Quantitative parameters from dynamic contrast-enhanced MRI in detecting prostate cancer

Congress:

ECR 2018

Poster Number:

C-1200

Type:

Scientific Exhibit

Keywords:

Oncology, Genital / Reproductive system male, MR physics, MR-Diffusion/Perfusion, MR-Functional imaging, MR-Angiography, Contrast agent-other, Image compression, Cancer, Neoplasia

Authors:

Y. Zhang, C. Wei, M. J. Szewczyk-Bieda, S. Lang, G. Nabi; Dundee/UK

DOI:

10.1594/ecr2018/C-1200

DOI-Link:

https://dx.doi.org/10.1594/ecr2018/C-1200

Fig. 1: Twelve regions segmentation method

Fig. 2: The ROC curves of quantitative DCE parameters in Tofts&Kromode model and the...

Fig. 3: The ROC curves of quantitative DCE parameters in Extended Tofts model and the...

Fig. 4: boxplot of Ve_TK and cancer grade

Aims and objectives

Prostate cancer (Pca) is the most common cancer for males and the second death rate caused by cancer in men [1]. 26% male cancer patients are diagnosed with prostate cancer (220800 of 848200) and making up 9% (27540 of 312150), which was the second grade of mortality for male in 2015[2]. In the European counties, incident of prostate cancer ranks the top of malignant tumours. The aim of this study was to assess the diagnostic accuracy of quantitative parameters fromDynamic contrast enhanced MRI (DCE-MRI) compartmental...

Methods and materials

•12 regions segmentation of prostate gland was established (Fig.1) and 348 regions (29*12) were segmented in the 29 patients who had proctectomy. •DCE-MRI quantitative parameters in two kinetic models (Tofts & Kermode, Extended Tofts) were analyzed using advanced MRI post-processing software Olea Sphere (Olea Medical, La Ciotat, France). •Prostate specimens which were detailly analysed by experienced pathologist were used as gold reference for validation of the cohort. •Roc curve and boxplot were mainly approached to analyse the cut-off point for quantitative parameters to distinguish benign...

Results

1.The cut-off value to differentiate benign and prostate cancer tissue were Kep (1.18 min-1; sensitivity: 55%, specificity: 68%) and Ve(0.24 min-1; sensitivity: 62%, specificity: 61%) in the Tofts&Kermode model(TK) (Fig.2), Ktrans(0.145 min-1; sensitivity:67%, specificity:57%) in extended Tofts model(ET) (Fig.3). 2.The quantitative parameters above cut-off value were analysed correlation with pathological results. Quantitative parameter Ve_TK (Fig.4) and Ktrans_ET (Fig.6) had strong positive correlation with Gleason Score. Medium value for Ve_TK with Gleason Score 6, 7 and >7 were 0.17 min-1, 0.32 min-1, 0.41 min-1 respectively and...

Conclusion

DCE-MRI can quantitatively detect prostate cancer using different models and quantitative parameters. Secondly, positive correlation was detected between different grades of cancer and quantitative parameters.

Personal information

Yilong Zhang, MSc Biomedical Engineering, School of Science and Engineering, University of Dundee, Dundee, DD1 4HN, UK E-mail :[email protected]

References

1.Jemal,A.et al. Global cancer statistics. CA Cancer J.Clin.61,67-90(2011) 2. Siegel, R.L., et al., Cancer statistics for Hispanics/Latinos, 2015. CA Cancer J Clin, 2015. 65(6): p. 457-480