Advertisement

Machine learning-based radiomic models to predict intensity-modulated radiation therapy response, Gleason score and stage in prostate cancer

  • Hamid Abdollahi
  • Bahram Mofid
  • Isaac Shiri
  • Abolfazl Razzaghdoust
  • Afshin Saadipoor
  • Arash Mahdavi
  • Hassan Maleki Galandooz
  • Seied Rabi MahdaviEmail author
ONCOLOGY IMAGING

Abstract

Objective

To develop different radiomic models based on the magnetic resonance imaging (MRI) radiomic features and machine learning methods to predict early intensity-modulated radiation therapy (IMRT) response, Gleason scores (GS) and prostate cancer (Pca) stages.

Methods

Thirty-three Pca patients were included. All patients underwent pre- and post-IMRT T2-weighted (T2 W) and apparent diffusing coefficient (ADC) MRI. IMRT response was calculated in terms of changes in the ADC value, and patients were divided as responders and non-responders. A wide range of radiomic features from different feature sets were extracted from all T2 W and ADC images. Univariate radiomic analysis was performed to find highly correlated radiomic features with IMRT response, and a paired t test was used to find significant features between responders and non-responders. To find high predictive radiomic models, tenfold cross-validation as the criterion for feature selection and classification was applied on the pre-, post- and delta IMRT radiomic features, and area under the curve (AUC) of receiver operating characteristics was calculated as model performance value.

Results

Of 33 patients, 15 patients (45%) were found as responders. Univariate analysis showed 20 highly correlated radiomic features with IMRT response (20 ADC and 20 T2). Two and fifteen T2 and ADC radiomic features were found as significant (P-value ≤ 0.05) features between responders and non-responders, respectively. Several cross-combined predictive radiomic models were obtained, and post-T2 radiomic models were found as high predictive models (AUC 0.632) followed by pre-ADC (AUC 0.626) and pre-T2 (AUC 0.61). For GS prediction, T2 W radiomic models were found as more predictive (mean AUC 0.739) rather than ADC models (mean AUC 0.70), while for stage prediction, ADC models had higher prediction performance (mean AUC 0.675).

Conclusions

Radiomic models developed by MR image features and machine learning approaches are noninvasive and easy methods for personalized prostate cancer diagnosis and therapy.

Keywords

Radiomics Prostate cancer MRI IMRT Prediction 

Notes

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

All procedures performed were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all participants included in the study.

Supplementary material

11547_2018_966_MOESM1_ESM.doc (145 kb)
Supplementary material 1 (DOC 145 kb)

