MRI-Based Radiomics Nomogram for Predicting Prostate Cancer with Gray-Zone Prostate-Specific Antigen Levels to Reduce Unnecessary Biopsies
Abstract
:1. Introduction
2. Materials and Methods
2.1. Patients
2.2. MRI Protocol
2.3. Reference Standard
2.4. Clinical Data
2.5. Radiomic Analysis: Segmentation and Extraction
2.6. Model Construction
2.7. Clinical Usefulness
2.8. Statistical Analysis
3. Results
3.1. Patient Characteristics
3.2. Clinical Model
3.3. Radiomics Model
3.4. Nomogram Model
3.5. Calibration
3.6. Clinical Use
4. Discussion
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Siegel, R.L.; Miller, K.D.; Fuchs, H.E.; Jemal, A. Cancer Statistics, 2021. CA Cancer J. Clin. 2021, 71, 7–33. [Google Scholar] [CrossRef]
- Roobol, M.J.; Kranse, R.; Bangma, C.H.; van Leenders, A.G.; Blijenberg, B.G.; van Schaik, R.H.; Kirkels, W.J.; Otto, S.J.; van der Kwast, T.H.; de Koning, H.J.; et al. Screening for prostate cancer: Results of the Rotterdam section of the European randomized study of screening for prostate cancer. Eur. Urol. 2013, 64, 530–539. [Google Scholar] [CrossRef]
- Chen, R.; Huang, Y.R.; Cai, X.B.; Xie, L.P.; He, D.L.; Zhou, L.Q.; Xu, C.L.; Gao, X.; Ren, S.C.; Wang, F.B.; et al. Age-Specific Cutoff Value for the Application of Percent Free Prostate-Specific Antigen (PSA) in Chinese Men with Serum PSA Levels of 4.0–10.0 ng/mL. PLoS ONE 2015, 10, e0130308. [Google Scholar] [CrossRef]
- Tang, P.; Du, W.; Xie, K.J.; Deng, X.R.; Fu, J.G.; Chen, H.; Yang, W.J. Transition zone PSA density improves the prostate cancer detection rate both in PSA 4.0-10.0 and 10.1-20.0 ng/ml in Chinese men. Urol. Oncol. 2013, 31, 744–748. [Google Scholar] [CrossRef]
- Thompson, I.M.; Ankerst, D.P.; Chi, C.; Lucia, M.S.; Goodman, P.J.; Crowley, J.J.; Parnes, H.L.; Coltman, C.A.J. Operating characteristics of prostate-specific antigen in men with an initial PSA level of 3.0 ng/ml or lower. JAMA 2005, 294, 66–70. [Google Scholar] [CrossRef]
- Park, T.Y.; Chae, J.Y.; Kim, J.W.; Kim, J.W.; Oh, M.M.; Yoon, C.Y.; Moon, D.G. Prostate-specific antigen mass and free prostate-specific antigen mass for predicting the prostate volume of korean men with biopsy-proven benign prostatic hyperplasia. Korean J. Urol. 2013, 54, 609–614. [Google Scholar] [CrossRef] [Green Version]
- Rocco, B.; Grasso, A.; Sosnowski, R.; Dell’orto, P.G.; Albo, G.; Castle, E.; Coelho, R.; Patel, V.; Mottrie, A. PSA mass screening: Is there enough evidence? Cent. Eur. J. Urol. 2012, 65, 4–6. [Google Scholar] [CrossRef] [Green Version]
- Huang, P.; Chen, Y.H.; Chen, S.H.; Li, X.D.; Chen, D.N.; Lin, T.T.; Wei, Y.; Zheng, Q.S.; Xu, N.; Xue, X.Y. Influence of prostatic calculi on the results of prostate biopsy in patients with a PSA level of 4–10 μg/L. Natl. J. Androl. 2021, 27, 718–724. [Google Scholar] [CrossRef]
