Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa
Background: Multiparametric MRI (mpMRI) prostate has become a fundamental part of prostate cancer primary diagnosis, active surveillance, and lesion localization. PI-RADS reporting lexicon is used to report mpMRI prostate examinations and is a risk- adjusted reporting structure designed by the Europ...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English English |
| Published: |
Division of Radiology
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613197689683968 |
|---|---|
| access_status_str | Open Access |
| author | Wegner, Brett |
| author2 | Human, Gercois |
| author_browse | Human, Gercois Wegner, Brett |
| author_facet | Human, Gercois Wegner, Brett |
| author_sort | Wegner, Brett |
| collection | Thesis |
| description | Background: Multiparametric MRI (mpMRI) prostate has become a fundamental part of prostate cancer primary diagnosis, active surveillance, and lesion localization. PI-RADS reporting lexicon is used to report mpMRI prostate examinations and is a risk- adjusted reporting structure designed by the European Society of Urogenital Radiology in combination with the American College of Radiology with the main aim to internationally standardized mpMRI prostate reporting into five risk categories, of which a higher PI-RADS category represents an increasing probability that the prostate pathology represents clinically significant cancer. Groote Schuur Hospital (GSH) radiology division has been reporting prostate mpMRI according to the updated PI-RADS v2.1 since its introduction in 2019, however no studies have been performed to assess the reliability of PI-RADS v2.1 at GSH for detecting clinically significant prostate cancer. This study aims to assess and validate the accurate use of PI-RADS v2.1 category 4 and 5 against internationally recognized standards of performance. The PI-RADS lexicon should be a reliable predictor of significant prostate cancer to promote confidence in its continued use and to justify the expenditure on mpMRI prostate examinations in the diagnostic pathway of prostate cancer at GSH. Objectives: The primary objective: To assess the diagnostic performance of PI-RADS v2.1 category 4 and 5 in predicting clinically significant prostate cancer confirmed on biopsy with appropriate histopathology. Secondary objective: - Correlate and validate the diagnostic performance of PI-RADS v2.1 category 4 and 5 lesions at GSH by comparing the outcomes with accepted international performance standards. - Identify problems with the use of PI-RADS v2.1 at GSH to improve prostate reporting with the ultimately goal of improving patient outcomes. Methods: Single institution retrospective study assessing mpMRI prostate PI-RADS (v2.1) category 4 and 5 reports performed at GSH between the dates of 01 June 2021 to 12 January 2024. Prostate reports categorised as PIRADS (v2.1) 4 or 5 are evaluated on PACS and correlated with the histological diagnosis on National Health Laboratory Service (NHLS). Prostate histology will be categorised into clinically significant prostate cancer or insignificant depending on the Gleason score. These outcomes will be measured against high powered international studies to assess the overall performance of the GSH radiology division when reporting PI-RADS 4 and 5 lesions. Results: This study incorporates a retrospective design structure and analysis where a grand total of 215 mpMRI prostate studies were considered for the period dated 01 June 2021 and 12 January 2024. From the 215 mpMRI prostate scans performed 89 studies met the strict inclusion and exclusion criteria: n=89. The sample size of 89 represented 37 PI-RADS category 4 and 52 PI-RADS category 5. The measures of sensitivity and specificity for the PI-RADS 4 and PI-RADS 5 categories in detecting clinically significant prostate cancer proven on biopsy are as follows: For PI-RADS 4: Sensitivity: 19.4, 95% CI: [5.4, 33.3] Specificity: 53.4, 95% CI: [40.6, 66.3] PPV : 16.2 95% CI:4.3-28.1 NPV: 51.9 95% CI:38.3-65.5 For PI-RADS 5: Sensitivity: 80.6, 95% CI: [66.7, 94.6] Specificity: 46.6, 95% CI: [33.7, 59.4] PPV: 48.1 95% CI:34.5- 61.7 NPV: 83.8 95% CI:71.9- 95.7 Conclusions: The findings in this study validates the accurate and reliable use of PI-RADS (v2.1) category at GSH when measured against the expected cancer detection rates reported in large high powered international studies. PI-RADS (v2.1) category 4 revealed a lower-than-expected correlation to international performance standards and is proposed to be the result of multiple interplaying factors such as reporting inexperience, inter-rater variability, suboptimal image acquisition and patient preparation and misrepresentative prostate biopsy. The Urology department, GSH management and patients can draw reserved confidence in the performance of the GSH radiology division reporting mpMRI prostate examinations according to PI-RADS v2.1 lexicon with a higher confidence in PI-RADS (v2.1) category 5 reports than PI-RADS (v2.1) category 4 reports. This study is not extrapolating to PI-RADS 1, 2 and 3 category lesions without further research. |
| format | Thesis |
| id | oai:open.uct.ac.za:11427/41983 |
| institution | University of Cape Town (South Africa) |
| language | English eng |
| last_indexed | 2026-06-10T12:32:18.917Z |
| license_str | Not specified — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from UCTD — University of Cape Town Open Access Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Division of Radiology |
| publisherStr | Division of Radiology |
| record_format | dspace |
| source_str | UCTD — University of Cape Town Open Access Repository |
