Diagnosing prostate cancer in the PSA gray zone through machine learning and transrectal ultrasound video

Background: We developed a machine learning-based predictive model for diagnosing prostate cancer within the gray zone of prostate-specific antigen (PSA) levels, leveraging transrectal prostate ultrasound video clips. Methods: Data were collected for patients with suspected prostate cancer, characterized by intermediate PSA levels between 4 and 10 ng/mL, who visited the Department of Urology, Dongyang People’s Hospital, which is affiliated with Wenzhou Medical University, from 20 August 2021 to 30 September 2023. Among the final selection of 508 patients, a total of 851 features were extracted from the ultrasound video clips, reduced the dimensionality using least absolute shrinkage and selection operator regression, and finally selected 25 features. The selected features were employed to construct radiomics models based on four machine learning algorithms support vector machine (SVM), random forest (RF), adaptive boosting (ADB) and gradient boosting machine (GBM). The performance of the model was comprehensively assessed using receiver operating characteristic (ROC) curve analysis, with diagnostic effectiveness measured through metrics such as the area under the curve (AUC), sensitivity, specificity and overall accuracy. Results: The RF model demonstrated an AUC of 0.89, accuracy of 0.81, sensitivity of 0.81, specificity of 0.79, positive predictive value of 0.91 and F1 score of 0.77. As compared to the RF model, the SVM, ADB and GBM models showed similar values for AUC (range 0.80–0.86), accuracy (range 0.75–0.79), sensitivity (range 0.80–0.81), specificity (range 0.65–0.75), positive predictive value (range 0.83–0.89) and F1 score (range 0.72–0.76). In the validation set, following comprehensive evaluation, the RF model exhibited the best performance among the four models. Conclusions: The four machine learning models each had diagnostic value for detecting prostate cancer in patients within the PSA “gray zone”, with the RF model demonstrating the highest predictive performance.

Prostate cancer; PSA; Gray zone; Machine Learning; Ultrasound

[1] Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians. 2021; 71: 209–249.

[2] Bergengren O, Pekala KR, Matsoukas K, Fainberg J, Mungovan SF, Bratt O, et al. 2022 update on prostate cancer epidemiology and risk factors—a systematic review. European Urology. 2023; 84: 191–206.

[3] Kachuri L, Hoffmann TJ, Jiang Y, Berndt SI, Shelley JP, Schaffer KR, et al. Genetically adjusted PSA levels for prostate cancer screening. Nature Medicine. 2023; 29: 1412–1423.

[4] Schaeffer EM, Srinivas S, Adra N, An Y, Barocas D, Bitting R, et al. Prostate cancer, Version 4.2023, NCCN clinical practice guidelines in oncology. Journal of the National Comprehensive Cancer Network. 2023; 21: 1067–1096.

[5] Ouyang M, Jia M, Chang Z, Wang Y, Wang K, Gao X, et al. Precise prostate cancer diagnosis using fluorescent nanoprobes for detecting PSA and PSMA in serum. Chemical Communications. 2024; 60: 5181–5184.

[6] Yang J, Li J, Xiao L, Zhou M, Fang Z, Cai Y, et al. 68Ga-PSMA PET/CT-based multivariate model for highly accurate and noninvasive diagnosis of clinically significant prostate cancer in the PSA gray zone. Cancer Imaging. 2023; 23: 81.

[7] Qian Y, Shi Y, Sheng X, Liao H, Chen H, Shi B, et al. Evaluating the role of morphological parameters in the prostate transition zone in PHI-Based predictive models for detecting gray zone prostate cancer. Clinical Medicine Insights: Oncology. 2023; 17: 11795549231201122.

[8] Siddiqui MR, Li EV, Kumar SKSR, Busza A, Lin JS, Mahenthiran AK, et al. Optimizing detection of clinically significant prostate cancer through nomograms incorporating MRI, clinical features, and advanced serum biomarkers in biopsy naïve men. Prostate Cancer and Prostatic Diseases. 2023; 26: 588–595.

[9] Thomas J, Atluri S, Zucker I, Reis I, Kwon D, Kim E, et al. A multi-institutional study of 1111 men with 4K score, multiparametric magnetic resonance imaging, and prostate biopsy. Urologic Oncology. 2023; 41: 430.e9–430.e16.

