Related Experiment Video
Updated: Feb 10, 2026

Assessment of Resistance to Tyrosine Kinase Inhibitors by an Interrogation of Signal Transduction Pathways by Antibody Arrays
Published on: September 19, 2018
Machine learning identifies a core gene set predictive of acquired resistance to EGFR tyrosine kinase inhibitor
Young Rae Kim1, Sung Young Kim2
1Department of Biochemistry, Konkuk University School of Medicine, Seoul, 143-701, Republic of Korea.
Purpose:
Acquired resistance (AR) to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) is a major issue worldwide, for both patients and healthcare providers. However, precise prediction is currently infeasible due to the lack of an appropriate model. This study was conducted to develop and validate an individualized prediction model for automated detection of acquired EGFR-TKI resistance.
Methods:
Penalized regression was applied to construct a predictive model using publically available genomic cohorts of acquired EGFR-TKI resistance. To develop a model with enhanced generalizability, we merged multiple cohorts then updated the learning parameter via robust cross-study validation. Model performance was evaluated mainly using the area under the receiver operating characteristic curve.
Results:
Using a multi-study-derived machine learning method, we developed an extremely parsimonious model with generalized predictors (DDK3, CPS1, MOB3B, KRT6A), which has excellent prediction performance on blind cohorts for AR to EGFR-TKIs (gefitinib, erlotinib and afatinib) and monoclonal antibody against EGFR (cetuximab). In addition, our model also showed high performance for predicting intrinsic resistance (IR) to EGFR-TKIs from two large-scale pharmacogenomic resources, the Cancer Genome Project and the Cancer Cell Line Encyclopedia, suggesting that these general predictive features may work across AR and IR.
Conclusions:
We successfully constructed a multi-study-derived prediction model for acquired EGFR-TKI resistance with excellent accuracy, generalizability and transferability.
Insights
A new model accurately predicts acquired resistance to EGFR-TKIs, a common challenge in cancer treatment. This machine learning approach identifies key predictors for better patient outcomes.
Area of Science:
- Oncology
- Genomics
- Machine Learning
Background:
- Acquired resistance (AR) to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) poses a significant global challenge in cancer care.
- Current prediction of AR is limited due to the absence of effective models.
Purpose of the Study:
- To develop and validate an individualized prediction model for automated detection of acquired EGFR-TKI resistance.
- To address the unmet need for precise prediction of treatment resistance in cancer patients.
Main Methods:
- Penalized regression was employed to build a predictive model using public genomic data.
- Multiple cohorts were merged and validated to enhance model generalizability.
- Model performance was assessed using the area under the receiver operating characteristic curve.
Main Results:
- A parsimonious model with predictors DDK3, CPS1, MOB3B, and KRT6A was developed using a multi-study machine learning approach.
- The model demonstrated excellent prediction performance for AR to EGFR-TKIs (gefitinib, erlotinib, afatinib) and cetuximab across blind cohorts.
- High performance was also observed in predicting intrinsic resistance (IR) to EGFR-TKIs, indicating broad applicability.
Conclusions:
- A novel multi-study-derived prediction model for acquired EGFR-TKI resistance was successfully developed.
- The model exhibits excellent accuracy, generalizability, and transferability for predicting treatment resistance.
- This tool has the potential to improve individualized cancer treatment strategies.
Related Concept Videos
Receptor Tyrosine Kinases
Treatment for Pulmonary Arterial Hypertension: Receptor Tyrosine Kinase Inhibitors and Calcium Channel Blockers
TKIs, such as imatinib (Gleevec), are particularly effective in tackling the growth and mitogenic factors that become upregulated in PAH patients. These factors contribute to the...
Protein Kinases and Phosphatases
Protein kinases
Many proteins in the cell are regulated by phosphorylation, the addition of a phosphate group. A family of enzymes called kinases...
The Nucleosome Core Particle
The paradox
Nucleosomes, paradoxically, perform two opposite functions simultaneously. On the one hand, their main responsibility is to protect the delicate DNA strands from physical damage and help achieve a higher compaction ratio. While on the other hand, they must allow polymerase enzymes to access DNA...
Eukaryotic Transcription Inhibitors
Eukaryotic transcription inhibitors usually contain two distinct domains, a...
Predicting Molecular Geometry