- Sandhu, J.S. Management of elevated prostate-specific antigen in men with nonbacterial chronic prostatitis. Curr. Urol. Rep. 2009, 10, 302–306. [Google Scholar] [CrossRef]
- Wang, R.; Wang, J.; Gao, G.; Hu, J.; Jiang, Y.Y.; Zhao, Z.L.; Zhang, X.D.; Zhang, Y.D.; Wang, X.Y. Prebiopsy mp-MRI Can Help to Improve the Predictive Performance in Prostate Cancer: A Prospective Study in 1,478 Consecutive Patients. Clin. Cancer Res. 2017, 23, 3692–3699. [Google Scholar] [CrossRef]
- MacAskill, F.; Lee, S.M.; Eldred-Evans, D.; Wulaningsih, W.; Popert, R.; Wolfe, K.; Van Hemelrijck, M.; Rottenberg, G.; Liyanage, S.H.; Acher, P. Diagnostic value of MRI-based PSA density in predicting transperineal sector-guided prostate biopsy outcomes. Int. Urol. Nephrol. 2017, 49, 1335–1342. [Google Scholar] [CrossRef] [Green Version]
- Xu, N.; Wu, Y.P.; Chen, D.N.; Ke, Z.B.; Cai, H.; Wei, Y.; Zheng, Q.S.; Huang, J.B.; Li, X.D.; Xue, X.Y. Can Prostate Imaging Reporting and Data System Version 2 reduce unnecessary prostate biopsies in men with PSA levels of 4-10 ng/ml? J. Cancer Res. Clin. Oncol. 2018, 144, 987–995. [Google Scholar] [CrossRef]
- Liu, C.; Liu, S.L.; Wang, Z.X.; Yu, K.; Feng, C.X.; Ke, Z.; Wang, L.; Zeng, X.Y. Using the prostate imaging reporting and data system version 2 (PI-RIDS v2) to detect prostate cancer can prevent unnecessary biopsies and invasive treatment. Asian J. Androl. 2018, 20, 459–464. [Google Scholar] [CrossRef]
- Qi, Y.F.; Zhang, S.T.; Wei, J.W.; Zhang, G.M.Y.; Lei, J.; Yan, W.G.; Xiao, Y.; Yan, S.; Xue, H.D.; Feng, F.; et al. Multiparametric MRI-Based Radiomics for Prostate Cancer Screening with PSA in 4-10 ng/mL to Reduce Unnecessary Biopsies. J. Magn. Reson. Imaging 2020, 51, 1890–1899. [Google Scholar] [CrossRef]
- Mussi, T.C.; Yamauchi, F.I.; Tridente, C.F.; Tachibana, A.; Tonso, V.M.; Recchimuzzi, D.R.; de Souza Leão, L.R.; Luz, D.C.; Martins, T.; Baroni, R.H. Interobserver Agreement and Positivity of PI-RADS Version 2 Among Radiologists with Different Levels of Experience. Acad. Radiol. 2019, 26, 1017–1022. [Google Scholar] [CrossRef]
- Aerts, H.J.; Velazquez, E.R.; Leijenaar, R.T.; Parmar, C.; Grossmann, P.; Carvalho, S.; Bussink, J.; Monshouwer, R.; Haibe-Kains, B.; Rietveld, D.; et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat. Commun. 2014, 5, 4006. [Google Scholar] [CrossRef] [Green Version]
- Gillies, R.J.; Kinahan, P.E.; Hricak, H. Radiomics: Images Are More than Pictures, They Are Data. Radiology 2016, 278, 563–577. [Google Scholar] [CrossRef] [Green Version]
- Gong, L.X.; Xu, M.; Fang, M.J.; Zou, J.; Yang, S.D.; Yu, X.Y.; Xu, D.D.; Zhou, L.J.; Li, H.L.; He, B.X.; et al. Noninvasive Prediction of High-Grade Prostate Cancer via Biparametric MRI Radiomics. J. Magn. Reson. Imaging 2020, 52, 1102–1109. [Google Scholar] [CrossRef]
