Multimodal Radiomic Features for the Predicting Gleason Score of Prostate Cancer

Ahmad Chaddad, Michael J Kucharczyk, Tamim Niazi, Ahmad Chaddad, Michael J Kucharczyk, Tamim Niazi

Abstract

Background: Novel radiomic features are enabling the extraction of biological data from routine sequences of MRI images. This study's purpose was to establish a new model, based on the joint intensity matrix (JIM), to predict the Gleason score (GS) of prostate cancer (PCa) patients.

Methods: A retrospective dataset comprised of the diagnostic imaging data of 99 PCa patients was used, extracted from The Cancer Imaging Archive's (TCIA) T2-Weighted (T2-WI) and apparent diffusion coefficient (ADC) images. Radiomic features derived from JIM and the grey level co-occurrence matrix (GLCM) were extracted from the reported tumor locations. The Kruskal-Wallis test and Spearman's rank correlation identified features related to the GS. The Random Forest classifier model was implemented to identify the best performing signature of JIM and GLCM radiomic features to predict for GS.

Results: Five JIM-derived features: contrast, homogeneity, difference variance, dissimilarity, and inverse difference were independent predictors of GS (p < 0.05). Combined JIM and GLCM analysis provided the best performing area-under-the-curve, with values of 78.40% for GS ≤ 6, 82.35% for GS = 3 + 4, and 64.76% for GS ≥ 4 + 3.

Conclusion: This retrospective study produced a novel predictive model for GS by the incorporation of JIM data from standard diagnostic MRI images.

Keywords: Gleason score; biomarkers; prostate cancer; radiomics.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Figure 1
Figure 1
Schema of the utilized methodology for a GS prediction model for prostate cancer. (A) T2-WI and ADC MRI images of 99 patients with biopsy-confirmed and MRI-localized PCa are extracted; (B) Co-occurrence matrices (i.e., JIM and GLCMs) were computed from ROI’s which corresponded to biopsy-proven sites of PCa; (C) The 19 features encoded within the GLCMs and JIM are extracted; (D) Statistical analyses performed uni- and multivariate analyses of these features to generate a model predictive of the GS.
Figure 2
Figure 2
Univariate prediction of GS using texture features (A) Kruskal-Wallis significance test compared the three GS groupings. (B) Spearman’s rank correlation coefficient between texture features and the GS groupings. Significant features (p < 0.05) are indicated with a black-green circle. (C) Average of the histogram features derived from T2-WI (i.e., first row) and ADC (i.e., second row) image value (Histogram features: Mean, Variance, Skewness, Kurtosis, Energy, and Entropy) across the three groups of GS (i.e., G1, G2, and G3).
Figure 3
Figure 3
Multivariate analysis of texture features and RF classifier model. The AUC for predicting G1 (GS ≤ 6), G2 (GS = 3+4), and G3 (GS ≥ 4 + 3) using the JIM (19 features), GLCMADC (19 features), GLCMT2-WI (19 features), all combined features (GLCMs + JIM, 19 × 3 features), and all of the standard features (histogram features derived from ADC and T2-WI images of PCa, 6 × 2 features). For training (n = 40 samples) and testing (n = 20 samples), balance samples were considered.
Figure 4
Figure 4
Importance of individual features for predicting GS groups using the RF classifier. Positive and negative values correspond respectively to predictive and non-predictive features.