References

  1. 1.
    Zelefsky MJ, Fuks Z, Hunt M, Yamada Y, Marion C, Ling CC et al (2002) High-dose intensity modulated radiation therapy for prostate cancer: early toxicity and biochemical outcome in 772 patients. Int J Radiat Oncol Biol Phys 53(5):6–1111Google Scholar
  2. 2.
    Kumar V, Bora GS, Kumar R, Jagannathan NR (2018) Multiparametric (mp) MRI of prostate cancer. Prog Nucl Magn Reson 105:23–40Google Scholar
  3. 3.
    Wallace T, Torre T, Grob M, Yu J, Avital I, Brücher B et al (2014) Current approaches, challenges and future directions for monitoring treatment response in prostate cancer. J Cancer 5(1):3Google Scholar
  4. 4.
    Kelloff GJ, Choyke P, Coffey DS (2009) Challenges in clinical prostate cancer: role of imaging. Am J Roentgenol 192(6):70–1455Google Scholar
  5. 5.
    Mazaheri Y, Akin O, Hricak H (2017) Dynamic contrast-enhanced magnetic resonance imaging of prostate cancer: a review of current methods and applications. World J Radiol 9(12):416Google Scholar
  6. 6.
    Aerts HJ, Velazquez ER, Leijenaar RT, Parmar C, Grossmann P, Carvalho S et al (2014) Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 5:4006Google Scholar
  7. 7.
    Abdollahi H, Mostafaei S, Cheraghi S, Shiri I, Mahdavi SR, Kazemnejad A (2018) Cochlea CT radiomics predicts chemoradiotherapy induced sensorineural hearing loss in head and neck cancer patients: a machine learning and multi-variable modelling study. Phys Med 45:7–192Google Scholar
  8. 8.
    Stoyanova R, Takhar M, Tschudi Y, Ford JC, Solórzano G, Erho N et al (2016) Prostate cancer radiomics and the promise of radiogenomics. Transl Cancer Res 5(4):432Google Scholar
  9. 9.
    X-k Niu, Z-f Chen, Chen L, Li J, Peng T, Li X (2018) Clinical application of biparametric MRI texture analysis for detection and evaluation of high-grade prostate cancer in zone-specific regions. Am J Roentgenol 210(3):56–549Google Scholar
  10. 10.
    Bates A, Miles K (2017) Prostate-specific membrane antigen PET/MRI validation of MR textural analysis for detection of transition zone prostate cancer. Eur Radiol 27(12):8–5290Google Scholar
  11. 11.
    Nketiah G, Elschot M, Kim E, Teruel JR, Scheenen TW, Bathen TF et al (2017) T2-weighted MRI-derived textural features reflect prostate cancer aggressiveness: preliminary results. Eur Radiol 27(7):9–3050Google Scholar
  12. 12.
    Gnep K, Fargeas A, Gutiérrez-Carvajal RE, Commandeur F, Mathieu R, Ospina JD et al (2017) Haralick textural features on T2-weighted MRI are associated with biochemical recurrence following radiotherapy for peripheral zone prostate cancer. J Magn Reson Imaging 45(1):17–103Google Scholar
  13. 13.
    Stoyanova R, Pollack A, Takhar M, Lynne C, Parra N, Lam LL et al (2016) Association of multiparametric MRI quantitative imaging features with prostate cancer gene expression in MRI-targeted prostate biopsies. Oncotarget 7(33):53362Google Scholar
  14. 14.
    Dzik-Jurasz A, Domenig C, George M, Wolber J, Padhani A, Brown G et al (2002) Diffusion MRI for prediction of response of rectal cancer to chemoradiation. Lancet 360(9329):8–307Google Scholar
  15. 15.
    Mardor Y, Roth Y, Lidar Z, Jonas T, Pfeffer R, Maier SE et al (2001) Monitoring response to convection-enhanced taxol delivery in brain tumor patients using diffusion-weighted magnetic resonance imaging. Cancer Res 61(13):3–4971Google Scholar
  16. 16.
    Pearson R, Pieniazek P, Thelwall P, Maxwell R, Plummer R, Frew J (2018) Diffusion-weighted MRI for early response assessment in the treatment of bladder cancer. Clin Oncol 30(3):193Google Scholar
  17. 17.
    Thoeny HC, Ross BD (2010) Predicting and monitoring cancer treatment response with diffusion-weighted MRI. J Magn Reson Imaging 32(1):2–16Google Scholar
  18. 18.
    Decker G, Mürtz P, Gieseke J, Träber F, Block W, Sprinkart AM et al (2014) Intensity-modulated radiotherapy of the prostate: dynamic ADC monitoring by DWI at 3.0 T. Radiother Oncol 113(1):20–115Google Scholar
  19. 19.