- Chen, T.; Li, M.J.; Gu, Y.F.; Zhang, Y.Y.; Yang, S.; Wei, C.G.; Wu, J.F.; Li, X.; Zhao, W.L.; Shen, J.K. Prostate Cancer Differentiation and Aggressiveness: Assessment with a Radiomic-Based Model vs. PI-RADS v2. J. Magn. Reson. Imaging 2019, 49, 875–884. [Google Scholar] [CrossRef] [Green Version]
- Luo, W.Q.; Huang, Q.X.; Huang, X.W.; Hu, H.T.; Zeng, F.Q.; Wang, W. Predicting Breast Cancer in Breast Imaging Reporting and Data System (BI-RADS) Ultrasound Category 4 or 5 Lesions: A Nomogram Combining Radiomics and BI-RADS. Sci. Rep. 2019, 9, 11921. [Google Scholar] [CrossRef]
- Niu, X.K.; Li, J.; Das, S.K.; Xiong, Y.; Yang, C.B.; Peng, T. Developing a nomogram based on multiparametric magnetic resonance imaging for forecasting high-grade prostate cancer to reduce unnecessary biopsies within the prostate-specific antigen gray zone. BMC Med. Imaging 2017, 17, 11. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Chawla, N.; Bowyer, K.; Hall, L.O.; Kegelmeyer, W.P. SMOTE: Synthetic minority over-sampling technique. J. Art. Intell. Res. 2002, 16, 321–357. [Google Scholar] [CrossRef]
- Coutant, C.; Olivier, C.; Lambaudie, E.; Fondrinier, E.; Marchal, F.; Guillemin, F.; Seince, N.; Thomas, V.; Levêque, J.; Barranger, E.; et al. Comparison of models to predict nonsentinel lymph node status in breast cancer patients with metastatic sentinel lymph nodes: A prospective multicenter study. J. Clin. Oncol. 2009, 27, 2800–2808. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Vickers, A.J.; Cronin, A.M.; Elkin, E.B.; Gonen, M. Extensions to decision curve analysis, a novel method for evaluating diagnostic tests, prediction models and molecular markers. BMC Med. Inform. Decis. Mak. 2008, 8, 53. [Google Scholar] [CrossRef] [Green Version]
- DeLong, E.R.; DeLong, D.M.; Clarke-Pearson, D.L. Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach. Biometrics 1988, 44, 837–845. [Google Scholar] [CrossRef] [PubMed]
- Wang, Z.B.; Wei, C.G.; Zhang, Y.Y.; Pan, P.; Dai, G.C.; Tu, J.; Shen, J.K. The Role of PSA Density among PI-RADS v2.1 Categories to Avoid an Unnecessary Transition Zone Biopsy in Patients with PSA 4-20 ng/mL. BioMed Res. Int. 2021, 1, 3995789. [Google Scholar] [CrossRef]
- Sun, J.L.; Zhang, Z.Y.; OuYang, J. A novel nomogram combined PIRADS v2 and neutrophil-to-lymphocyte ratio to predict the risk of clinically significant prostate cancer in men with PSA < 10 ng/ml at first biopsy. Urol. Oncol. 2020, 38, 401–409. [Google Scholar] [CrossRef]
- Parekh, V.; Jacobs, M.A. Radiomics: A new application from established techniques. Expert Rev. Precis. Med. Drug. Dev. 2016, 1, 207–226. [Google Scholar] [CrossRef] [Green Version]
- Wu, P.Q.; Liu, Z.Y.; He, L.; Huang, Y.Q.; Liang, C.H. The status of combining radiomics with big data. Chin. J. Radiol. 2017, 51, 554–558. [Google Scholar] [CrossRef]
