Journal of Engineering, Project, and Production Management, 2026, 16(5), 2025-315
Engineering Predictive AI/ML Pipelines for EGFR-TKI Resistance in NSCLC: A Systematic Review
1 Instructor,
Department of Computer Science, College of Information Technology, Misr
University for Science and Technology (MUST), P.O. Box 77, Giza, Egypt.
E-mail: faris.eltomy@must.edu.eg (corresponding author), https://orcid.org/0009-0003-0590-6186
2 Assistant Professor, Department
of Artificial Intelligence, College of Information Technology, Misr
University for Science and Technology (MUST), P.O. Box 77, Giza, Egypt.
E-mail: mohanad.deif@must.edu.eg, https://orcid.org/0000-0002-4388-1480
3 Instructor, Department of
Computer Science, College of Information Technology, Misr University for
Science and Technology (MUST), P.O. Box 77, Giza, Egypt.
E-mail: alaa.zaghloul@must.edu.eg, https://orcid.org/0000-0001-8171-3659
4 Professor, Department of
Information Systems, College of Computer and Information Sciences, Ain
Shams University, Cairo, Egypt.E-mail: rania.elgohary@cis.asu.edu.eg,
https://orcid.org/0000-0001-5284-8343
Engineering Management
Received October 14, 2025; revised December 29, 2025; accepted January 29, 2026
Available online June 17, 2026
Abstract: Resistance to Epidermal Growth Factor Receptor (EGFR) Tyrosine Kinase Inhibitors (TKIs) is a significant challenge in Non-Small Cell Lung Cancer (NSCLC), often resulting in disease progression and reduced survival. Artificial Intelligence (AI) and Machine Learning (ML) models have recently been evaluated for their ability to predict EGFR-TKI resistance and guide treatment. This systematic review assesses studies using AI/ML to forecast EGFR-TKI resistance from radiomics, transcriptomics, clinical, or multi-omics data. A systematic search was conducted across PubMed, Web of Science, and Scopus through March 15, 2025. Studies included applied AI/ML models to predict resistance to EGFR-TKIs in NSCLC. Extracted data included model inputs, performance, and methodological quality, with risk of bias assessed using the PROBAST tool. Ten studies met the inclusion criteria. Models showed promising predictive performance, including radiomics-based models Area Under the Curve (AUC) up to 0.86) and molecular dynamics approaches (accuracy up to 97.5%). Deep learning models stratified patients by mutation status and survival. However, most studies had methodological limitations, including suboptimal measurement of predictors, missing outcomes, lack of external validation, and overfitting. Only two studies had a low or unclear risk of bias across all PROBAST domains. AI/ML models have the potential to predict EGFR-TKI resistance in NSCLC, but methodological heterogeneity and quality issues limit their current clinical utility. Rigorous study design, external validation, and transparent reporting are needed for reliable integration into precision oncology.
Keywords: Non-small cell lung cancer (NSCLC); artificial intelligence (AI); machine learning (ML); EGFR-TKI resistance; predictive modeling.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Hassan, F., Deif, M. A., Zaghloul, A., and Elgohary, R. (2026). Engineering Predictive AI/ML Pipelines for EGFR-TKI Resistance in NSCLC: A Systematic Review. Journal of Engineering, Project, and Production Management, 16(5), 2025-315.