    Desouza N, Reinsberg S, Scurr E, Brewster J, Payne G (2007) Magnetic resonance imaging in prostate cancer: the value of apparent diffusion coefficients for identifying malignant nodules. Br J Radiol 80(950):5–90Google Scholar
  20. 20.
    Wolf MB, Edler C, Tichy D, Röthke MC, Schlemmer HP, Herfarth K et al (2017) Diffusion-weighted MRI treatment monitoring of primary hypofractionated proton and carbon ion prostate cancer irradiation using raster scan technique. J Magn Reson Imaging 46(3):60–850Google Scholar
  21. 21.
    Parmar C, Grossmann P, Bussink J, Lambin P, Aerts HJ (2015) Machine learning methods for quantitative radiomic biomarkers. Sci Rep 5:13087Google Scholar
  22. 22.
    Parmar C, Grossmann P, Rietveld D, Rietbergen MM, Lambin P, Aerts HJ (2015) Radiomic machine-learning classifiers for prognostic biomarkers of head and neck cancer. Front Oncol 5:272Google Scholar
  23. 23.
    Zhou M, Scott J, Chaudhury B, Hall L, Goldgof D, Yeom K et al (2018) Radiomics in brain tumor: image assessment, quantitative feature descriptors, and machine-learning approaches. Am J Neuroradiol 39(2):16–208Google Scholar
  24. 24.
    Zhang B, He X, Ouyang F, Gu D, Dong Y, Zhang L et al (2017) Radiomic machine-learning classifiers for prognostic biomarkers of advanced nasopharyngeal carcinoma. Cancer Lett 403:7–21Google Scholar
  25. 25.
    Song I, Kim CK, Park BK, Park W (2010) Assessment of response to radiotherapy for prostate cancer: value of diffusion-weighted MRI at 3 T. Am J Roentgenol 194(6):W82–W477Google Scholar
  26. 26.
    Wahba MH, Morad MM (2015) The role of diffusion-weighted MRI: in assessment of response to radiotherapy for prostate cancer. Egypt J Radiol Nucl Med 46(1):8–183Google Scholar
  27. 27.
    Belli P, Costantini M, Ierardi C, Bufi E, Amato D, Mule A et al (2011) Diffusion-weighted imaging in evaluating the response to neoadjuvant breast cancer treatment. Breast J 17(6):9–610Google Scholar
  28. 28.
    Khalil RF, Abdelhamid AEM, Darwish AMA, Hassan HHM (2017) Diffusion weighted imaging in early prediction of neoadjuvant chemotherapy response in breast cancer. Egypt J Radiol Nucl Med 48(2):35–529Google Scholar
  29. 29.
    Collewet G, Strzelecki M, Mariette F (2004) Influence of MRI acquisition protocols and image intensity normalization methods on texture classification. Magn Reson Imaging 22(1):81–91Google Scholar
  30. 30.
    Bollineni VR, Widder J, Pruim J, Langendijk JA, Wiegman EM (2012) Residual 18F-FDG-PET uptake 12 weeks after stereotactic ablative radiotherapy for stage I non-small-cell lung cancer predicts local control. Int J Radiat Oncol Biol Phys 83(4):e5–e551Google Scholar
  31. 31.
    Liu Z, Zhang X-Y, Shi Y-J, Wang L, Zhu H-T, Tang Z-C, et al. Radiomics analysis for evaluation of pathological complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer. Clin Cancer Res. 2017:clincanres. 1038.2017Google Scholar
  32. 32.
    Aerts HJ, Grossmann P, Tan Y, Oxnard GR, Rizvi N, Schwartz LH et al (2016) Defining a radiomic response phenotype: a pilot study using targeted therapy in NSCLC. Sci Rep 6:33860Google Scholar
  33. 33.
    Yue Y, Osipov A, Fraass B, Sandler H, Zhang X, Nissen N et al (2017) Identifying prognostic intratumor heterogeneity using pre-and post-radiotherapy 18F-FDG PET images for pancreatic cancer patients. J Gastrointest Oncol 8(1):127Google Scholar
  34. 34.
    Rao S-X, Lambregts DM, Schnerr RS, Beckers RC, Maas M, Albarello F et al (2016) CT texture analysis in colorectal liver metastases: a better way than size and volume measurements to assess response to chemotherapy? United Eur Gastroenterol J 4(2):63–257Google Scholar
  35. 35.
    Goh V, Ganeshan B, Nathan P, Juttla JK, Vinayan A, Miles KA (2011) Assessment of response to tyrosine kinase inhibitors in metastatic renal cell cancer: CT texture as a predictive biomarker. Radiology 261(1):71–165Google Scholar
  36. 36.