- Li, M.j.; Chen, T.; Zhao, W.l.; Wei, C.g.; Li, X.B.; Duan, S.F.; Ji, L.B.; Lu, Z.H.; Shen, J.K. Radiomics prediction model for the improved diagnosis of clinically significant prostate cancer on biparametric MRI. Quant. Imaging Med. Surg. 2020, 10, 368–379. [Google Scholar] [CrossRef]
- Fang, D.; Ren, D.; Zhao, C.L.; Li, X.S.; Yu, W.; Wang, R.; Wang, H.H.; Xi, C.G.; He, Q.; Wang, X.Y.; et al. Prevalence and Risk Factors of Prostate Cancer in Chinese Men with PSA 4-10 ng/mL Who Underwent TRUS-Guided Prostate Biopsy: The Utilization of PAMD Score. BioMed Res. Int. 2015, 1, 596797. [Google Scholar] [CrossRef] [Green Version]
- Xiao, M.L.; Ma, F.H.; Li, Y.; Li, Y.A.; Li, M.D.; Zhang, G.F.; Qiang, J.W. Multiparametric MRI-Based Radiomics Nomogram for Predicting Lymph Node Metastasis in Early-Stage Cervical Cancer. J. Magn. Reson. Imaging. 2020, 52, 885–896. [Google Scholar] [CrossRef]
- Liang, L.; Zhi, X.; Sun, Y.; Li, H.R.; Wang, J.J.; Xu, J.X.; Guo, J. A Nomogram Based on a Multiparametric Ultrasound Radiomics Model for Discrimination Between Malignant and Benign Prostate Lesions. Front. Oncol. 2021, 11, 610785. [Google Scholar] [CrossRef]
- Fiorentino, V.; Martini, M.; Dell’Aquila, M.; Musarra, T.; Orticelli, E.; Larocca, L.M.; Rossi, E.; Totaro, A.; Pinto, F.; Lenci, N.; et al. Histopathological Ratios to Predict Gleason Score Agreement between Biopsy and Radical Prostatectomy. Diagnostics 2020, 11, 10. [Google Scholar] [CrossRef] [PubMed]
Characteristics | PCa (n = 90) | Non-PCa (n = 174) | p | Training Cohort (n = 191) | Validation Cohort (n = 83) | p |
---|---|---|---|---|---|---|
Age (y) | 75.0 [53.0, 98.0] | 69.5 [47.0, 93.0] | 0.015 | 71.0 [47.0, 98.0] | 74.0 [57.0, 93.0] | 0.360 |
Volume (mL) | 39.0 [14.5, 138] | 60.5 [2.70, 432] | <0.001 | 54.1 [14.2, 432] | 53.6 [2.70, 186] | 0.484 |
tPSA | 7.66 [4.00, 10.7] | 6.71 [3.26, 9.95] | 0.005 | 6.98 [3.26, 10.7] | 7.35 [4.21, 9.95] | 0.749 |
fPSA | 1.19 [0.01, 6.06] | 1.18 [0.10, 5.13] | 0.979 | 1.20 [0.01, 6.06] | 1.14 [0.20, 3.12] | 0.512 |
f/tPSA | 0.161 [0.0025, 0.652] | 0.180 [0.025, 0.523] | 0.157 | 0.177 [0.0025, 0.652] | 0.167 [0.0274, 0.382] | 0.339 |
PSAD | 0.193 [0.0474, 0.513] | 0.113 [0.0095, 2.93] | 0.037 | 0.129 [0.0095, 0.513] | 0.129 [0.0095, 0.513] | 0.173 |
PI-RADS v2.1 score | <0.001 | 0.993 | ||||
1 | 5.00 (5.6%) | 52.0 (28.3%) | 41.0 (21.5%) | 16.0 (19.3%) | ||
2 | 7.00 (7.8%) | 82.0 (44.6%) | 58.0 (30.4%) | 31.0 (37.3%) | ||
3 | 13.0 (14.4%) | 29.0 (15.8%) | 30.0 (15.7%) | 12.0 (14.5%) | ||
4 | 39.0 (43.3%) | 19.0 (10.3%) | 41.0 (21.5%) | 17.0 (20.5%) | ||
5 | 26.0 (28.9%) | 2.00 (1.1%) | 21.0 (11.0%) | 7.00 (8.4%) |
Predictor | Univariate Analysis | Multiple Analysis | ||||
---|---|---|---|---|---|---|
β | OR (95%CI) | p Value | β | OR (95%CI) | p Value | |