References

    1. Lindenberg M.L., Turkbey B., Mena E., Choyke P.L. Imaging Locally Advanced, Recurrent, and Metastatic Prostate Cancer: A Review. JAMA Oncol. 2017;3:1415–1422. doi: 10.1001/jamaoncol.2016.5840.
    1. Fitzmaurice C., Allen C., Barber R.M., Barregard L., Bhutta Z.A., Brenner H., Dicker D.J., Chimed-Orchir O., Dandona R., Dandona L., et al. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncol. 2018 doi: 10.1001/jamaoncol.2018.2706.
    1. Canadian Cancer Statistics Publication—Canadian Cancer Society. [(accessed on 22 May 2018)]; Available online: .
    1. Pinsky P.F., Parnes H.L., Andriole G. Mortality and complications after prostate biopsy in the P rostate, L ung, C olorectal and O varian C ancer S creening (PLCO) trial. BJU Int. 2014;113:254–259. doi: 10.1111/bju.12368.
    1. Epstein J.I. Prostate cancer grading: A decade after the 2005 modified system. Mod. Pathol. 2018;31:S47–S63. doi: 10.1038/modpathol.2017.133.
    1. Truong M., Frye T., Messing E., Miyamoto H. Historical and contemporary perspectives on cribriform morphology in prostate cancer. Nat. Rev. Urol. 2018 doi: 10.1038/s41585-018-0013-1.
    1. D’Amico A.V., Moul J., Carroll P.R., Sun L., Lubeck D., Chen M.-H. Cancer-specific mortality after surgery or radiation for patients with clinically localized prostate cancer managed during the prostate-specific antigen era. J. Clin. Oncol. Off. J. Am. Soc. Clin. Oncol. 2003;21:2163–2172. doi: 10.1200/JCO.2003.01.075.
    1. Andriole G.L., Crawford E.D., Grubb R.L.I., Buys S.S., Chia D., Church T.R., Fouad M.N., Gelmann E.P., Kvale P.A., Reding D.J., et al. Mortality Results from a Randomized Prostate-Cancer Screening Trial. N. Engl. J. Med. 2009;360:1310–1319. doi: 10.1056/NEJMoa0810696.
    1. Schröder F.H., Hugosson J., Roobol M.J., Tammela T.L.J., Ciatto S., Nelen V., Kwiatkowski M., Lujan M., Lilja H., Zappa M., et al. Screening and Prostate-Cancer Mortality in a Randomized European Study. N. Engl. J. Med. 2009;360:1320–1328. doi: 10.1056/NEJMoa0810084.
    1. Heidegger I., Skradski V., Steiner E., Klocker H., Pichler R., Pircher A., Horninger W., Bektic J. High risk of under-grading and -staging in prostate cancer patients eligible for active surveillance. PLoS ONE. 2015;10:e0115537. doi: 10.1371/journal.pone.0115537.
    1. Griffiths D.F.R., Melia J., McWilliam L.J., Ball R.Y., Grigor K., Harnden P., Jarmulowicz M., Montironi R., Moseley R., Waller M., et al. A study of Gleason score interpretation in different groups of UK pathologists; techniques for improving reproducibility. Histopathology. 2006;48:655–662. doi: 10.1111/j.1365-2559.2006.02394.x.
    1. Harbias A., Salmo E., Crump A. Implications of Observer Variation in Gleason Scoring of Prostate Cancer on Clinical Management: A Collaborative Audit. Gulf J. Oncol. 2017;1:41–45.
    1. Nakai Y., Tanaka N., Shimada K., Konishi N., Miyake M., Anai S., Fujimoto K. Review by urological pathologists improves the accuracy of Gleason grading by general pathologists. BMC Urol. 2015;15:70. doi: 10.1186/s12894-015-0066-x.
    1. Chen Y.-J., Chu W.-C., Pu Y.-S., Chueh S.-C., Shun C.-T., Tseng W.-Y.I. Washout gradient in dynamic contrast-enhanced MRI is associated with tumor aggressiveness of prostate cancer. J. Magn. Reson. Imaging JMRI. 2012;36:912–919. doi: 10.1002/jmri.23723.
    1. Vos E.K., Litjens G.J.S., Kobus T., Hambrock T., Hulsbergen-van de Kaa C.A., Barentsz J.O., Huisman H.J., Scheenen T.W.J. Assessment of prostate cancer aggressiveness using dynamic contrast-enhanced magnetic resonance imaging at 3 T. Eur. Urol. 2013;64:448–455. doi: 10.1016/j.eururo.2013.05.045.
