Multiparametric Prostate MRI in Biopsy-Naïve Men: A Prospective Evaluation of Performance and Biopsy Strategies

Krüger-Stokke, Brage; Bertilsson, Helena; Langørgen, Sverre; Sjøbakk, Torill Anita Eidhammer; Bathen, Tone Frost; Selnæs, Kirsten Margrete

doi:10.3389/fonc.2021.745657

Cited by 9 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to a study by Natale et al, PSAD is a useful tool that can be consulted to determine whether a biopsy in patients with PI-RADS3 is required. [ 32 ]. Zhang et al also reported that PSAD could improve the detection of PCa, as well as csPCa, for PI-RADSv2 score-3 lesions [ 33 ].…”

Section: Discussionmentioning

confidence: 99%

Clinical Utility of Prostate Health Index for Diagnosis of Prostate Cancer in Patients with PI-RADS 3 Lesions

Lee

Chung

et al. 2022

Cancers

View full text Add to dashboard Cite

The risk of prostate cancer (PCa) in prostate imaging reporting and data system version 2 (PI-RADSv2) score-3 lesions is equivocal; it is regarded as an intermediate status of presented PCa. In this study, we evaluated the clinical utility of the prostate health index (PHI) for the diagnosis of PCa and clinically significant PCa (csPCa) in patients with PI-RADSv2 score-3 lesions. The study cohort included patients who underwent a transrectal ultrasound (TRUS)-guided, cognitive-targeted biopsy for PI-RADSv2 score-3 lesions between November 2018 and April 2021. Before prostate biopsy, the prostate-specific antigen (PSA) derivatives, such as total PSA (tPSA), [-2] proPSA (p2PSA) and free PSA (fPSA) were determined. The calculation equation of PHI is as follows: [(p2PSA/fPSA) × tPSA ½]. Using a receiver operating characteristic (ROC) curve analysis, the values of PSA derivatives measured by the area under the ROC curve (AUC) were compared. For this study, csPCa was defined as Gleason grade 2 or higher. Of the 392 patients with PI-RADSv2 score-3 lesions, PCa was confirmed in 121 (30.9%) patients, including 59 (15.1%) confirmed to have csPCa. Of all the PSA derivatives, PHI and PSA density (PSAD) showed better performance in predicting overall PCa and csPCa, compared with PSA (all p < 0.05). The AUC of the PHI for predicting overall PCa and csPCa were 0.807 (95% confidence interval (CI): 0.710–0.906, p = 0.001) and 0.819 (95% CI: 0.723–0.922, p < 0.001), respectively. By the threshold of 30, PHI was 91.7% sensitive and 46.1% specific for overall PCa, and was 100% sensitive for csPCa. Using 30 as a threshold for PHI, 34.4% of unnecessary biopsies could have been avoided, at the cost of 8.3% of overall PCa, but would include all csPCa.

show abstract

Section: Discussionmentioning

confidence: 99%

Clinical Utility of Prostate Health Index for Diagnosis of Prostate Cancer in Patients with PI-RADS 3 Lesions

Lee

Chung

et al. 2022

Cancers

View full text Add to dashboard Cite

show abstract

“…The in-house collected dataset was obtained from St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway between March 2015 and December 2017 as part of a previous prospective study [ 12 ]. It consists of mpMR images from 233 patients (median age = 65; range: 44–76 years) who were examined due to suspicion of prostate cancer, via the Norwegian standardized care pathway, in which patients with elevated PSA and/or abnormal digital rectal exam results are referred for an initial mpMRI scan to identify suspicious cancerous tissue.…”

