Clinical decision support algorithm based on machine learning to assess the clinical response to anti–programmed death-1 therapy in patients with non–small-cell lung cancer

Beung Chul Ahn, Jea Woo So, Chun Bong Synn, Tae Hyung Kim, Jae Hwan Kim, Yeongseon Byeon, Young Seob Kim, Seong Gu Heo, San Duk Yang, Mi Ran Yun, Sangbin Lim, Su Jin Choi, Wongeun Lee, Dong Kwon Kim, Eun Ji Lee, Seul Lee, Doo Jae Lee, Chang Gon Kim, Sun Min Lim, Min Hee HongByoung Chul Cho, Kyoung Ho Pyo, Hye Ryun Kim

Research output: Contribution to journalArticlepeer-review

Abstract

Objective: Anti–programmed death (PD)-1 therapy confers sustainable clinical benefits for patients with non–small-cell lung cancer (NSCLC), but only some patients respond to the treatment. Various clinical characteristics, including the PD-ligand 1 (PD-L1) level, are related to the anti–PD-1 response; however, none of these can independently serve as predictive biomarkers. Herein, we established a machine learning (ML)–based clinical decision support algorithm to predict the anti–PD-1 response by comprehensively combining the clinical information. Materials and methods: We collected clinical data, including patient characteristics, mutations and laboratory findings, from the electronic medical records of 142 patients with NSCLC treated with anti–PD-1 therapy; these were analysed for the clinical outcome as the discovery set. Nineteen clinically meaningful features were used in supervised ML algorithms, including LightGBM, XGBoost, multilayer neural network, ridge regression and linear discriminant analysis, to predict anti–PD-1 responses. Based on each ML algorithm's prediction performance, the optimal ML was selected and validated in an independent validation set of PD-1 inhibitor–treated patients. Results: Several factors, including PD-L1 expression, tumour burden and neutrophil-to-lymphocyte ratio, could independently predict the anti–PD-1 response in the discovery set. ML platforms based on the LightGBM algorithm using 19 clinical features showed more significant prediction performance (area under the curve [AUC] 0.788) than on individual clinical features and traditional multivariate logistic regression (AUC 0.759). Conclusion: Collectively, our LightGBM algorithm offers a clinical decision support model to predict the anti–PD-1 response in patients with NSCLC.

Original languageEnglish
Pages (from-to)179-189
Number of pages11
JournalEuropean Journal of Cancer
Volume153
DOIs
Publication statusPublished - 2021 Aug

Bibliographical note

Funding Information:
This research was supported by the Bio & Medical Technology Development Program of the National Research Foundation funded by the Ministry of Science and ICT [grant numbers NRF-2019M3A9B6065231 , 2017M3A9E8029717 , 2017M3A9E9072669 ]. This study was supported by a Dongin Sports research grant of Yonsei University College of Medicine (6-2019-0128).

Publisher Copyright:
© 2021

All Science Journal Classification (ASJC) codes

  • Oncology
  • Cancer Research

Fingerprint

Dive into the research topics of 'Clinical decision support algorithm based on machine learning to assess the clinical response to anti–programmed death-1 therapy in patients with non–small-cell lung cancer'. Together they form a unique fingerprint.

Cite this