[10] Sachpekidis C, Pan L, Groezinger M, Strauss DS, Dimitrakopoulou-Strauss A. Combined whole-body dynamic and static PET/CT with low-dose [18F] PSMA-1007 in prostate cancer patients. European Journal of Nuclear Medicine and Molecular Imaging. 2024; 51: 2137–2150.

[11] Hamm CA, Baumgärtner GL, Biessmann F, Beetz NL, Hartenstein A, Savic LJ, et al. Interactive explainable deep learning model informs prostate cancer diagnosis at MRI. Radiology. 2023; 307: e222276.

[12] Chen G, Dai X, Zhang M, Tian Z, Jin X, Mei K, et al. Machine learning-based prediction model and visual interpretation for prostate cancer. BMC Urology. 2023; 23: 164.

[13] van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Research. 2017; 77: e104–107.

[14] Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. Journal of Machine Learning Research. 2011; 12: 2825–2830.

[15] Alonzo DG, Mure AL, Soloway MS. Prostate cancer and the increasing role of active surveillance. Postgraduate Medicine. 2013; 125: 109–116.

[16] van Griethuysen JJM, Fedorov A, Parmar C, Hosny A, Aucoin N, Narayan V, et al. Computational radiomics system to decode the radiographic phenotype. Cancer Research. 2017; 77: e104–e107.

[17] Kaneko M, Sugano D, Lebastchi AH, Duddalwar V, Nabhani J, Haiman C, et al. Techniques and outcomes of MRI-TRUS fusion prostate biopsy. Current Urology Reports. 2021; 22: 27.

[18] Merriel SWD, Pocock L, Gilbert E, Creavin S, Walter FM, Spencer A, et al. Systematic review and meta-analysis of the diagnostic accuracy of prostate-specific antigen (PSA) for the detection of prostate cancer in symptomatic patients. BMC Medicine. 2022; 20: 54.

[19] Van den Broeck T, van den Bergh RCN, Briers E, Cornford P, Cumberbatch M, Tilki D, et al. Biochemical recurrence in prostate cancer: the European association of urology prostate cancer guidelines panel recommendations. European Urology Focus. 2020; 6: 231–234.

[20] Smeenge M, Barentsz J, Cosgrove D, de la Rosette J, de Reijke T, Eggener S, et al. Role of transrectal ultrasonography (TRUS) in focal therapy of prostate cancer: report from a Consensus Panel. BJU International. 2012; 110: 942–948.

[21] Tomczak K, Czerwińska P, Wiznerowicz M. The cancer genome atlas (TCGA): An immeasurable source of knowledge. Contemporary Oncology. 2015; 19: A68–A77.

[22] Smith DS, Humphrey PA, Catalona WJ. The early detection of prostate carcinoma with prostate specific antigen: the Washington University experience. Cancer. 1997; 80: 1852–1856.

[23] Ignatiadis M, Sledge GW, Jeffrey SS. Liquid biopsy enters the clinic—implementation issues and future challenges. Nature Reviews Clinical Oncology. 2021; 18: 297–312.

[24] Wang Y, Wang Z, Gang X, Wang G. Liquid biopsy in prostate cancer: current status and future challenges of clinical application. The Aging Male. 2021; 24: 58–71.

[25] Urabe F, Sumiyoshi T, Tashiro K, Goto T, Kimura T, Kobayashi T. Prostate cancer and liquid biopsies: clinical applications and challenges. International Journal of Urology. 2024; 31: 617–626.

[26] Kim CJ, Dong L, Amend SR, Cho YK, Pienta KJ. The role of liquid biopsies in prostate cancer management. Lab on a Chip. 2021; 21: 3263–3288.

[27] Qin B, Liang Y, Liang Y, Zhang J, Wu Q, Ji H, et al. Comparative study on the diagnosis of gray zone prostate cancer by transperineal and transrectal prostate biopsy guided by transrectal ultrasound. Chinese Journal of Andrology. 2016; 22: 477–479. (In Chinese)

[28] Liu J, Wang Z, Li M, Zhou M, Yu Y, Zhan W. Establishment of two new predictive models for prostate cancer to determine whether to require prostate biopsy when the PSA level is in the diagnostic gray zone (4–10 ng ml−1). Asian Journal of Andrology. 2020; 22: 213–216.