Age (y) | 0.0748 | 2.455 [1.501, 4.015] | <0.001 | 0.0774 | 2.533 [1.378, 4.656] | 0.001 |
Volume (mL) | −0.0186 | 0.486 [0.308, 0.768] | 0.002 | NA | NA | 0.486 |
tPSA | 0.2307 | 1.871 [1.445, 3.058] | 0.013 | NA | NA | 0.225 |
fPSA | 0.0283 | 1.012 [0.749, 1.393] | 0.893 | NA | NA | NA |
f/tPSA | −2.0633 | 0.803 [0.552, 1.165] | 0.247 | NA | NA | NA |
PSAD | 8.1946 | 2.956 [1.568, 3.362] | <0.001 | NA | NA | 0.714 |
PI-RADS v2.1 score | 1.2241 | 11.569 [5.811, 23.032] | <0.001 | 1.2439 | 12.034 [5.847, 24.77] | <0.001 |
Training Cohort | Validation Cohort | |||||
---|---|---|---|---|---|---|
Clinical Model | Radiomics Model | Clinical-Radiomics Combined Model | Clinical Model | Radiomics Model | Clinical-Radiomics Combined Model | |
AUC (95%CI) | 0.868 [0.813, 0.922] | 0.982 [0.964, 0.999] | 0.984 [0.968, 1.000] | 0.866 [0.783, 0.950] | 0.941 [0.888, 0.995] | 0.953 [0.907, 0.999] |
Sensitivity (95%CI) | 0.750 [0.630–0.841] | 0.953 [0.865, 0.985] | 0.953 [0.865, 0.985] | 0.846 [0.655, 0.941] | 0.808 [0.613, 0.918] | 0.885 [0.697, 0.962] |
Specificity (95%CI) | 0.827 [0.751–0.883] | 0.961 [0.909, 0.984] | 0.984 [0.939, 0.996] | 0.842 [0.724, 0.916] | 0.965 [0.870,0.991] | 0.930 [0.827, 0.973] |
PPV | 0.686 [0.562, 0.789] | 0.924 [0.825, 0.972] | 0.968 [0.880, 0.994] | 0.710 [0.518, 0.851] | 0.913 [0.705, 0.985] | 0.852 [0.654, 0.951] |
NPV | 0.868 [0.791, 0.920] | 0.976 [0.926, 0.994] | 0.977 [0.928, 0.994] | 0.923 [0.806, 0.975] | 0.917 [0.809, 0.969] | 0.946 [0.842, 0.986] |
LR+ | 4.330 [2.886, 6.494] | 24.209 [10.236, 57.257] | 60.523 [15.287, 239.619] | 5.359 [2.878, 9.977] | 23.019 [5.825, 90.973] | 12.606 [4.850, 32.762] |
LR− | 0.302 [0.197, 0.464] | 0.049 [0.016, 0.147] | 0.048 [0.016, 0.144] | 0.183 [0.073, 0.453] | 0.199 [0.091, 0.439] | 0.124 [0.043, 0.361] |
Z | −4.391 | −0.952 | NA | −2.154 | −1.1227 | NA |
p value | <0.001 | 0.341 | NA | 0.031 | 0.262 | NA |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, L.; Zhang, J.; Tang, M.; Lei, X.-Y.; Li, L.-C. MRI-Based Radiomics Nomogram for Predicting Prostate Cancer with Gray-Zone Prostate-Specific Antigen Levels to Reduce Unnecessary Biopsies. Diagnostics 2022, 12, 3005. https://doi.org/10.3390/diagnostics12123005
Zhang L, Zhang J, Tang M, Lei X-Y, Li L-C. MRI-Based Radiomics Nomogram for Predicting Prostate Cancer with Gray-Zone Prostate-Specific Antigen Levels to Reduce Unnecessary Biopsies. Diagnostics. 2022; 12(12):3005. https://doi.org/10.3390/diagnostics12123005
Chicago/Turabian StyleZhang, Li, Jing Zhang, Min Tang, Xiao-Yan Lei, and Long-Chao Li. 2022. "MRI-Based Radiomics Nomogram for Predicting Prostate Cancer with Gray-Zone Prostate-Specific Antigen Levels to Reduce Unnecessary Biopsies" Diagnostics 12, no. 12: 3005. https://doi.org/10.3390/diagnostics12123005