Skip to main content
SpringerLink
Log in

Zone-specific logistic regression models improve classification of prostate cancer on multi-parametric MRI

  • Magnetic Resonance
  • Published:
European Radiology Aims and scope Submit manuscript

Abstract

Objectives

To assess the interchangeability of zone-specific (peripheral-zone (PZ) and transition-zone (TZ)) multiparametric-MRI (mp-MRI) logistic-regression (LR) models for classification of prostate cancer.

Methods

Two hundred and thirty-one patients (70 TZ training-cohort; 76 PZ training-cohort; 85 TZ temporal validation-cohort) underwent mp-MRI and transperineal-template-prostate-mapping biopsy. PZ and TZ uni/multi-variate mp-MRI LR-models for classification of significant cancer (any cancer-core-length (CCL) with Gleason > 3 + 3 or any grade with CCL ≥ 4 mm) were derived from the respective cohorts and validated within the same zone by leave-one-out analysis. Inter-zonal performance was tested by applying TZ models to the PZ training-cohort and vice-versa. Classification performance of TZ models for TZ cancer was further assessed in the TZ validation-cohort. ROC area-under-curve (ROC-AUC) analysis was used to compare models.

Results

The univariate parameters with the best classification performance were the normalised T2 signal (T2nSI) within the TZ (ROC-AUC = 0.77) and normalized early contrast-enhanced T1 signal (DCE-nSI) within the PZ (ROC-AUC = 0.79). Performance was not significantly improved by bi-variate/tri-variate modelling. PZ models that contained DCE-nSI performed poorly in classification of TZ cancer. The TZ model based solely on maximum-enhancement poorly classified PZ cancer.

Conclusion

LR-models dependent on DCE-MRI parameters alone are not interchangable between prostatic zones; however, models based exclusively on T2 and/or ADC are more robust for inter-zonal application.

Key points

• The ADC and T2-nSI of benign/cancer PZ are higher than benign/cancer TZ.

• DCE parameters are significantly different between benign PZ and TZ, but not between cancerous PZ and TZ.

• Diagnostic models containing contrast enhancement parameters have reduced performance when applied across zones.

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
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Abbreviations

PSA:

prostate specific antigen

TRUS:

transrectal ultrasound guided biopsy

CAD:

computer assisted diagnosis

LR:

logistic regression

TPM:

template mapping biopsy

DWI:

diffusion weighted imaging

DCE:

dynamic contrast enhanced

mp:

multi-parametric

SI:

signal intensity

T2-nSI:

normalized T2 signal intensity

DCE-nSI:

early contrast enhanced T1 signal intensity

SoE:

slope of enhancement

ME:

maximum enhancement

Etype:

curve type

AUCtot:

total area under the dynamic contrast enhanced signal intensity time curve

ROC:

receiver operator characteristic

ROI:

region of interest

AUC:

area under curve

References

  1. Prostate cancer: diagnosis and treatment. NICE clinical guideline 175 Issued: January 2014 guidance.nice.org.uk/cg175

  2. Barentsz JO, Richenberg J, Clements R et al (2012) ESUR prostate MR guidelines 2012. Eur Radiol 22:746–757

    Article  PubMed Central  PubMed  Google Scholar 

  3. Dickinson L, Ahmed HU, Allen C et al (2011) Magnetic resonance imaging for the detection, localisation, and characterisation of prostate cancer: recommendations from a European consensus meeting. Eur Urol 59(4):477–494

    Article  PubMed  Google Scholar 

  4. Kirkham AP, Haslam P, Keanie JY et al (2013) Prostate MRI: who, when, and how? Report from a UK consensus meeting. Clin Radiol 68(10):1016–1023

    Article  CAS  PubMed  Google Scholar 

  5. Hoeks CM, Barentsz JO, Hambrock T et al (2011) Prostate cancer: multiparametric MR imaging for detection, localization, and staging. Radiology 261:46–66

    Article  PubMed  Google Scholar 

  6. Ruprecht O, Weisser P, Bodelle B et al (2012) MRI of the prostate: interobserver agreement compared with histopathologic outcome after radical prostatectomy. Eur J Radiol 81(3):456–460

    Article  PubMed  Google Scholar 

  7. Fütterer JJ (2007) MR imaging in local staging of prostate cancer. Eur J Radiol 63:328–334

    Article  PubMed  Google Scholar 

  8. Puech P, Betrouni N, Makni N et al (2009) Computer-assisted diagnosis of prostate cancer using DCE-MRI data: design, implementation and preliminary results. Int J Comput Assist Radiol Surg 4(1):1–10