Section: Methodsmentioning

confidence: 99%

Label-set impact on deep learning-based prostate segmentation on MRI

Meglič,

Sunoqrot,

Bathen

et al. 2023

Insights Imaging

Self Cite

View full text Add to dashboard Cite

Background Prostate segmentation is an essential step in computer-aided detection and diagnosis systems for prostate cancer. Deep learning (DL)-based methods provide good performance for prostate gland and zones segmentation, but little is known about the impact of manual segmentation (that is, label) selection on their performance. In this work, we investigated these effects by obtaining two different expert label-sets for the PROSTATEx I challenge training dataset (n = 198) and using them, in addition to an in-house dataset (n = 233), to assess the effect on segmentation performance. The automatic segmentation method we used was nnU-Net. Results The selection of training/testing label-set had a significant (p < 0.001) impact on model performance. Furthermore, it was found that model performance was significantly (p < 0.001) higher when the model was trained and tested with the same label-set. Moreover, the results showed that agreement between automatic segmentations was significantly (p < 0.0001) higher than agreement between manual segmentations and that the models were able to outperform the human label-sets used to train them. Conclusions We investigated the impact of label-set selection on the performance of a DL-based prostate segmentation model. We found that the use of different sets of manual prostate gland and zone segmentations has a measurable impact on model performance. Nevertheless, DL-based segmentation appeared to have a greater inter-reader agreement than manual segmentation. More thought should be given to the label-set, with a focus on multicenter manual segmentation and agreement on common procedures. Critical relevance statement Label-set selection significantly impacts the performance of a deep learning-based prostate segmentation model. Models using different label-set showed higher agreement than manual segmentations. Key points • Label-set selection has a significant impact on the performance of automatic segmentation models. • Deep learning-based models demonstrated true learning rather than simply mimicking the label-set. • Automatic segmentation appears to have a greater inter-reader agreement than manual segmentation. Graphical Abstract

show abstract

“…Therefore, many efforts have been made to identify new biomarkers that could improve PCa detection. PSA density (PSAD) is a clinically useful tool to compensate for the low sensitivity of PSA, and recent studies have reported that PSAD is effective in determining whether to perform a prostate biopsy and improves the detection of PCa and clinically significant PCa (csPCa) [7][8][9].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Prostate-Specific Antigen and Its Density and Prostate Health Index and Its Density for Detection of Prostate Cancer

et al. 2023

View full text Add to dashboard Cite

As the incidence of prostate cancer (PCa) has increased, screening based on prostate-specific antigen (PSA) has become controversial due to the low specificity of PSA. Therefore, we investigated the diagnostic performance of prostate health index (PHI) density (PHID) for the detection of PCa and clinically significant PCa (csPCa) compared to PSA, PSA density (PSAD), and PHI as a triaging test. We retrospectively reviewed 306 men who underwent prostate biopsy for PSA levels of 2.5 to 10 ng/mL between January 2020 and April 2023. Of all cohorts, 86 (28.1%) and 48 (15.7%) men were diagnosed with PCa and csPCa, respectively. In ROC analysis, the highest AUC was identified for PHID (0.812), followed by PHI (0.791), PSAD (0.650), and PSA (0.571) for PCa. A similar trend was observed for csPCa: PHID (AUC 0.826), PHI (AUC 0.796), PSAD (AUC 0.671), and PSA (0.552). When the biopsy was restricted to men with a PHID ≥ 0.56, 26.5% of unnecessary biopsies could be avoided; however, 9.3% of PCa cases and one csPCa case (2.1%) remained undiagnosed. At approximately 90% sensitivity for csPCa, at the given cut-off values of PHI ≥ 36.4, and PHID ≥ 0.91, 48.7% and 49.3% of unnecessary biopsies could be avoided. In conclusion, PHID had a small advantage over PHI, about 3.6%, for the reduction in unnecessary biopsies for PCa. The PHID and PHI showed almost the same diagnostic performance for csPCa detection. PHID can be used as a triaging test in a clinical setting to pre-select the risk of PCa and csPCa.

show abstract

Multiparametric Prostate MRI in Biopsy-Naïve Men: A Prospective Evaluation of Performance and Biopsy Strategies

Cited by 9 publications

References 29 publications

Clinical Utility of Prostate Health Index for Diagnosis of Prostate Cancer in Patients with PI-RADS 3 Lesions

Clinical Utility of Prostate Health Index for Diagnosis of Prostate Cancer in Patients with PI-RADS 3 Lesions

Label-set impact on deep learning-based prostate segmentation on MRI

Comparison of Prostate-Specific Antigen and Its Density and Prostate Health Index and Its Density for Detection of Prostate Cancer

Contact Info

Product

Resources

About