Skip to main content

Advertisement

SpringerLink
Log in

Multivariate machine learning models for prediction of pathologic response to neoadjuvant therapy in breast cancer using MRI features: a study using an independent validation set

  • Epidemiology
  • Published:
Breast Cancer Research and Treatment Aims and scope Submit manuscript

Abstract

Purpose

To determine whether a multivariate machine learning-based model using computer-extracted features of pre-treatment dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict pathologic complete response (pCR) to neoadjuvant therapy (NAT) in breast cancer patients.

Methods

Institutional review board approval was obtained for this retrospective study of 288 breast cancer patients at our institution who received NAT and had a pre-treatment breast MRI. A comprehensive set of 529 radiomic features was extracted from each patient’s pre-treatment MRI. The patients were divided into equal groups to form a training set and an independent test set. Two multivariate machine learning models (logistic regression and a support vector machine) based on imaging features were trained to predict pCR in (a) all patients with NAT, (b) patients with neoadjuvant chemotherapy (NACT), and (c) triple-negative or human epidermal growth factor receptor 2-positive (TN/HER2+) patients who had NAT. The multivariate models were tested using the independent test set, and the area under the receiver operating characteristics (ROC) curve (AUC) was calculated.

Results

Out of the 288 patients, 64 achieved pCR. The AUC values for predicting pCR in TN/HER+ patients who received NAT were significant (0.707, 95% CI 0.582–0.833, p < 0.002).

Conclusions

The multivariate models based on pre-treatment MRI features were able to predict pCR in TN/HER2+ patients.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

References

  1. von Minckwitz G, Untch M, Blohmer J-U et al (2012) Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 30(15):1796–1804. https://doi.org/10.1200/JCO.2011.38.8595

    Article  Google Scholar 

  2. Kaufmann M, von Minckwitz G, Bear HD et al (2007) Recommendations from an international expert panel on the use of neoadjuvant (primary) systemic treatment of operable breast cancer: new perspectives 2006. Ann Oncol 18(12):1927–1934. https://doi.org/10.1093/annonc/mdm201

    Article  PubMed  CAS  Google Scholar 

  3. Kaufmann M, von Minckwitz G, Smith R et al (2003) International expert panel on the use of primary (preoperative) systemic treatment of operable breast cancer: review and recommendations. J Clin Oncol 21(13):2600–2608. https://doi.org/10.1200/JCO.2003.01.136

    Article  PubMed  Google Scholar 

  4. Heys SD, Hutcheon AW, Sarkar TK et al (2002) Neoadjuvant docetaxel in breast cancer: 3-year survival results from the Aberdeen trial. Clin Breast Cancer 3:S69–S74. https://doi.org/10.3816/CBC.2002.s.015

    Article  PubMed  Google Scholar 

  5. van der Hage JH, van de Velde CC, Mieog SJ (2007) Preoperative chemotherapy for women with operable breast cancer. In: Mieog SJ (ed) Cochrane database of systematic reviews. John Wiley & Sons, Ltd, Chichester. https://doi.org/10.1002/14651858.CD005002.pub2

    Chapter  Google Scholar 

  6. Fisher B, Bryant J, Wolmark N et al (1998) Effect of preoperative chemotherapy on the outcome of women with operable breast cancer. J Clin Oncol 16(8):2672–2685. https://doi.org/10.1200/JCO.1998.16.8.2672

    Article  PubMed  CAS  Google Scholar 

  7. van der Hage JA, van de Velde CJ, Julien JP et al (2001) Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 19(22):4224–4237. https://doi.org/10.1200/JCO.2001.19.22.4224

    Article  PubMed  Google Scholar 

  8. Michoux N, Van den Broeck S, Lacoste L et al (2015) Texture analysis on MR images helps predicting non-response to NAC in breast cancer. BMC Cancer 15(1):574. https://doi.org/10.1186/s12885-015-1563-8

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  9. Barbi GP, Marroni P, Bruzzi P, Nicolò G, Paganuzzi M, Ferrara GB (1987) Correlation between steroid hormone receptors and prognostic factors in human breast cancer. Oncology 44(5):265–269. https://doi.org/10.1159/000226492