    Article  PubMed  Google Scholar 

  9. Hambrock T, Vos PC (2013) Hulsbergen-van de Kaa CA, et al. Prostate cancer: computer-aided diagnosis with multiparametric 3-T MR imaging–effect on observer performance. Radiology 266(2):521–530

    Article  PubMed  Google Scholar 

  10. Sung YS, Kwon HJ, Park BW et al (2011) Prostate cancer detection on dynamic contrast-enhanced MRI: computer-aided diagnosis versus single perfusion parameter maps. AJR Am J Roentgenol 197(5):1122–1129

    Article  PubMed  Google Scholar 

  11. Niaf E, Rouvière O, Mège-Lechevallier F et al (2012) Computer-aided diagnosis of prostate cancer in the peripheral zone using multiparametric MRI. Phys Med Biol 57(12):3833–3851

    Article  PubMed  Google Scholar 

  12. Yoshizako T, Wada A, Hayashi T et al (2008) Usefulness of diffusion-weighted imaging and dynamic contrast-enhanced magnetic resonance imaging in the diagnosis of prostate transition-zone cancer. Acta Radiol 49:1207–1213

    Article  CAS  PubMed  Google Scholar 

  13. Hoeks CM, Hambrock T, Yakar D et al (2013) Transition zone prostate cancer: detection and localization with 3-T multiparametric MR imaging. Radiology 266(1):207–217

    Article  PubMed  Google Scholar 

  14. Engelbrecht MR, Huisman HJ, Laheij RJ et al (2003) Discrimination of prostate cancer from normal peripheral zone and central gland tissue by using dynamic contrast-enhanced MR imaging. Radiology 229:248–254

    Article  PubMed  Google Scholar 

  15. Greene DR, Wheeler TM, Egawa S et al (1991) A comparison of the morphological features of cancer arising in the transition zone and in the peripheral zone of the prostate. J Urol 146(4):1069–1076

    CAS  PubMed  Google Scholar 

  16. Akin O, Sala E, Moskowitz CS et al (2006) Transition zone prostate cancers: features, detection, localization, and staging at endorectal MR imaging. Radiology 239(3):784–792

    Article  PubMed  Google Scholar 

  17. Erbersdobler A, Fritz H, Schnöger S et al (2002) Tumour grade, proliferation, apoptosis, microvessel density, p53, and bcl-2 in prostate cancers: differences between tumours located in the transition zone and in the peripheral zone. Eur Urol 41(1):40–46

    Article  CAS  PubMed  Google Scholar 

  18. Buckley DL, Roberts C, Parker GJ et al (2004) Prostate cancer: evaluation of vascular characteristics with dynamic contrast-enhanced T1-weighted MR imaging–initial experience. Radiology 233(3):709–715

    Article  PubMed  Google Scholar 

  19. Langer DL, van der Kwast TH, Evans AJ et al (2009) Prostate cancer detection with multi-parametric MRI: logistic regression analysis of quantitative T2, diffusion-weighted imaging, and dynamic contrast-enhanced MRI. J Magn Reson Imaging 30:327–334

    Article  PubMed  Google Scholar 

  20. Choi YJ, Kim JK, Kim N et al (2007) Functional MR imaging of prostate cancer. Radiographics 27(1):63–75

    Article  PubMed  Google Scholar 

  21. van Niekerk CG, Witjes JA, Barentsz JO et al (2013) Microvascularity in transition zone prostate tumors resembles normal prostatic tissue. Prostate 73(5):467–475

    Article  PubMed  Google Scholar 

  22. Onik G, Barzell W (2008) Transperineal 3D mapping biopsy of the prostate: an essential tool in selecting patients for focal prostate cancer therapy. Urol Oncol 26:506–510

    Article  PubMed  Google Scholar 

  23. Taira AV, Merrick GS, Galbreath RW et al (2010) Performance of transperineal template-guided mapping biopsy in detecting prostate cancer in the initial and repeat biopsy setting. Prostate Cancer Prostatic Dis 13:71–77

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  24. Barzell WE, Melamed MR (2007) Appropriate patient selection in the focal treatment of prostate cancer: the role of transperineal 3-dimensional pathologic mapping of the prostate–a 4-year experience. Urology 70:27–35

    Article  PubMed  Google Scholar 

  25. Ahmed HU, Hu Y, Carter T et al (2011) Characterizing clinically significant prostate cancer using template prostate mapping biopsy. J Urol 186:458–464

    Article  PubMed  Google Scholar 

  26. Zelhof B, Lowry M, Rodrigues G et al (2009) Description of magnetic resonance imaging-derived enhancement variables in pathologically confirmed prostate cancer and normal peripheral zone regions. BJU Int 104:621–627

