Personalized Prediction of Acquired Resistance to EGFR-Targeted Inhibitors Using a Pathway-Based Machine Learning Approach.

Young Kim,Hye Yang,Sung Kim,Suh Lee,Yong Kim

doi:10.3390/cancers11010045

Young Kim, Hye Yang + Show 3 more

Open Access

https://doi.org/10.3390/cancers11010045

Copy DOI

Journal: Cancers	Publication Date: Jan 4, 2019
Citations: 6	License type: CC BY 4.0

Affiliation: Trinity College Dublin, Konkuk University

Abstract

Epidermal growth factor receptor (EGFR) inhibitors have benefitted cancer patients worldwide, but resistance inevitably develops over time, resulting in treatment failures. An accurate prediction model for acquired resistance (AR) to EGFR inhibitors is critical for early diagnosis and according intervention, but is not yet available due to personal variations and the complex mechanisms of AR. Here, we have developed a novel pipeline to build a meta-analysis-based, multivariate model for personalized pathways in AR to EGFR inhibitors, using sophisticated machine learning algorithms. Surprisingly, the model achieved excellent predictive performance, with a cross-study validation area under curve (AUC) of over 0.9, and generalization performance on independent cohorts of samples, with a perfect AUC score of 1. Furthermore, the model showed excellent transferability across different cancer cell lines and EGFR inhibitors, including gefitinib, erlotinib, afatinib, and cetuximab. In conclusion, our model achieved high predictive accuracy through robust cross study validation, and enabled individualized prediction on newly introduced data. We also discovered common pathway alteration signatures for AR to EGFR inhibitors, which can provide directions for other follow-up studies.

Highlights

Despite the initial benefits of Epidermal growth factor receptor (EGFR) inhibitors in cancer patients harboring EGFR mutations, the rapid development of acquired resistance (AR) is a major obstacle in clinical practice and often leads to therapeutic failure and disease recurrence
To build a robust and generalized prediction model based on individualized pathway information, we developed a novel pipeline that integrates meta-analysis-based regularized regression with pathway-level measurement of abnormality (Figure 1)
The study cohort was very heterogeneous in terms of the types of EGFR inhibitors, platforms, and cancer cell lines (Table S1)

Summary

Introduction

Despite the initial benefits of EGFR inhibitors in cancer patients harboring EGFR mutations, the rapid development of acquired resistance (AR) is a major obstacle in clinical practice and often leads to therapeutic failure and disease recurrence. A broad range of mechanisms of AR to EGFR inhibitors have been proposed, from mutational to non-mutation-based mechanisms. The exact mechanisms still remain unclear due to the multifactorial natures of cancer and intracellular signaling networks. Inherent crosstalk and redundancy of signaling pathways introduces huge complexity [1,2]. Inhibiting a single signaling network via drugs may trigger other survival pathways and limit efficacy. These complex dynamics make it more difficult to understand the underlying causes of AR and predict potential EGFR inhibitor sensitivity

Methods

Results

Discussion

Conclusion