    Article  PubMed  CAS  Google Scholar 

  10. von Minckwitz G, Sinn H-P, Raab G et al (2008) Clinical response after two cycles compared to HER2, Ki-67, p53, and bcl-2 in independently predicting a pathological complete response after preoperative chemotherapy in patients with operable carcinoma of the breast. Breast Cancer Res 10(2):R30. https://doi.org/10.1186/bcr1989

    Article  CAS  Google Scholar 

  11. Esserman L, Kaplan E, Partridge S et al (2001) MRI phenotype is associated with response to doxorubicin and cyclophosphamide neoadjuvant chemotherapy in stage III breast cancer. Ann Surg Oncol 8(6):549–559. https://doi.org/10.1007/s10434-001-0549-8

    Article  PubMed  CAS  Google Scholar 

  12. Nishimura R, Osako T, Okumura Y, Hayashi M, Arima N (2010) Clinical significance of Ki-67 in neoadjuvant chemotherapy for primary breast cancer as a predictor for chemosensitivity and for prognosis. Breast Cancer 17(4):269–275. https://doi.org/10.1007/s12282-009-0161-5

    Article  PubMed  Google Scholar 

  13. Fangberget A, Nilsen LB, Hole KH et al (2011) Neoadjuvant chemotherapy in breast cancer-response evaluation and prediction of response to treatment using dynamic contrast-enhanced and diffusion-weighted MR imaging. Eur Radiol 21(6):1188–1199. https://doi.org/10.1007/s00330-010-2020-3

    Article  PubMed  CAS  Google Scholar 

  14. Uematsu T, Kasami M, Yuen S (2010) Neoadjuvant chemotherapy for breast cancer: correlation between the baseline MR imaging findings and responses to therapy. Eur Radiol 20(10):2315–2322. https://doi.org/10.1007/s00330-010-1813-8

    Article  PubMed  Google Scholar 

  15. Pickles MD, Manton DJ, Lowry MTL (2009) Prognostic value of pre-treatment DCE-MRI parameters in predicting disease free and overall survival for breast cancer patients undergoing neoadjuvant chemotherapy. Eur J Radiol 71(3):498–505. https://doi.org/10.1016/J.EJRAD.2008.05.007

    Article  PubMed  Google Scholar 

  16. Craciunescu OI, Blackwell KL, Jones EL et al (2009) DCE-MRI parameters have potential to predict response of locally advanced breast cancer patients to neoadjuvant chemotherapy and hyperthermia: A pilot study. Int J Hyperth 25(6):405–415. https://doi.org/10.1080/02656730903022700

    Article  CAS  Google Scholar 

  17. Smith IC, Heys SD, Hutcheon AW et al (2002) Neoadjuvant chemotherapy in breast cancer: significantly enhanced response with docetaxel. J Clin Oncol 20(6):1456–1466. https://doi.org/10.1200/JCO.2002.20.6.1456

    Article  PubMed  CAS  Google Scholar 

  18. Kawamura M, Satake H, Ishigaki S, Nishio A, Sawaki M, Naganawa S (2011) Early prediction of response to neoadjuvant chemotherapy for locally advanced breast cancer using MRI. Nagoya J Med Sci 73(3–4):147–156

    Google Scholar 

  19. Goorts B, van Nijnatten TJA, de Munck L et al (2017) Clinical tumor stage is the most important predictor of pathological complete response rate after neoadjuvant chemotherapy in breast cancer patients. Breast Cancer Res Treat 163(1):83–91. https://doi.org/10.1007/s10549-017-4155-2

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  20. Braman NM, Etesami M, Prasanna P et al (2017) Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI. Breast Cancer Res 19(1):57. https://doi.org/10.1186/s13058-017-0846-1

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Fan M, Wu G, Cheng H, Zhang J, Shao G, Li L (2017) Radiomic analysis of DCE-MRI for prediction of response to neoadjuvant chemotherapy in breast cancer patients. Eur J Radiol 94:140–147. https://doi.org/10.1016/J.EJRAD.2017.06.019

    Article  PubMed  CAS  Google Scholar 

  22. Eom H-J, Cha JH, Choi WJ, Chae EY, Shin HJ, Kim HH (2017) Predictive clinicopathologic and dynamic contrast-enhanced MRI findings for tumor response to neoadjuvant chemotherapy in triple-negative breast cancer. Am J Roentgenol 208(6):W225–W230. https://doi.org/10.2214/AJR.16.17125

