{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T23:28:17Z","timestamp":1792193297152,"version":"4.4.0"},"posted":{"date-parts":[[2026,4,24]]},"group-title":"Oncology","reference-count":16,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T00:00:00Z","timestamp":1776988800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/publicdomain\/zero\/1.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,4,24]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Objective<\/jats:title>\n                  <jats:p>To demonstrate the proof of principle that machine learning (ML) can be used to quantify Gleason Pattern (GP) 4 on digitized biopsy slides using multiple measurement approaches, allowing direct comparison of their prognostic performance.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>We assembled a convenience sample of 726 patients with grade group 2-4 prostate cancer on systematic biopsy who underwent radical prostatectomy between 2014 and 2023. Digitized biopsy slides were analyzed using a machine-learning algorithm (PAIGE-AI) to quantify GP4 using multiple measurement approaches, particularly with respect to how gaps between cancer foci (\u201cinterfocal stroma\u201d) were handled. GP4 extent was quantified using linear measurements or a pixel-based area metric. Discrimination of each GP4 quantification approach, along with Grade Group (GG), was assessed for adverse radical prostatectomy pathology and biochemical recurrence.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>We identified 15 different quantification approaches and observed differences between their discrimination. The highest discrimination was in the pixel-countingmethod (AUC 0.648). GP4 quantification outperformed GG for predicting adverse pathology (AUC 0.627 vs 0.608). Amount of GP3 was non-predictive once GP4 was known. These findings were consistent for BCR.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>We were able to measure slides using 15 distinct measurement approaches and replicated prior findings using ML to quantify GP4. Our findings support the use of ML as a research tool to compare different GP4 quantification approaches. We intend to use our method on larger cohorts to determine with which measurement approach best predicts oncologic outcome.<\/jats:p>\n                <\/jats:sec>","DOI":"10.64898\/2026.04.23.26351615","type":"posted-content","created":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T21:55:16Z","timestamp":1777067716000},"source":"Crossref","is-referenced-by-count":1,"title":["Comparing Gleason Pattern 4 Measurement Approaches on Prostate Biopsy Using Machine Learning: A Proof-of-Principle Study"],"prefix":"10.64898","author":[{"given":"Matei M.","family":"Buzoianu","sequence":"first","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rebecca","family":"Yu","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melissa","family":"Assel","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alican","family":"Bozkurt","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hamed","family":"Aghdam","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samson W.","family":"Fine","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew J.","family":"Vickers","sequence":"additional","affiliation":[{"name":"Departments of Epidemiology and Biostatistics (MMB, RY, MA, AJV) and Pathology & Laboratory Medicine (SWF), Memorial Sloan Kettering Cancer Center PAIGE-AI (AB, HA)"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026042709301011000_2026.04.23.26351615v1.1","doi-asserted-by":"publisher","DOI":"10.1097\/PAS.0b013e3182556dcd"},{"key":"2026042709301011000_2026.04.23.26351615v1.2","unstructured":"Vickers, A.J. , S. Fine , and G. ProQuant , Evaluating Tumor Quantification in Place of Proportions in Prostate Cancer: Principles of the ProQuant Group. Eur Urol, 2025."},{"key":"2026042709301011000_2026.04.23.26351615v1.3","doi-asserted-by":"publisher","DOI":"10.1097\/JU.0000000000002685"},{"issue":"1","key":"2026042709301011000_2026.04.23.26351615v1.4","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.eururo.2024.01.005","article-title":"Amount of Gleason Pattern 3 Is Not Predictive of Risk in Grade Group 2-4 Prostate Cancer","volume":"86","year":"2024","journal-title":"European Urology"},{"key":"2026042709301011000_2026.04.23.26351615v1.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.urology.2022.08.014"},{"key":"2026042709301011000_2026.04.23.26351615v1.6","doi-asserted-by":"publisher","DOI":"10.1111\/his.15372"},{"key":"2026042709301011000_2026.04.23.26351615v1.7","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-019-0508-1"},{"key":"2026042709301011000_2026.04.23.26351615v1.8","doi-asserted-by":"publisher","DOI":"10.2307\/2531595"},{"issue":"1","key":"2026042709301011000_2026.04.23.26351615v1.9","first-page":"570","article-title":"Reproducible Summary Tables with the gtsummary Package","volume":"13","year":"2021","journal-title":"R Journal"},{"key":"2026042709301011000_2026.04.23.26351615v1.10","unstructured":"Paner, G.P. , Srigley, J. R. , Giannico, G. A. , & Sirintrapun, S. J. , Protocol for the Examination of Prostate Needle Biopsies From Patients With Carcinoma of the Prostate Gland: Specimen Level Reporting. 2021, College of American Pathologists: Northfield, IL."},{"key":"2026042709301011000_2026.04.23.26351615v1.11","doi-asserted-by":"publisher","DOI":"10.1097\/JU.0000000000004484"},{"key":"2026042709301011000_2026.04.23.26351615v1.12","doi-asserted-by":"crossref","unstructured":"Pickersgill, N.A. , et al., Total Gleason Pattern 4 Length Outperforms Grade Group and Clinical Models in Predicting Oncologic Outcomes in Grade Group 2-4 Prostate Cancer. medRxiv, 2026: p. 2026.02.27.26347289.","DOI":"10.64898\/2026.02.27.26347289"},{"key":"2026042709301011000_2026.04.23.26351615v1.13","doi-asserted-by":"publisher","DOI":"10.1111\/bju.16884"},{"key":"2026042709301011000_2026.04.23.26351615v1.14","doi-asserted-by":"publisher","DOI":"10.1097\/PAS.0b013e3182217b79"},{"issue":"2","key":"2026042709301011000_2026.04.23.26351615v1.15","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1097\/PAS.0000000000000344","article-title":"Does Discontinuous Involvement of a Prostatic Needle Biopsy Core by Adenocarcinoma Correlate With a Large Tumor Focus at Radical Prostatectomy?","volume":"39","year":"2015","journal-title":"American Journal of Surgical Pathology"},{"issue":"4","key":"2026042709301011000_2026.04.23.26351615v1.16","doi-asserted-by":"crossref","first-page":"646","DOI":"10.1038\/s41391-022-00600-y","article-title":"Oncological outcomes of cribriform histology pattern in prostate cancer patients: a systematic review and meta-analysis","volume":"26","year":"2023","journal-title":"Prostate Cancer and Prostatic Diseases"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.04.23.26351615","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,27]],"date-time":"2026-04-27T16:31:26Z","timestamp":1777307486000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.04.23.26351615"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,24]]},"references-count":16,"URL":"https:\/\/doi.org\/10.64898\/2026.04.23.26351615","relation":{},"subject":[],"published":{"date-parts":[[2026,4,24]]},"subtype":"preprint"}}