Advertisement

Journal of Cancer Research and Clinical Oncology

, Volume 144, Issue 8, pp 1435–1444 | Cite as

Machine learning identifies a core gene set predictive of acquired resistance to EGFR tyrosine kinase inhibitor

  • Young Rae Kim
  • Sung Young Kim
Original Article – Cancer Research
  • 87 Downloads

Abstract

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.

Keywords

Epidermal growth factor receptor Protein tyrosine kinases Drug resistance Transcriptomics Computer modeling 

Notes

Acknowledgements

This paper was supported by Konkuk University in 2015.

Author contributions

SYK and YRK conceived, designed the experiments and performed and analyzed the experiments. SYK performed the mathematical and statistical analyses. All authors wrote the paper. All authors analyzed the results and approved the final version of the article.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Informed consent

No human participants were used in this study.

Supplementary material

432_2018_2676_MOESM1_ESM.pdf (1.5 mb)
Supplementary material 1 (PDF 1494 KB)

References

  1. Bailey ST, Miron PL, Choi YJ et al (2014) NF-κB activation-induced anti-apoptosis renders HER2-positive cells drug resistant and accelerates tumor growth. Mol Cancer Res 12:408–420.  https://doi.org/10.1158/1541-7786.MCR-13-0206-T CrossRefPubMedGoogle Scholar
  2. Chang L-C, Lin H-M, Sibille E, Tseng GC (2013) Meta-analysis methods for combining multiple expression profiles: comparisons, statistical characterization and an application guideline. BMC Bioinform 14:368.  https://doi.org/10.1186/1471-2105-14-368 CrossRefGoogle Scholar
  3. Clarke R, Ressom HW, Wang A et al (2008) The properties of high-dimensional data spaces: implications for exploring gene and protein expression data. Nat Rev Cancer 8:37–49.  https://doi.org/10.1038/nrc2294 CrossRefPubMedPubMedCentralGoogle Scholar
  4. Ding Z, Zu S, Gu J (2016) Evaluating the molecule-based prediction of clinical drug responses in cancer. Bioinformatics 32:2891–2895.  https://doi.org/10.1093/bioinformatics/btw344 CrossRefPubMedGoogle Scholar
  5. Eberlein CA, Stetson D, Markovets AA et al (2015) Acquired resistance to the mutant-selective EGFR inhibitor AZD9291 is associated with increased dependence on RAS signaling in preclinical models. Cancer Res 75:2489–2500.  https://doi.org/10.1158/0008-5472.CAN-14-3167 CrossRefPubMedPubMedCentralGoogle Scholar
  6. Friedman J, Hastie T, Tibshirani R (2010) Regularization paths for generalized linear models via coordinate descent. J Stat Softw 33:1–22.  https://doi.org/10.18637/jss.v033.i01 CrossRefPubMedPubMedCentralGoogle Scholar
  7. Giles KM, Kalinowski FC, Candy PA et al (2013) Axl mediates acquired resistance of head and neck cancer cells to the epidermal growth factor receptor inhibitor erlotinib. Mol Cancer Ther 12:2541–2558.  https://doi.org/10.1158/1535-7163.MCT-13-0170 CrossRefPubMedGoogle Scholar
  8. Guix M, Faber AC, Wang SE et al (2008) Acquired resistance to EGFR tyrosine kinase inhibitors in cancer cells is mediated by loss of IGF-binding proteins. J Clin Investig 118:2609–2619.  https://doi.org/10.1172/JCI34588 PubMedGoogle Scholar
  9. Hatakeyama H, Cheng H, Wirth P et al (2010) Regulation of heparin-binding EGF-like growth factor by miR-212 and acquired cetuximab-resistance in head and neck squamous cell carcinoma. PLoS One 5:e12702.  https://doi.org/10.1371/journal.pone.0012702 CrossRefPubMedPubMedCentralGoogle Scholar
  10. Hu T, Li C (2010) Convergence between Wnt-β-catenin and EGFR signaling in cancer. Mol Cancer 9:236.  https://doi.org/10.1186/1476-4598-9-236 CrossRefPubMedPubMedCentralGoogle Scholar
  11. Huang L, Fu L (2015) Mechanisms of resistance to EGFR tyrosine kinase inhibitors. Acta Pharm Sin B 5:390–401.  https://doi.org/10.1016/j.apsb.2015.07.001 CrossRefPubMedPubMedCentralGoogle Scholar