    1. Mehralivand S., Shih J.H., Rais-Bahrami S., Oto A., Bednarova S., Nix J.W., Thomas J.V., Gordetsky J.B., Gaur S., Harmon S.A., et al. A Magnetic Resonance Imaging–Based Prediction Model for Prostate Biopsy Risk Stratification. JAMA Oncol. 2018;4:678–685. doi: 10.1001/jamaoncol.2017.5667.
    1. Hassanzadeh E., Glazer D.I., Dunne R.M., Fennessy F.M., Harisinghani M.G., Tempany C.M. Prostate Imaging Reporting and Data System Version 2 (PI-RADS v2): A pictorial review. Abdom. Radiol. N.Y. 2017;42:278–289. doi: 10.1007/s00261-016-0871-z.
    1. Donati O.F., Afaq A., Vargas H.A., Mazaheri Y., Zheng J., Moskowitz C.S., Hricak H., Akin O. Prostate MRI: Evaluating tumor volume and apparent diffusion coefficient as surrogate biomarkers for predicting tumor Gleason score. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 2014;20:3705–3711. doi: 10.1158/1078-0432.CCR-14-0044.
    1. Hambrock T., Somford D.M., Huisman H.J., van Oort I.M., Witjes J.A., Hulsbergen-van de Kaa C.A., Scheenen T., Barentsz J.O. Relationship between apparent diffusion coefficients at 3.0-T MR imaging and Gleason grade in peripheral zone prostate cancer. Radiology. 2011;259:453–461. doi: 10.1148/radiol.11091409.
    1. Wang Q., Li H., Yan X., Wu C.-J., Liu X.-S., Shi H.-B., Zhang Y.-D. Histogram analysis of diffusion kurtosis magnetic resonance imaging in differentiation of pathologic Gleason grade of prostate cancer. Urol. Oncol. 2015;33:337.e15–337.e24. doi: 10.1016/j.urolonc.2015.05.005.
    1. Tiwari P., Kurhanewicz J., Madabhushi A. Multi-kernel graph embedding for detection, Gleason grading of prostate cancer via MRI/MRS. Med. Image Anal. 2013;17:219–235. doi: 10.1016/j.media.2012.10.004.
    1. Vargas H.A., Akin O., Franiel T., Mazaheri Y., Zheng J., Moskowitz C., Udo K., Eastham J., Hricak H. Diffusion-weighted Endorectal MR Imaging at 3 T for Prostate Cancer: Tumor Detection and Assessment of Aggressiveness. Radiology. 2011;259:775–784. doi: 10.1148/radiol.11102066.
    1. Wibmer A., Hricak H., Gondo T., Matsumoto K., Veeraraghavan H., Fehr D., Zheng J., Goldman D., Moskowitz C., Fine S.W., et al. 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. 2015;25:2840–2850. doi: 10.1007/s00330-015-3701-8.
    1. Li J., Weng Z., Xu H., Zhang Z., Miao H., Chen W., Liu Z., Zhang X., Wang M., Xu X., et al. Support Vector Machines (SVM) classification of prostate cancer Gleason score in central gland using multiparametric magnetic resonance images: A cross-validated study. Eur. J. Radiol. 2018;98:61–67. doi: 10.1016/j.ejrad.2017.11.001.
    1. Haralick R.M., Shanmugam K., Dinstein I. Textural Features for Image Classification. IEEE Trans. Syst. Man Cybern. 1973;SMC-3:610–621. doi: 10.1109/TSMC.1973.4309314.
    1. Litjens G., Debats O., Barentsz J., Karssemeijer N., Huisman H. Computer-aided detection of prostate cancer in MRI. IEEE Trans. Med. Imaging. 2014;33:1083–1092. doi: 10.1109/TMI.2014.2303821.
    1. Chaddad A., Daniel P., Desrosiers C., Toews M., Abdulkarim B. Novel Radiomic Features based on Joint Intensity Matrices for Predicting Glioblastoma Patient Survival Time. IEEE J. Biomed. Health Inform. 2018 doi: 10.1109/JBHI.2018.2825027.
    1. Haralick R.M. Statistical and structural approaches to texture. Proc. IEEE. 1979;67:786–804. doi: 10.1109/PROC.1979.11328.
    1. Zar J.H. Significance Testing of the Spearman Rank Correlation Coefficient. J. Am. Stat. Assoc. 1972;67:578–580. doi: 10.1080/01621459.1972.10481251.
    1. Holm S. A Simple Sequentially Rejective Multiple Test Procedure. Scand. J. Stat. 1979;6:65–70.
    1. Archer K.J., Kimes R.V. Empirical characterization of random forest variable importance measures. Comput. Stat. Data Anal. 2008;52:2249–2260. doi: 10.1016/j.csda.2007.08.015.
    1. Breiman L. Random Forests. Mach. Learn. 2001;45:5–32. doi: 10.1023/A:1010933404324.