    Cunliffe A, Armato SG III, Castillo R, Pham N, Guerrero T, Al-Hallaq HA (2015) Lung texture in serial thoracic computed tomography scans: correlation of radiomics-based features with radiation therapy dose and radiation pneumonitis development. Int J Radiat Oncol Biol Phys 91(5):56–1048Google Scholar
  37. 37.
    Fave X, Zhang L, Yang J, Mackin D, Balter P, Gomez D et al (2017) Delta-radiomics features for the prediction of patient outcomes in non-small cell lung cancer. Sci Rep 7(1):588Google Scholar
  38. 38.
    Coroller TP, Agrawal V, Narayan V, Hou Y, Grossmann P, Lee SW et al (2016) Radiomic phenotype features predict pathological response in non-small cell lung cancer. Radiother Oncol 119(3):6–480Google Scholar
  39. 39.
    Mattonen SA, Palma DA, Haasbeek CJ, Senan S, Ward AD (2014) Early prediction of tumor recurrence based on CT texture changes after stereotactic ablative radiotherapy (SABR) for lung cancer. Med Phys 41(3):033502Google Scholar
  40. 40.
    Lucia F, Visvikis D, Desseroit M-C, Miranda O, Malhaire J-P, Robin P et al (2018) Prediction of outcome using pretreatment 18 F-FDG PET/CT and MRI radiomics in locally advanced cervical cancer treated with chemoradiotherapy. Eur J Nucl Med Mol Imaging 45(5):86–768Google Scholar
  41. 41.
    Hawkins SH, Korecki JN, Balagurunathan Y, Gu Y, Kumar V, Basu S et al (2014) Predicting outcomes of nonsmall cell lung cancer using CT image features. IEEE Access 2:26–1418Google Scholar
  42. 42.
    Parmar C, Leijenaar RT, Grossmann P, Velazquez ER, Bussink J, Rietveld D et al (2015) Radiomic feature clusters and prognostic signatures specific for lung and head & neck cancer. Sci Rep 5:11044Google Scholar
  43. 43.
    Fehr D, Veeraraghavan H, Wibmer A, Gondo T, Matsumoto K, Vargas HA et al (2015) Automatic classification of prostate cancer Gleason scores from multiparametric magnetic resonance images. Proc Natl Acad Sci USA 112(46):E73–E6265Google Scholar
  44. 44.
    Wibmer A, Hricak H, Gondo T, Matsumoto K, Veeraraghavan H, Fehr D et al (2015) Haralick texture analysis of prostate MRI: utility for differentiating non-cancerous prostate from prostate cancer and differentiating prostate cancers with different Gleason scores. Eur Radiol 25(10):50–2840Google Scholar
  45. 45.
    Shiri I, Rahmim A, Ghaffarian P, Geramifar P, Abdollahi H, Bitarafan-Rajabi A (2017) The impact of image reconstruction settings on 18F-FDG PET radiomic features: multi-scanner phantom and patient studies. Eur Radiol 27(11):509–4498Google Scholar
  46. 46.
    Shiri I, Abdollahi H, Shaysteh S, Mahdavi SR (2017) Test-retest reproducibility and robustness analysis of recurrent glioblastoma MRI radiomics texture features. Iran J Radiol.  https://doi.org/10.5812/iranjradiol.48035
  47. 47.
    Saeedi E, Dejkam A, Beigi J, Rastegar S, Yousei Z, Mehdipour A et al (2018) Radiomic feature robustness and reproducibility in quantitative bone radiography: a study on radiologic parameter changes. J Clin Densitom.  https://doi.org/10.1016/j.jocd.2018.06.004

Copyright information

© Italian Society of Medical Radiology 2019

Authors and Affiliations

  1. 1.Department of Medical Physics, School of MedicineIran University of Medical SciencesTehranIran
  2. 2.Shohada-e-Tajrish Medical CenterShahid Beheshti University of Medical SciencesTehranIran
  3. 3.Biomedical and Health Informatics, Rajaie Cardiovascular Medical and Research CenterIran University of Medical SciencesTehranIran
  4. 4.Research Center for Molecular and Cellular ImagingTehran University of Medical SciencesTehranIran
  5. 5.Urology and Nephrology Research Center, Student Research CommitteeShahid Beheshti University of Medical SciencesTehranIran
  6. 6.Department of Radiology, Modarres HospitalShahid Beheshti University of Medical SciencesTehranIran
  7. 7.Faculty of Computer Science and Engineering, Image Processing and Distributed System LabShahid Beheshti UniversityTehranIran
  8. 8.Radiation Biology Research CenterIran University of Medical SciencesTehranIran

Personalised recommendations