  12. Hughey JJ, Butte AJ (2015) Robust meta-analysis of gene expression using the elastic net. Nucleic Acids Res 43:e79.  https://doi.org/10.1093/nar/gkv229 CrossRefPubMedPubMedCentralGoogle Scholar
  13. Irizarry RA, Hobbs B, Collin F et al (2003) Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4:249–264.  https://doi.org/10.1093/biostatistics/4.2.249 CrossRefPubMedGoogle Scholar
  14. Johnson WE, Li C, Rabinovic A (2007) Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics 8:118–127.  https://doi.org/10.1093/biostatistics/kxj037 CrossRefPubMedGoogle Scholar
  15. Kim E-A, Kim Y-H, Kang HW et al (2015) Lower levels of human MOB3B are associated with prostate cancer susceptibility and aggressive clinicopathological characteristics. J Korean Med Sci 30:937–942.  https://doi.org/10.3346/jkms.2015.30.7.937 CrossRefPubMedPubMedCentralGoogle Scholar
  16. Komurov K, Tseng J-T, Muller M et al (2012) The glucose-deprivation network counteracts lapatinib-induced toxicity in resistant ErbB2-positive breast cancer cells. Mol Syst Biol 8:596.  https://doi.org/10.1038/msb.2012.25 CrossRefPubMedPubMedCentralGoogle Scholar
  17. Kuhn M (2008) Building predictive models in R using the caret Package. J Stat Softw.  https://doi.org/10.18637/jss.v028.i05 Google Scholar
  18. Lever J, Krzywinski M, Altman N (2016) Points of significance: regularization. Nat Methods 13:803–804.  https://doi.org/10.1038/nmeth.4014 CrossRefGoogle Scholar
  19. Liu L, Greger J, Shi H et al (2009) Novel mechanism of lapatinib resistance in HER2-positive breast tumor cells: activation of AXL. Cancer Res 69:6871–6878.  https://doi.org/10.1158/0008-5472.CAN-08-4490 CrossRefPubMedGoogle Scholar
  20. Lorsy E, Topuz AS, Geisler C et al (2016) Loss of dickkopf 3 promotes the tumorigenesis of basal breast cancer. PLoS One 11:e0160077.  https://doi.org/10.1371/journal.pone.0160077 CrossRefPubMedPubMedCentralGoogle Scholar
  21. Sill M, Hielscher T, Becker N, Zucknick M (2014) c060: extended inference with Lasso and elastic-net regularized Cox and generalized linear models. J Stat Softw 62:1–22.  https://doi.org/10.18637/jss.v062.i05 CrossRefGoogle Scholar
  22. Stanam A, Love-Homan L, Joseph TS et al (2015) Upregulated interleukin-6 expression contributes to erlotinib resistance in head and neck squamous cell carcinoma. Mol Oncol 9:1371–1383.  https://doi.org/10.1016/j.molonc.2015.03.008 CrossRefPubMedPubMedCentralGoogle Scholar
  23. Tai W, Mahato R, Cheng K (2010) The role of HER2 in cancer therapy and targeted drug delivery. J Control Release 146:264–275.  https://doi.org/10.1016/j.jconrel.2010.04.009 CrossRefPubMedPubMedCentralGoogle Scholar
  24. Troyanskaya O, Cantor M, Sherlock G et al (2001) Missing value estimation methods for DNA microarrays. Bioinformatics 17:520–525.  https://doi.org/10.1093/bioinformatics/17.6.520 CrossRefPubMedGoogle Scholar
  25. Waldmann P, Mészáros G, Gredler B et al (2013) Evaluation of the lasso and the elastic net in genome-wide association studies. Front Genet 4:270.  https://doi.org/10.3389/fgene.2013.00270 CrossRefPubMedPubMedCentralGoogle Scholar
  26. Wood ER, Truesdale AT, McDonald OB et al (2004) A unique structure for epidermal growth factor receptor bound to GW572016 (Lapatinib): relationships among protein conformation, inhibitor off-rate, and receptor activity in tumor cells. Cancer Res 64:6652–6659.  https://doi.org/10.1158/0008-5472.CAN-04-1168 CrossRefPubMedGoogle Scholar
  27. Zhang Z, Lee JC, Lin L et al (2012) Activation of the AXL kinase causes resistance to EGFR-targeted therapy in lung cancer. Nat Genet 44:852–860.  https://doi.org/10.1038/ng.2330 CrossRefPubMedPubMedCentralGoogle Scholar
  28. Zou H, Hastie T (2005) Regularization and variable selection via the elastic net. J R Stat Soc B 67:301–320.  https://doi.org/10.1111/j.1467-9868.2005.00503.x CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Department of BiochemistryKonkuk University School of MedicineSeoulRepublic of Korea

Personalised recommendations