| spelling | oai:open.uct.ac.za:11427/41983 Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa Wegner, Brett Human, Gercois Moosa, Sulaiman PI-RADS v2.1 PI-RADS 4 PI-RADS 5 Clinically significant prostate cancer mpMRI prostate. Background: Multiparametric MRI (mpMRI) prostate has become a fundamental part of prostate cancer primary diagnosis, active surveillance, and lesion localization. PI-RADS reporting lexicon is used to report mpMRI prostate examinations and is a risk- adjusted reporting structure designed by the European Society of Urogenital Radiology in combination with the American College of Radiology with the main aim to internationally standardized mpMRI prostate reporting into five risk categories, of which a higher PI-RADS category represents an increasing probability that the prostate pathology represents clinically significant cancer. Groote Schuur Hospital (GSH) radiology division has been reporting prostate mpMRI according to the updated PI-RADS v2.1 since its introduction in 2019, however no studies have been performed to assess the reliability of PI-RADS v2.1 at GSH for detecting clinically significant prostate cancer. This study aims to assess and validate the accurate use of PI-RADS v2.1 category 4 and 5 against internationally recognized standards of performance. The PI-RADS lexicon should be a reliable predictor of significant prostate cancer to promote confidence in its continued use and to justify the expenditure on mpMRI prostate examinations in the diagnostic pathway of prostate cancer at GSH. Objectives: The primary objective: To assess the diagnostic performance of PI-RADS v2.1 category 4 and 5 in predicting clinically significant prostate cancer confirmed on biopsy with appropriate histopathology. Secondary objective: - Correlate and validate the diagnostic performance of PI-RADS v2.1 category 4 and 5 lesions at GSH by comparing the outcomes with accepted international performance standards. - Identify problems with the use of PI-RADS v2.1 at GSH to improve prostate reporting with the ultimately goal of improving patient outcomes. Methods: Single institution retrospective study assessing mpMRI prostate PI-RADS (v2.1) category 4 and 5 reports performed at GSH between the dates of 01 June 2021 to 12 January 2024. Prostate reports categorised as PIRADS (v2.1) 4 or 5 are evaluated on PACS and correlated with the histological diagnosis on National Health Laboratory Service (NHLS). Prostate histology will be categorised into clinically significant prostate cancer or insignificant depending on the Gleason score. These outcomes will be measured against high powered international studies to assess the overall performance of the GSH radiology division when reporting PI-RADS 4 and 5 lesions. Results: This study incorporates a retrospective design structure and analysis where a grand total of 215 mpMRI prostate studies were considered for the period dated 01 June 2021 and 12 January 2024. From the 215 mpMRI prostate scans performed 89 studies met the strict inclusion and exclusion criteria: n=89. The sample size of 89 represented 37 PI-RADS category 4 and 52 PI-RADS category 5. The measures of sensitivity and specificity for the PI-RADS 4 and PI-RADS 5 categories in detecting clinically significant prostate cancer proven on biopsy are as follows: For PI-RADS 4: Sensitivity: 19.4, 95% CI: [5.4, 33.3] Specificity: 53.4, 95% CI: [40.6, 66.3] PPV : 16.2 95% CI:4.3-28.1 NPV: 51.9 95% CI:38.3-65.5 For PI-RADS 5: Sensitivity: 80.6, 95% CI: [66.7, 94.6] Specificity: 46.6, 95% CI: [33.7, 59.4] PPV: 48.1 95% CI:34.5- 61.7 NPV: 83.8 95% CI:71.9- 95.7 Conclusions: The findings in this study validates the accurate and reliable use of PI-RADS (v2.1) category at GSH when measured against the expected cancer detection rates reported in large high powered international studies. PI-RADS (v2.1) category 4 revealed a lower-than-expected correlation to international performance standards and is proposed to be the result of multiple interplaying factors such as reporting inexperience, inter-rater variability, suboptimal image acquisition and patient preparation and misrepresentative prostate biopsy. The Urology department, GSH management and patients can draw reserved confidence in the performance of the GSH radiology division reporting mpMRI prostate examinations according to PI-RADS v2.1 lexicon with a higher confidence in PI-RADS (v2.1) category 5 reports than PI-RADS (v2.1) category 4 reports. This study is not extrapolating to PI-RADS 1, 2 and 3 category lesions without further research. 2025-10-02T12:22:12Z 2025-10-02T12:22:12Z 2025 2025-10-02T10:57:58Z Thesis / Dissertation Masters MMed http://hdl.handle.net/11427/41983 en eng application/pdf Division of Radiology Faculty of Health Sciences University of Cape Town |
| spellingShingle | PI-RADS v2.1 PI-RADS 4 PI-RADS 5 Clinically significant prostate cancer mpMRI prostate. Wegner, Brett Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| thesis_degree_str | Master's |
| title | Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| title_full | Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| title_fullStr | Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| title_full_unstemmed | Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| title_short | Retrospective review: diagnostic performance of PI-RADS V2.1 4 and 5 lesions in predicting clinically significant prostate cancer at Groote Schuur Hospital, South Africa |
| title_sort | retrospective review diagnostic performance of pi rads v2 1 4 and 5 lesions in predicting clinically significant prostate cancer at groote schuur hospital south africa |
| topic | PI-RADS v2.1 PI-RADS 4 PI-RADS 5 Clinically significant prostate cancer mpMRI prostate. |
| url | http://hdl.handle.net/11427/41983 |
| work_keys_str_mv | AT wegnerbrett retrospectivereviewdiagnosticperformanceofpiradsv214and5lesionsinpredictingclinicallysignificantprostatecanceratgrooteschuurhospitalsouthafrica |