[29] Spohn SKB, Bettermann AS, Bamberg F, Benndorf M, Mix M, Nicolay NH, et al. Radiomics in prostate cancer imaging for a personalized treatment approach-current aspects of methodology and a systematic review on validated studies. Theranostics. 2021; 11: 8027–8042.

[30] Mayerhoefer ME, Materka A, Langs G, Häggström I, Szczypiński P, Gibbs P, et al. Introduction to radiomics. The Journal of Nuclear Medicine. 2020; 61: 488–495.

[31] Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, van Stiphout RG, Granton P, et al. Radiomics: extracting more information from medical images using advanced feature analysis. European Journal of Cancer. 2012; 48: 441–446.

[32] Monti S, Brancato V, Di Costanzo G, Basso L, Puglia M, Ragozzino A, et al. Multiparametric MRI for prostate cancer detection: new insights into the combined use of a radiomic approach with advanced acquisition protocol. Cancers. 2020; 12: 390.

[33] Abboud B, Smayra T, Jabbour H, Ghorra C, Abadjian G. Correlations of neck ultrasound and pathology in cervical lymph node of papillary thyroid carcinoma. Acta Chirurgica Belgica. 2020; 120: 238–244.

[34] Zhang H, Li X, Zhang Y, Huang C, Wang Y, Yang P, et al. Diagnostic nomogram based on intralesional and perilesional radiomics features and clinical factors of clinically significant prostate cancer. Journal of Magnetic Resonance Imaging. 2021; 53: 1550–1558.

[35] Xian M, Zheng X, Xu J, Li X, Chen L, Wang W. Prediction of lymph node metastasis in rectal cancer: comparison between shear-wave elastography based ultrasomics and MRI. Diagnostic and Interventional Radiology. 2021; 27: 424–431.

[36] Jin X, Ai Y, Zhang J, Zhu H, Jin J, Teng Y, et al. Noninvasive prediction of lymph node status for patients with early-stage cervical cancer based on radiomics features from ultrasound images. European Radiology. 2020; 30: 4117–4124.

[37] Wang K, Chen P, Feng B, Tu J, Hu Z, Zhang M, et al. Machine learning prediction of prostate cancer from transrectal ultrasound video clips. Frontiers in Oncology. 2022; 12: 948662.

[38] Li M, Chen T, Zhao W, Wei C, Li X, Duan S, et al. Radiomics prediction model for the improved diagnosis of clinically significant prostate cancer on biparametric MRI. Quantitative Imaging in Medicine and Surgery. 2020; 10: 368–379.

[39] Qi X, Wang K, Feng B, Sun X, Yang J, Hu Z, et al. Comparison of machine learning models based on multi-parametric magnetic resonance imaging and ultrasound videos for the prediction of prostate cancer. Frontiers in Oncology. 2023; 13: 1157949.

[40] Nketiah G, Elschot M, Kim E, Teruel JR, Scheenen TW, Bathen TF, et al. T2-weighted MRI-derived textural features reflect prostate cancer aggressiveness: preliminary results. European Radiology. 2017; 27: 3050–3059.

[41] Chen M, Ma T, Li J, Zhang H, Li Q, Wang J, et al. Diagnosis of prostate cancer in patients with prostate-specific antigen (PSA) in the gray area: construction of 2 predictive models. Medical Science Monitor. 2021; 27: e929913.

[42] Liu T, Zhang X, Chen R, Deng X, Fu B. Development, comparison, and validation of four intelligent, practical machine learning models for patients with prostate-specific antigen in the gray zone. Frontiers in Oncology. 2023; 13: 1157384.

[43] Soeterik TFW, van Melick HHE, Dijksman LM, Biesma DH, Witjes JA, van Basten JA. Active surveillance for prostate cancer in a real-life cohort: comparing outcomes for PRIAS-eligible and PRIAS-ineligible patients. European Urology Oncology. 2018; 1: 231–237.