{\`o}nica Mir{\'o} and Sandra Castro and Coro Miranda and Xavier Monzonis-Hern{\'a}ndez and Manuel Pera",

note = "Publisher Copyright: {\textcopyright} 2024 by the authors.",

year = "2024",

month = jul,

day = "5",

doi = "10.3390/cancers16132463",

language = "English",

volume = "16",

journal = "Cancers",

issn = "2072-6694",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "13",

}

Dal Cero, M, Gibert, J, Grande, L, on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group, Gimeno, M, Osorio, J, Bencivenga, M, Fumagalli Romario, U, Rosati, R, Morgagni, P, Gisbertz, S, Polkowski, WP, Lara Santos, L, Kołodziejczyk, P, Kielan, W, Reddavid, R, van Sandick, JW, Baiocchi, GL, Gockel, I, Davies, A, Wijnhoven, BPL, Reim, D, Costa, P, Allum, WH, Piessen, G, Reynolds, JV, Mönig, SP, Schneider, PM, Garsot, E, Eizaguirre, E, Miró, M, Castro, S, Miranda, C, Monzonis-Hernández, X & Pera, M 2024, 'International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning', Cancers, vol. 16, no. 13, 2463. https://doi.org/10.3390/cancers16132463

TY - JOUR

T1 - International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning

AU - Dal Cero, Mariagiulia

AU - Gibert, Joan

AU - Grande, Luis

AU - on behalf of the Spanish EURECCA Esophagogastric Cancer Group and the European GASTRODATA Study Group

AU - Gimeno, Marta

AU - Osorio, Javier

AU - Bencivenga, Maria

AU - Fumagalli Romario, Uberto

AU - Rosati, Riccardo

AU - Morgagni, Paolo

AU - Gisbertz, Suzanne

AU - Polkowski, Wojciech P.

AU - Lara Santos, Lucio

AU - Kołodziejczyk, Piotr

AU - Kielan, Wojciech

AU - Reddavid, Rossella

AU - van Sandick, Johanna W.

AU - Baiocchi, Gian Luca

AU - Gockel, Ines

AU - Davies, Andrew

AU - Wijnhoven, Bas P.L.

AU - Reim, Daniel

AU - Costa, Paulo

AU - Allum, William H.

AU - Piessen, Guillaume

AU - Reynolds, John V.

AU - Mönig, Stefan P.

AU - Schneider, Paul M.

AU - Garsot, Elisenda

AU - Eizaguirre, Emma

AU - Miró, Mònica

AU - Castro, Sandra

AU - Miranda, Coro

AU - Monzonis-Hernández, Xavier

AU - Pera, Manuel

PY - 2024/7/5

Y1 - 2024/7/5

N2 - Background: Radical gastrectomy remains the main treatment for gastric cancer, despite its high mortality. A clinical predictive model of 90-day mortality (90DM) risk after gastric cancer surgery based on the Spanish EURECCA registry database was developed using a matching learning algorithm. We performed an external validation of this model based on data from an international multicenter cohort of patients. Methods: A cohort of patients from the European GASTRODATA database was selected. Demographic, clinical, and treatment variables in the original and validation cohorts were compared. The performance of the model was evaluated using the area under the curve (AUC) for a random forest model. Results: The validation cohort included 2546 patients from 24 European hospitals. The advanced clinical T- and N-category, neoadjuvant therapy, open procedures, total gastrectomy rates, and mean volume of the centers were significantly higher in the validation cohort. The 90DM rate was also higher in the validation cohort (5.6%) vs. the original cohort (3.7%). The AUC in the validation model was 0.716. Conclusion: The externally validated model for predicting the 90DM risk in gastric cancer patients undergoing gastrectomy with curative intent continues to be as useful as the original model in clinical practice.

AB - Background: Radical gastrectomy remains the main treatment for gastric cancer, despite its high mortality. A clinical predictive model of 90-day mortality (90DM) risk after gastric cancer surgery based on the Spanish EURECCA registry database was developed using a matching learning algorithm. We performed an external validation of this model based on data from an international multicenter cohort of patients. Methods: A cohort of patients from the European GASTRODATA database was selected. Demographic, clinical, and treatment variables in the original and validation cohorts were compared. The performance of the model was evaluated using the area under the curve (AUC) for a random forest model. Results: The validation cohort included 2546 patients from 24 European hospitals. The advanced clinical T- and N-category, neoadjuvant therapy, open procedures, total gastrectomy rates, and mean volume of the centers were significantly higher in the validation cohort. The 90DM rate was also higher in the validation cohort (5.6%) vs. the original cohort (3.7%). The AUC in the validation model was 0.716. Conclusion: The externally validated model for predicting the 90DM risk in gastric cancer patients undergoing gastrectomy with curative intent continues to be as useful as the original model in clinical practice.

UR - http://www.scopus.com/inward/record.url?scp=85198515513&partnerID=8YFLogxK

U2 - 10.3390/cancers16132463

DO - 10.3390/cancers16132463

M3 - Article

C2 - 39001525

AN - SCOPUS:85198515513

SN - 2072-6694

VL - 16

JO - Cancers

JF - Cancers

IS - 13

M1 - 2463

ER -

International External Validation of Risk Prediction Model of 90-Day Mortality after Gastrectomy for Cancer Using Machine Learning

Abstract

Bibliographical note

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this