    Article  PubMed  Google Scholar 

  27. Tofts PS (1997) Modeling tracer kinetics in dynamic Gd-DTPA MR imaging. J Magn Reson Imaging 7(1):91–101

    Article  CAS  PubMed  Google Scholar 

  28. S M Applied logistic regression analysis Sage University Paper Series on Quantitative Applications in the Social Sciences. 1995: 7–106

  29. Fukunaga K, Hummels DM (1989) Leave-one-out procedures for nonparametric error estimates. IEEE Trans Pattern Anal Mach Intell 11:421–423

    Article  Google Scholar 

  30. Ahmed HU, Emberton M, Kepner G et al (2012) A biomedical engineering approach to mitigate the errors of prostate biopsy. Nat Rev Urol 9:227–231

    Article  CAS  PubMed  Google Scholar 

  31. Roy C, Foudi F, Charton J et al (2013) Comparative sensitivities of functional MRI sequences in detection of local recurrence of prostate carcinoma after radical prostatectomy or external-beam radiotherapy. AJR Am J Roentgenol 200:361–368

    Article  Google Scholar 

  32. Turkbey B, Pinto PA, Mani H et al (2010) Prostate cancer: value of multiparametric MR imaging at 3 T for detection–histopathologic correlation. Radiology 255(1):89–99

    Article  PubMed Central  PubMed  Google Scholar 

  33. Kim CK, Park BK, Han JJ et al (2007) Diffusion-weighted imaging of the prostate at 3 T for differentiation of malignant and benign tissue in transition and peripheral zones: preliminary results. J Comput Assist Tomogr 31:449–454

    Article  PubMed  Google Scholar 

  34. Chesnais AL, Niaf E, Bratan F et al (2013) Differentiation of transitional zone prostate cancer from benign hyperplasia nodules: evaluation of discriminant criteria at multiparametric MRI. Clin Radiol 68(6):e323–e330

    Article  CAS  PubMed  Google Scholar 

  35. Oto A, Kayhan A, Jiang Y et al (2010) Prostate cancer: differentiation of central gland cancer from benign prostatic hyperplasia by using diffusion-weighted and dynamic contrast-enhanced MR imaging. Radiology 257:715–723

    Article  PubMed  Google Scholar 

Download references

Acknowledgments

The scientific guarantor of this publication is Shonit Punwani. The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article. This work was undertaken at the NIHR UCLH/UCL Biomedical Research Centre, which received a portion of the funding from the National Institute for Health Research. The views expressed in this publication are those of the authors and not necessarily those of the UK Department of Health.

This work was supported by the CRUK/EPSRC KCL/UCL Comprehensive Cancer Imaging Centre. ND was supported by UK EPSRC grants EP/I018700/1 and EP/H046410/1.

No complex statistical methods were necessary for this paper. Institutional Review Board approval was obtained. Written informed consent was waived by the Institutional Review Board. Methodology: retrospective, experimental, performed at one institution.

Author information

Authors and Affiliations

  1. Centre for Medical Imaging, University College London, Level 3 East, 250 Euston Road, London, NW1 2PG, UK

    Nikolaos Dikaios, Jokha Alkalbani, Harbir Singh Sidhu, Steve Halligan, Stuart Taylor, David Atkinson & Shonit Punwani

  2. Departments of Radiology, University College London Hospital, 235 Euston Road, London, NW1 2BU, UK

    Nikolaos Dikaios, Alex Kirkham, Steve Halligan, Stuart Taylor, David Atkinson & Shonit Punwani

  3. Research Department of Urology, Division of Surgery and Interventional Science, University College London, London, NW1 2PG, UK

    Mohamed Abd-Alazeez, Hashim U. Ahmed & Mark Emberton

  4. Department of Histopathology, University College London Hospital, London, NW1 2PG, UK

    Alex Freeman

  5. Centre for Medical Imaging, University College London, 3rd Floor East Wing, 250 Euston Road, London, NW1 2PG, UK

    Shonit Punwani

Authors
  1. Nikolaos Dikaios
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jokha Alkalbani
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mohamed Abd-Alazeez
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Harbir Singh Sidhu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Alex Kirkham
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Hashim U. Ahmed
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Mark Emberton
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Alex Freeman
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Steve Halligan
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Stuart Taylor
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. David Atkinson
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Shonit Punwani
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Shonit Punwani.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Dikaios, N., Alkalbani, J., Abd-Alazeez, M. et al. Zone-specific logistic regression models improve classification of prostate cancer on multi-parametric MRI. Eur Radiol 25, 2727–2737 (2015). https://doi.org/10.1007/s00330-015-3636-0

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00330-015-3636-0

Keywords

Search

Navigation