Prostate cancer incidence and diagnosis in men with PSA levels >20 ng/ml: is it possible to decrease the number of biopsy cores?

Comparison of machine learning models based on multi-parametric magnetic resonance imaging and ultrasound videos for the prediction of prostate cancer

Objective

To establish machine learning (ML) prediction models for prostate cancer (PCa) using transrectal ultrasound videos and multi-parametric magnetic resonance imaging (mpMRI) and compare their diagnostic performance.

Materials and methods

We systematically collated the data of 383 patients, including 187 with PCa and 196 with benign lesions. Of them, 307 patients (150 with PCa and 157 with benign lesions) were randomly selected to train and validate the ML models, 76 patients were used as test set. B-Ultrasound videos (BUS), mpMRI T2 sequence (T2), and ADC sequence (ADC) were obtained from all patients. We extracted 851 features of each patient in the BUS, T2, and ADC groups and used a t-test, the Mann-Whitney U test, and LASSO regression to screen the features. Support vector machine (SVM), random forest (RF), adaptive boosting (ADB), and gradient boosting machine (GBM) models were used to establish radiomics models. In addition, we fused the features screened via LASSO regression from three groups as new features and rebuilt ML models. The performance of the ML models in diagnosing PCa in the BUS, T2, ADC, and fusion groups was compared using the area under the ROC curve (AUC), sensitivity, specificity, and accuracy.

Results

In the test cohort, the AUC of each model in the ADC group was higher than that of in the.BUS and T2 groups. Among the models, the RF model had the best diagnostic performance, with an AUC of 0.85, sensitivity of 0.78 (0.61-0.89), specificity of 0.84 (0.69-0.94), and accuracy of 0.83 (0.66-0.93). The SVM model in both the BUS and T2 groups performed best. Based on the features screened in the BUS, T2, and ADC groups fused to construct the models, the SVM model was found to perform best, with an AUC of 0.87, sensitivity of 0.73 (0.56-0.86), specificity of 0.79 (0.63-0.90), and accuracy of 0.77 (0.59-0.89). The difference in the results was statistically significant (p<0.05).

Conclusion

The ML prediction models had a good diagnostic ability for PCa. Among them, the SVM model in the fusion group showed the best performance in diagnosing PCa. These prediction models can help radiologists make better diagnoses.

How can we predict the active surveillance candidates meeting all Epstein criteria prior to prostate biopsy to avoid overdiagnosis?

OBJECTIVE

To investigate the effectiveness of PSA, prostate volume (PV) and free-to-total PSA ratio (fPSA%) in predicting patients meeting all active surveillance criteria, including Epstein criteria.

METHOD

Retrospective analysis was made of the data of 1901 men who underwent transrectal ultrasound (TRUS)-guided prostate biopsy in our clinic between January 2015 and December 2019. The active surveillance criteria were determined as Gleason score ≤6, when specified ≤2 positive cores with <50% cancer involvement in every positive core, a clinical T1c, a PSA <10ng/mL and a PSA density <0.15 ng/mL/cc. Patients who met all active surveillance criteria were included in Group 1, and other patients with prostate cancer were included in Group 2.

RESULTS

The study included 336 patients with available data of age, total-free PSA levels, PV calculated by TRUS. Group 1 consisted of 82 patients and Group 2 consisted of 254 patients. PV and fPSA% were significantly higher and PSA was significantly lower in Group 1 than in Group 2. On multivariate analysis, the independent predictors were determined to be PSA and PV while fPSA% was not.

CONCLUSION

By using PSA and PV in predicting patients meeting all active surveillance criteria, unnecessary biopsies and ultimately overdiagnosis can be reduced.

Use of Prostate Systematic and Targeted Biopsy on the Basis of Bi-Parametric Magnetic Resonance Imaging in Biopsy-Naïve Patients

OBJECTIVES

To explore the performance of targeted biopsy (TB) in combination with systematic biopsy (SB) in the detection of prostate cancer (PCa) in biopsy naïve patients.

METHODS

From May 2018 to January 2020, 230 biopsy-naïve men with suspicious bi-parametric MRI [bpMRI; Prostate Imaging Reporting and Data System (PI-RADS) score ≥3] were enrolled. All patients had prostate-specific antigen (PSA) levels of 20 ng/ml or less. For each patient, transrectal ultrasound-guided prostate biopsy was performed. The primary endpoint was the detection rate of CSPC [clinically-significant PCa, International Society of Urological Pathology grade group (ISUP GG) 2 or higher tumors]. The secondary endpoints were the detection rates of CIPC (clinically insignificant PCa, ISUP GG 1 tumors).

RESULTS

CSPC was detected in 90 patients. Twelve (13.33%) of them were detected by TB only and 18 (20.00%) by SB only. Detection of CSPC by SB and TB did not differ significantly (p = .36). In 4.35% of 230 patients, CSPC would have been missed if we performed SB only, and in 6.09% of patients if we performed TB only. Moreover, combination of TB and SB did not increase the detection of CIPC.

CONCLUSIONS

No significant difference was found in the detection of CSPC between TB and SB; however, both techniques revealed substantial added value and combination of TB and SB could further improve this detection rate without increasing the detection of CIPC.