Manual prostate MRI segmentation by readers with different experience: a study of the learning progress.

OBJECTIVE: To evaluate the learning progress of less experienced readers in prostate MRI segmentation. MATERIALS AND METHODS: One hundred bi-parametric prostate MRI scans were retrospectively selected from the Göteborg Prostate Cancer Screening 2 Trial (single center). Nine readers with varying degrees of segmentation experience were involved: one expert radiologist, two experienced radiology residents, two inexperienced radiology residents, and four novices. The task was to segment the whole prostate gland. The expert's segmentations were used as reference. For all other readers except three novices, the 100 MRI scans were divided into five rounds (cases 1-10, 11-25, 26-50, 51-76, 76-100). Three novices segmented only 50 cases (three rounds). After each round, a one-on-one feedback session between the expert and the reader was held, with feedback on systematic errors and potential improvements for the next round. Dice similarity coefficient (DSC) > 0.8 was considered accurate. RESULTS: Using DSC > 0.8 as the threshold, the novices had a total of 194 accurate segmentations out of 250 (77.6%). The residents had a total of 397/400 (99.2%) accurate segmentations. In round 1, the novices had 19/40 (47.5%) accurate segmentations, in round 2 41/60 (68.3%), and in round 3 84/100 (84.0%) indicating learning progress. CONCLUSIONS: Radiology residents, regardless of prior experience, showed high segmentation accuracy. Novices showed larger interindividual variation and lower segmentation accuracy than radiology residents. To prepare datasets for artificial intelligence (AI) development, employing radiology residents seems safe and provides a good balance between cost-effectiveness and segmentation accuracy. Employing novices should only be considered on an individual basis. CLINICAL RELEVANCE STATEMENT: Employing radiology residents for prostate MRI segmentation seems safe and can potentially reduce the workload of expert radiologists. Employing novices should only be considered on an individual basis. KEY POINTS: • Using less experienced readers for prostate MRI segmentation is cost-effective but may reduce quality. • Radiology residents provided high accuracy segmentations while novices showed large inter-reader variability. • To prepare datasets for AI development, employing radiology residents seems safe and might provide a good balance between cost-effectiveness and segmentation accuracy while novices should only be employed on an individual basis.

[Prostate cancer - diagnostics and screening]. / Prostatacancer ­ utredning, klinisk diagnostik och screening.

Prostate-specific antigen (PSA) based screening is controversial, even though randomised trials show that screening can reduce prostate cancer mortality. The main reason is that screening leads to overdiagnosis of indolent cancers that would never have surfaced clinically in the absence of screening. Recently, several large studies have shown that magnetic resonance imaging (MRI) improves prostate cancer diagnostics. With MRI, up to half of all men with elevated PSA values can be spared a biopsy. When a biopsy is needed, the needles can be directed towards the suspicious area in the prostate, which increases the detection of clinically significant tumors. In Sweden, regional programmes with organised prostate cancer testing were introduced in 2020. These programmes aim to make prostate cancer testing more standardized, efficient, and equitable. In the future, biomarkers and AI-based systems will likely be important to further improve prostate cancer diagnostics.

Prostate Cancers in the Prostate-specific Antigen Interval of 1.8-3 ng/ml: Results from the Göteborg-2 Prostate Cancer Screening Trial.

BACKGROUND AND OBJECTIVE: Magnetic resonance imaging(MRI) and targeted biopsies reduce overdiagnosis of prostate cancer(PC). It is uncertain how this strategy performs for low PSA levels. The objective was to investigate the PI-RADS distribution, frequency and characteristics of screen-detected PC with PSA of 1.8-<3ng/ml and 3-<10ng/ml. METHODS: In the population-based Göteborg-2 screening study, 17 974 men choose to participate by having a PSA(2015-2020). One-third of the participants(n=6006) were randomized to Arm 3, men with a PSA ≥1.8ng/ml were recommended MRI. Men with positive MRI(PI-RADS3-5) had four targeted biopsies from each MRI-visible lesion. Clinically significant PC was defined as Gleason score ≥3+4. KEY FINDINGS AND LIMITATIONS: 6006 men were included. Median age was 55.9 years. 670(11%) had PSA of 1.8-<3ng/ml(low-PSA group), median PSA 2.1ng/ml, and 377(6.3%) had PSA of 3-<10ng/ml(high-PSA group), median PSA 3.9ng/ml. PI-RADS scores of 3, 4, and 5 were observed in 7.8%, 15%, and 1.0% in the low-PSA group, and in 6.9%, 17%, and 5.3% in the high-PSA group, respectively. PC was found in 64 men (41%, 95%CI 0.33-0.49) with positive MRI findings in the low-PSA group, 33(21%) had Gleason 6 and 31(20%) had Gleason ≥7. In the high-PSA group, PC was detected in 61men (56%, 95%CI 0.46-0.66), 26(24%) had Gleason 6 and 35(32%) had Gleason ≥7. Limitations include results from only a single screening round. CONCLUSIONS AND CLINICAL IMPLICATIONS: A non-negligible number of men with PSA 1.8-3ng/ml have clinically significant PC. Whether a delay in the diagnosis of these tumors until they reached PSA ≥3ng/ml would impair their chance of cure remains to be evaluated.