    1. Oshiro T.M., Perez P.S., Baranauskas J.A. Machine Learning and Data Mining in Pattern Recognition. Springer; Berlin, Heidelberg: 2012. How Many Trees in a Random Forest? pp. 154–168. Lecture Notes in Computer Science.
    1. Vignati A., Mazzetti S., Giannini V., Russo F., Bollito E., Porpiglia F., Stasi M., Regge D. Texture features on T2-weighted magnetic resonance imaging: New potential biomarkers for prostate cancer aggressiveness. Phys. Med. Biol. 2015;60:2685–2701. doi: 10.1088/0031-9155/60/7/2685.
    1. Aerts H.J.W.L. The Potential of Radiomic-Based Phenotyping in Precision Medicine: A Review. JAMA Oncol. 2016;2:1636–1642. doi: 10.1001/jamaoncol.2016.2631.
    1. Mankoff D.A., Farwell M.D., Clark A.S., Pryma D.A. Making Molecular Imaging a Clinical Tool for Precision Oncology: A Review. JAMA Oncol. 2017;3:695–701. doi: 10.1001/jamaoncol.2016.5084.
    1. Siddiqui M.M., Rais-Bahrami S., Turkbey B., George A.K., Rothwax J., Shakir N., Okoro C., Raskolnikov D., Parnes H.L., Linehan W.M., et al. Comparison of MR/Ultrasound Fusion–Guided Biopsy With Ultrasound-Guided Biopsy for the Diagnosis of Prostate Cancer. JAMA. 2015;313:390–397. doi: 10.1001/jama.2014.17942.
    1. Chaddad A., Daniel P., Niazi T. Radiomics Evaluation of Histological Heterogeneity Using Multiscale Textures Derived From 3D Wavelet Transformation of Multispectral Images. Front. Oncol. 2018;8 doi: 10.3389/fonc.2018.00096.
    1. Chaddad A., Sabri S., Niazi T., Abdulkarim B. Prediction of survival with multi-scale radiomic analysis in glioblastoma patients. Med. Biol. Eng. Comput. 2018;1–14:1–14. doi: 10.1007/s11517-018-1858-4.
    1. Chaddad A., Desrosiers C., Toews M., Abdulkarim B., Chaddad A., Desrosiers C., Toews M., Abdulkarim B. Predicting survival time of lung cancer patients using radiomic analysis. Oncotarget. 2017;8:104393–104407. doi: 10.18632/oncotarget.22251.
    1. Viswanath S.E., Bloch N.B., Chappelow J.C., Toth R., Rofsky N.M., Genega E.M., Lenkinski R.E., Madabhushi A. Central gland and peripheral zone prostate tumors have significantly different quantitative imaging signatures on 3 Tesla endorectal, in vivo T2-weighted MR imagery. J. Magn. Reson. Imaging JMRI. 2012;36:213–224. doi: 10.1002/jmri.23618.
    1. Moradi M., Salcudean S.E., Chang S.D., Jones E.C., Buchan N., Casey R.G., Goldenberg S.L., Kozlowski P. Multiparametric MRI maps for detection and grading of dominant prostate tumors. J. Magn. Reson. Imaging JMRI. 2012;35:1403–1413. doi: 10.1002/jmri.23540.
    1. Madabhushi A., Feldman M.D., Metaxas D.N., Tomaszeweski J., Chute D. Automated detection of prostatic adenocarcinoma from high-resolution ex vivo MRI. IEEE Trans. Med. Imaging. 2005;24:1611–1625. doi: 10.1109/TMI.2005.859208.
    1. Niaf E., Rouvière O., Mège-Lechevallier F., Bratan F., Lartizien C. Computer-aided diagnosis of prostate cancer in the peripheral zone using multiparametric MRI. Phys. Med. Biol. 2012;57:3833–3851. doi: 10.1088/0031-9155/57/12/3833.
    1. Fehr D., Veeraraghavan H., Wibmer A., Gondo T., Matsumoto K., Vargas H.A., Sala E., Hricak H., Deasy J.O. Automatic classification of prostate cancer Gleason scores from multiparametric magnetic resonance images. Proc. Natl. Acad. Sci. USA. 2015;112:E6265–6273. doi: 10.1073/pnas.1505935112.
    1. Chatterjee A., Bourne R.M., Wang S., Devaraj A., Gallan A.J., Antic T., Karczmar G.S., Oto A. Diagnosis of Prostate Cancer with Noninvasive Estimation of Prostate Tissue Composition by Using Hybrid Multidimensional MR Imaging: A Feasibility Study. Radiology. 2018;171130 doi: 10.1148/radiol.2018171130.

Source: PubMed

Sponsorzy i współpracownicy

Warunki medyczne

Interwencje lekowe

3
Subskrybuj