    Article  Google Scholar 

  23. Chamming’s F, Ueno Y, Ferré R et al (2017) Features from computerized texture analysis of breast cancers at pretreatment MR imaging are associated with response to neoadjuvant chemotherapy. Radiology. https://doi.org/10.1148/radiol.2017170143

    Article  PubMed  Google Scholar 

  24. Aghaei F, Tan M, Hollingsworth AB, Qian W, Liu H, Zheng B (2015) Computer-aided breast MR image feature analysis for prediction of tumor response to chemotherapy. Med Phys 42(11):6520–6528. https://doi.org/10.1118/1.4933198

    Article  PubMed  PubMed Central  Google Scholar 

  25. Aghaei F, Tan M, Hollingsworth AB, Zheng B, Cheng S (2016) Computer-aided global breast MR image feature analysis for prediction of tumor response to chemotherapy: performance assessment. In: Tourassi GD, Armato SG (eds) International society for optics and photonics. https://doi.org/10.1117/12.2216326

  26. Ahmed A, Gibbs P, Pickles M, Turnbull L (2013) Texture analysis in assessment and prediction of chemotherapy response in breast cancer. J Magn Reson Imaging 38(1):89–101. https://doi.org/10.1002/jmri.23971

    Article  PubMed  Google Scholar 

  27. Nilsen L, Fangberget A, Geier O, Olsen DR, Seierstad T (2010) Diffusion-weighted magnetic resonance imaging for pretreatment prediction and monitoring of treatment response of patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy. Acta Oncol (Madr) 49(3):354–360. https://doi.org/10.3109/02841861003610184

    Article  Google Scholar 

  28. Tudorica A, Oh KY, Chui SY-C et al (2016) Early prediction and evaluation of breast cancer response to neoadjuvant chemotherapy using quantitative DCE-MRI 1. Transl Oncol 9:8–17. https://doi.org/10.1016/j.tranon.2015.11.016

    Article  PubMed  PubMed Central  Google Scholar 

  29. New response evaluation criteria in solid tumours (2009) Revised RECIST guideline (version 1.1). Eur J Cancer 45(2):228–247. https://doi.org/10.1016/J.EJCA.2008.10.026

    Article  Google Scholar 

  30. Bezdek JC (2013) Pattern recognition with fuzzy objective function algorithms. Springer Science & Business Media, New York

    Google Scholar 

  31. Saha A, Yu X, Sahoo D, Mazurowski MA (2017) Effects of MRI scanner parameters on breast cancer radiomics. Expert Syst Appl 87:384–391. https://doi.org/10.1016/j.eswa.2017.06.029

    Article  PubMed  PubMed Central  Google Scholar 

  32. Team RC (2014) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna 2014.

    Google Scholar 

  33. DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44(3):837. https://doi.org/10.2307/2531595

    Article  PubMed  CAS  Google Scholar 

  34. Guyon I, Elisseeff A, De AM (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182

    Google Scholar 

  35. Dave RV, Millican-Slater R, Dodwell D, Horgan K, Sharma N (2017) Neoadjuvant chemotherapy with MRI monitoring for breast cancer. Br J Surg 104(9):1177–1187. https://doi.org/10.1002/bjs.10544

    Article  PubMed  CAS  Google Scholar 

  36. Sharma U, Danishad KKA, Seenu V, Jagannathan NR (2009) Longitudinal study of the assessment by MRI and diffusion-weighted imaging of tumor response in patients with locally advanced breast cancer undergoing neoadjuvant chemotherapy. NMR Biomed 22(1):104–113. https://doi.org/10.1002/nbm.1245

    Article  PubMed  Google Scholar 

  37. Cleator S, Heller W, Coombes RC (2007) Triple-negative breast cancer: therapeutic options. Lancet Oncol 8(3):235–244. https://doi.org/10.1016/S1470-2045(07)70074-8

    Article  PubMed  Google Scholar 

  38. Liedtke C, Mazouni C, Hess KR et al (2008) Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol 26(8):1275–1281. https://doi.org/10.1200/JCO.2007.14.4147

    Article  PubMed  Google Scholar 

  39. Carey LA, Dees EC, Sawyer L et al (2007) The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res 13(8):2329–2334. https://doi.org/10.1158/1078-0432.CCR-06-1109