Performance of 4Kscore as a Reflex Test to Prostate-specific Antigen in the GÖTEBORG-2 Prostate Cancer Screening Trial.

BACKGROUND AND OBJECTIVE: We investigated whether adding 4Kscore as a reflex test to prostate-specific antigen (PSA) could improve the screening algorithm for prostate cancer (PC). METHODS: In the GÖTEBORG-2 PC screening trial, 38 000men (50-60 yr) were invited to PSA testing and, if elevated, followed by magnetic resonance imaging (MRI). For 571 men with PSA ≥3.0 ng/ml and evaluable outcomes, 4Kscore was calculated. The performance using a prespecified 4Kscore cutoff of 7.5% was evaluated. KEY FINDINGS AND LIMITATIONS: The area under the curve for 4Kscore to identify intermediate- and high-risk PC was 0.84 (95% confidence interval 0.79-0.89), and the positive predictive value, and negative predictive value were 15% (0.12-0.20) and 99% (97-100%), respectively. Of the 54 men diagnosed with intermediate- or high-grade PC, two had a 4Kscore cutoff below 7.5%, both with organ-confined intermediate-risk PC. Per 1000 men with elevated PSA, adding 4Kscore would have resulted in avoidance of MRI for 408 (41%) men, biopsies for 95 (28% reduction) men, and diagnosis of 23 low-grade cancers (23% reduction) while delaying the diagnosis of four men with intermediate-grade cancers (4%). CONCLUSIONS AND CLINICAL IMPLICATIONS: Including 4Kscore as a reflex test for men with elevated PSA reduces the need for MRI and biopsy markedly, and results in less overdiagnosis of low-grade PC at the cost of delaying the diagnosis of intermediate-grade PC in a few men. These results add further evidence for including new blood-based biomarkers in addition to PSA to improve the harm and benefit ratio of PC screening and reduce the need for resource-demanding MRI and biopsies. PATIENT SUMMARY: In this study, 4Kscore, a blood-based biomarker, as a reflex test for men with elevated prostate-specific antigen (PSA), reduces the need for magnetic resonance imaging and biopsy. These results support the inclusion of new blood-based biomarkers in addition to PSA.

How a population-based cohort of men estimate lifetime risk of prostate cancer in a survey before entering a prostate cancer screening trial in Sweden?

OBJECTIVES: Investigating men's perceived lifetime risk of prostate cancer. DESIGN: Survey-based study to men invited for prostate-specific antigen (PSA) screening in the GÖTEBORG-2 trial between September 2015 and June 2020. SETTING: 38 775 men in the Gothenburg area, Sweden, were invited for PSA-testing and participated in a survey. PARTICIPANTS: 17 980 men participated in PSA-testing, of whom 13 189 completed the survey. In addition, 1264 men answered the survey only. INTERVENTIONS: Before having the PSA-test, men answered an electronic survey and estimated their lifetime risk of receiving a prostate cancer diagnosis on a visual analogue scale from 0% to 100%. MAIN OUTCOME MEASURES: The primary outcome was the median lifetime risk estimation, which was compared with Wilcoxon test to an anticipated lifetime risk of 20% (based on GÖTEBORG-1 trial). The secondary outcome was to determine factors associated with risk estimation in a multivariable linear regression model: previous prostate examination, family history, physical exercise, healthy diet, comorbidity, alcohol consumption, smoking, education level, marital status, urinary symptoms and erectile dysfunction. RESULTS: Among PSA-tested men, the median estimated lifetime risk of prostate cancer was 30% (IQR 19% to 50%), corresponding to a 10 percentage-points higher estimation compared with the anticipated risk (p<0.001). Family history of prostate cancer, moderate to severe urinary symptoms and mild to moderate erectile dysfunction were associated with >5 percentage-points higher risk estimation. Similar results were obtained for non-PSA-tested men. CONCLUSIONS: Most men overestimated their prostate cancer risk which underscores the importance of providing them accurate information about prostate cancer. TRIAL REGISTRATION NUMBER: ISRCTN94604465.