Integration of Medical and Genomic Information to Enhance
Relapse Prediction in Early Stage Lung Cancer Patients

D 2022

Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients

TIMILSINA, Mohan, Dirk FEY, Adrianna JANIK, Maria TORRENTE, Mariano PROVENCIO et. al.

Basic information

Original name

Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients

Authors

Edition

Washington, USA, Proceedings of the Annual Symposium of the American Medical Informatics Association, p. 1082-1091, 10 pp. 2022

Publisher

AMIA

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

10201 Computer sciences, information science, bioinformatics

Country of publisher

United States of America

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

electronic version available online

References:

URL

Organization unit

Faculty of Informatics

ISSN

Keywords in English

relapse; lung cancer; imputation; machine learning; genomic scores

Abstract

V originále

Early-stage lung cancer is crucial clinically due to its insidious nature and rapid progression. Most of the prediction models designed to predict tumour recurrence in the early stage of lung cancer rely on the clinical or medical history of the patient. However, their performance could likely be improved if the input patient data contained genomic information. Unfortunately, such data is not always collected. This is the main motivation of our work, in which we have imputed and integrated specific type of genomic data with clinical data to increase the accuracy of machine learning models for prediction of relapse in early-stage, non-small cell lung cancer patients. Using a publicly available TCGA lung adenocarcinoma cohort of 501 patients, their aneuploidy scores were imputed into similar records in the Spanish Lung Cancer Group (SLCG) data, more specifically a cohort of 1348 early-stage patients. First, the tumor recurrence in those patients was predicted without the imputed aneuploidy scores. Then, the SLCG data were enriched with the aneuploidy scores imputed from TCGA. This integrative approach improved the prediction of the relapse risk, achieving area under the precision-recall curve (PR-AUC) score of 0.74, and area under the ROC (ROC-AUC) score of 0.79. Using the prediction explanation model SHAP (SHapley Additive exPlanations), we further explained the predictions performed by the machine learning model. We conclude that our explainable predictive model is a promising tool for oncologists that addresses an unmet clinical need of post-treatment patient stratification based on the relapse risk, while also improving the predictive power by incorporating proxy genomic data not available for the actual specific patients.

Citovat

TIMILSINA, Mohan, Dirk FEY, Adrianna JANIK, Maria TORRENTE, Mariano PROVENCIO, Alberto Cruz BERMUDEZ, Enric CARCERENY, Luca COSTABELLO, Delvys Rodrıguez ABREU, Manuel COBO, Rafael Lopez CASTRO, Reyes BERNABE, Maria GUIRADO, Pasquale MINERVINI and Vít NOVÁČEK. Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients. Online. In Proceedings of the Annual Symposium of the American Medical Informatics Association. Washington, USA: AMIA, 2022, p. 1082-1091. ISSN 1559-4076.

@inproceedings{2241663,
   author = {Timilsina, Mohan and Fey, Dirk and Janik, Adrianna and Torrente, Maria and Provencio, Mariano and Bermudez, Alberto Cruz and Carcereny, Enric and Costabello, Luca and Abreu, Delvys Rodrıguez and Cobo, Manuel and Castro, Rafael Lopez and Bernabe, Reyes and Guirado, Maria and Minervini, Pasquale and Nováček, Vít},
   address = {Washington, USA},
   booktitle = {Proceedings of the Annual Symposium of the American Medical Informatics Association},
   keywords = {relapse; lung cancer; imputation; machine learning; genomic scores},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {Washington, USA},
   pages = {1082-1091},
   publisher = {AMIA},
   title = {Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients},
   url = {https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10148374/},
   year = {2022}
}

TY  - JOUR
ID  - 2241663
AU  - Timilsina, Mohan - Fey, Dirk - Janik, Adrianna - Torrente, Maria - Provencio, Mariano - Bermudez, Alberto Cruz - Carcereny, Enric - Costabello, Luca - Abreu, Delvys Rodrıguez - Cobo, Manuel - Castro, Rafael Lopez - Bernabe, Reyes - Guirado, Maria - Minervini, Pasquale - Nováček, Vít
PY  - 2022
TI  - Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients
PB  - AMIA
CY  - Washington, USA
KW  - relapse
KW  - lung cancer
KW  - imputation
KW  - machine learning
KW  - genomic scores
UR  - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10148374/
N2  - Early-stage lung cancer is crucial clinically due to its insidious nature and rapid progression. Most of the prediction models designed to predict tumour recurrence in the early stage of lung cancer rely on the clinical or medical history of the patient. However, their performance could likely be improved if the input patient data contained genomic information. Unfortunately, such data is not always collected. This is the main motivation of our work, in which we have imputed and integrated specific type of genomic data with clinical data to increase the accuracy of machine learning models for prediction of relapse in early-stage, non-small cell lung cancer patients. Using a publicly available TCGA lung adenocarcinoma cohort of 501 patients, their aneuploidy scores were imputed into similar records in the Spanish Lung Cancer Group (SLCG) data, more specifically a cohort of 1348 early-stage patients. First, the tumor recurrence in those patients was predicted without the imputed aneuploidy scores. Then, the SLCG data were enriched with the aneuploidy scores imputed from TCGA. This integrative approach improved the prediction of the relapse risk, achieving area under the precision-recall curve (PR-AUC) score of 0.74, and area under the ROC (ROC-AUC) score of 0.79. Using the prediction explanation model SHAP (SHapley Additive exPlanations), we further explained the predictions performed by the machine learning model. We conclude that our explainable predictive model is a promising tool for oncologists that addresses an unmet clinical need of post-treatment patient stratification based on the relapse risk, while also improving the predictive power by incorporating proxy genomic data not available for the actual specific patients.
ER  -

TIMILSINA, Mohan, Dirk FEY, Adrianna JANIK, Maria TORRENTE, Mariano PROVENCIO, Alberto Cruz BERMUDEZ, Enric CARCERENY, Luca COSTABELLO, Delvys Rodrıguez ABREU, Manuel COBO, Rafael Lopez CASTRO, Reyes BERNABE, Maria GUIRADO, Pasquale MINERVINI and Vít NOVÁČEK. Integration of Medical and Genomic Information to Enhance Relapse Prediction in Early Stage Lung Cancer Patients. Online. In \textit{Proceedings of the Annual Symposium of the American Medical Informatics Association}. Washington, USA: AMIA, 2022, p.~1082-1091. ISSN~1559-4076.

Detailed Information on Publication Record