    Article  PubMed  CAS  Google Scholar 

  40. Gianni L, Eiermann W, Semiglazov V et al (2010) Neoadjuvant chemotherapy with trastuzumab followed by adjuvant trastuzumab versus neoadjuvant chemotherapy alone, in patients with HER2-positive locally advanced breast cancer (the NOAH trial): a randomised controlled superiority trial with a parallel HER. Lancet 375(9712):377–384. https://doi.org/10.1016/S0140-6736(09)61964-4

    Article  PubMed  CAS  Google Scholar 

  41. Buzdar AU, Suman VJ, Meric-Bernstam F et al (2013) Fluorouracil, epirubicin, and cyclophosphamide (FEC-75) followed by paclitaxel plus trastuzumab versus paclitaxel plus trastuzumab followed by FEC-75 plus trastuzumab as neoadjuvant treatment for patients with HER2-positive breast cancer (Z1041): a random. Lancet Oncol 14(13):1317–1325. https://doi.org/10.1016/S1470-2045(13)70502-3

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Robidoux A, Tang G, Rastogi P et al (2013) Lapatinib as a component of neoadjuvant therapy for HER2-positive operable breast cancer (NSABP protocol B-41): an open-label, randomised phase 3 trial. Lancet Oncol 14(12):1183–1192. https://doi.org/10.1016/S1470-2045(13)70411-X

    Article  PubMed  CAS  Google Scholar 

  43. Untch M, Loibl S, Bischoff J et al (2012) Lapatinib versus trastuzumab in combination with neoadjuvant anthracycline-taxane-based chemotherapy (GeparQuinto, GBG 44): a randomised phase 3 trial. Lancet Oncol 13(2):135–144. https://doi.org/10.1016/S1470-2045(11)70397-7

    Article  PubMed  CAS  Google Scholar 

  44. Cortazar P, Zhang L, Untch M et al (2014) Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet 384(9938):164–172. https://doi.org/10.1016/S0140-6736(13)62422-8

    Article  PubMed  Google Scholar 

  45. Chen X, Ye G, Zhang C et al (2013) Superior outcome after neoadjuvant chemotherapy with docetaxel, anthracycline, and cyclophosphamide versus docetaxel plus cyclophosphamide: results from the NATT trial in triple negative or HER2 positive breast cancer. Breast Cancer Res Treat 142(3):549–558. https://doi.org/10.1007/s10549-013-2761-1

    Article  PubMed  CAS  Google Scholar 

  46. Earl H, Provenzano E, Abraham J et al (2015) Neoadjuvant trials in early breast cancer: pathological response at surgery and correlation to longer term outcomes—what does it all mean? BMC Med 13(1):234. https://doi.org/10.1186/s12916-015-0472-7

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

Funding was provided by North Carolina Biotechnology Center (Grant No. 2016-BIG-6520), National Institutes of Health (Grant No. 1R01EB021360).

Author information

Authors and Affiliations

  1. Department of Radiology, Duke University School of Medicine, 2301 Erwin Road, Durham, NC, 27705, USA

    Elizabeth Hope Cain, Ashirbani Saha, Michael R. Harowicz & Maciej A. Mazurowski

  2. Department of Radiology, Johns Hopkins University, Baltimore, MD, USA

    Michael R. Harowicz

  3. Department of Surgery, Duke University School of Medicine, 2301 Erwin Road, Durham, NC, 27705, USA

    Jeffrey R. Marks

  4. Department of Medicine, Duke University School of Medicine, 2301 Erwin Road, Durham, NC, 27705, USA

    P. Kelly Marcom

  5. Department of Electrical and Computer Engineering, Duke University, Durham, NC, USA

    Maciej A. Mazurowski

Authors
  1. Elizabeth Hope Cain
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Ashirbani Saha
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Michael R. Harowicz
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jeffrey R. Marks
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. P. Kelly Marcom
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Maciej A. Mazurowski
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Elizabeth Hope Cain.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cain, E.H., Saha, A., Harowicz, M.R. et al. Multivariate machine learning models for prediction of pathologic response to neoadjuvant therapy in breast cancer using MRI features: a study using an independent validation set. Breast Cancer Res Treat 173, 455–463 (2019). https://doi.org/10.1007/s10549-018-4990-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10549-018-4990-9

Keywords

Search

Navigation