{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T03:04:07Z","timestamp":1772766247599,"version":"3.50.1"},"posted":{"date-parts":[[2026,3,3]]},"group-title":"Oncology","reference-count":47,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T00:00:00Z","timestamp":1772496000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,3,3]]},"abstract":"<jats:title>Summary<\/jats:title>\n                <jats:sec>\n                  <jats:title>Background<\/jats:title>\n                  <jats:p>\n                    Once the treatment starts, early prediction of treatment benefit and its correlation with overall survival (OS) remains challenging in metastatic castration-resistant prostate cancer (mCRPC). Existing prognostic models require long-term follow-up, limiting their ability to inform timely treatment decisions. To address this gap, we evaluated tumor growth rate (\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate)\u2013based survival models across multiple treatment lines to assess their ability to predict OS and support early clinical decision-making.\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>\n                    We developed GxSurv, a Random Survival Forest (RSF)\u2013based framework that incorporates baseline clinical variables and\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate calculated from serial on-treatment PSA, to construct line-specific prediction models of OS, a direct measure of treatment outcome. Three variants were developed: G3Surv, using the 3-month\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate; G6Surv, using the 6-month\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate; and GfSurv, using the final observed\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate. Model performance was evaluated using Harrell\u2019s C-index, Uno\u2019s C-index, Integrated Brier Score (IBS), time-dependent area under the curve (tAUC). Model interpretability was assessed using permutation importance to quantify predictor contributions within the GxSurv framework.\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Findings<\/jats:title>\n                  <jats:p>\n                    The study included 15912 treatment records from 11014 patients with mCPRC across four lines of therapy. We found that incorporation of\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate consistently improved model performance across all treatment lines, with all GxSurv models outperforming Cox proportional hazards (CoxPH). As the earliest prognostic model, our G3Surv demonstrated strong early predictive performance, with Harrell\u2019s C-index values ranging from 0\u00b7700 to 0\u00b7746 and tAUC values of 0\u00b7766 to 0\u00b7822 across all lines, representing 5-8% and 4-5% improvements over CoxPH, respectively. These results indicate that G3Surv accurately predicts individual treatment outcomes at 3 months after treatment initiation. Feature importance analyses consistently identified\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate as a top predictor, followed by baseline PSA and hemoglobin, with relative variation across treatment lines.\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Interpretation<\/jats:title>\n                  <jats:p>\n                    Integrating\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate calculated from on-treatment PSA values enables accurate, line-specific prediction of treatment outcomes in mCRPC, with the 3-month\n                    <jats:bold>\n                      <jats:italic>g<\/jats:italic>\n                    <\/jats:bold>\n                    -rate providing robust early prognostic information to support timely, personalized clinical decision-making.\n                  <\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Funding<\/jats:title>\n                  <jats:p>U.S. National Science Foundation, National Institutes of Health, American Cancer Society.<\/jats:p>\n                <\/jats:sec>","DOI":"10.64898\/2026.02.26.26346987","type":"posted-content","created":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T00:25:19Z","timestamp":1772583919000},"source":"Crossref","is-referenced-by-count":0,"title":["Early treatment outcome prediction in metastatic castration-resistant prostate cancer utilizing 3-month tumor growth rate (\n                  <i>g<\/i>\n                  -rate) based machine learning model"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-5947-3422","authenticated-orcid":false,"given":"Esther C.","family":"Ugwueke","sequence":"first","affiliation":[{"name":"Department of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mohamed","family":"Azzam","sequence":"additional","affiliation":[{"name":"Department of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mengxi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Memorial Sloan Kettering Cancer Center, New York, NY, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Benjamin A.","family":"Teply","sequence":"additional","affiliation":[{"name":"Department of Internal Medicine, Division of Medical Oncology\/Hematology, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Raymond C.","family":"Bergan","sequence":"additional","affiliation":[{"name":"Fred and Pamela Buffett Cancer Center, University of Nebraska Medical Center, Omaha, NE, USA"},{"name":"Stony Brook Cancer Center, Stony Brook University, Stony Brook, NY, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shibiao","family":"Wan","sequence":"additional","affiliation":[{"name":"Department of Genetics, Cell Biology and Anatomy, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Antonio Tito","family":"Fojo","sequence":"additional","affiliation":[{"name":"Department of Medicine, Division of Hematology\/Oncology, Columbia University Medical Center, New York, NY, USA"},{"name":"Hematology\/Oncology, James J. Peters VA Medical Center, Bronx, NY, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Harshraj","family":"Leuva","sequence":"additional","affiliation":[{"name":"Department of Internal Medicine, Division of Medical Oncology\/Hematology, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jieqiong","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2026030518051616000_2026.02.26.26346987v1.1","unstructured":"Cancer of the Prostate - Cancer Stat Facts. SEER. SEER, National Cancer Institute, Bethesda, MD. Accessed 2025."},{"key":"2026030518051616000_2026.02.26.26346987v1.2","doi-asserted-by":"crossref","first-page":"11156","DOI":"10.3390\/ijerph182111156","article-title":"Prostate Cancer Survival by Risk and Other Prognostic Factors in Mallorca, Spain","volume":"18","year":"2021","journal-title":"Int J Environ Res Public Health"},{"key":"2026030518051616000_2026.02.26.26346987v1.3","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2017.7248"},{"key":"2026030518051616000_2026.02.26.26346987v1.4","doi-asserted-by":"publisher","DOI":"10.1038\/onc.2013.206"},{"key":"2026030518051616000_2026.02.26.26346987v1.5","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2025.0228"},{"key":"2026030518051616000_2026.02.26.26346987v1.6","doi-asserted-by":"publisher","DOI":"10.1002\/cncr.28890"},{"key":"2026030518051616000_2026.02.26.26346987v1.7","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1186\/s13014-017-0786-z","article-title":"An appraisal of analytical tools used in predicting clinical outcomes following radiation therapy treatment of men with prostate cancer: a systematic review","volume":"12","year":"2017","journal-title":"Radiat Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.8","doi-asserted-by":"crossref","first-page":"e158","DOI":"10.1016\/S2589-7500(20)30314-9","article-title":"Application of a novel machine learning framework for predicting non-metastatic prostate cancer-specific mortality in men using the Surveillance, Epidemiology, and End Results (SEER) database","volume":"3","year":"2021","journal-title":"Lancet Digit Health"},{"key":"2026030518051616000_2026.02.26.26346987v1.9","doi-asserted-by":"crossref","first-page":"155687","DOI":"10.1016\/j.prp.2024.155687","article-title":"Prostate cancer prognosis using machine learning: A critical review of survival analysis methods","volume":"264","year":"2024","journal-title":"Pathol Res Pract"},{"key":"2026030518051616000_2026.02.26.26346987v1.10","doi-asserted-by":"crossref","first-page":"1232","DOI":"10.1200\/JCO.2003.06.100","article-title":"Prognostic model for predicting survival in men with hormone-refractory metastatic prostate cancer","volume":"21","year":"2003","journal-title":"J Clin Oncol Off J Am Soc Clin Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.11","doi-asserted-by":"crossref","first-page":"1502629","DOI":"10.3389\/fonc.2024.1502629","article-title":"Harnessing machine learning to predict prostate cancer survival: a review","volume":"14","year":"2025","journal-title":"Front Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.12","doi-asserted-by":"publisher","DOI":"10.1002\/1097-0142(19921215)70:12<2870::AID-CNCR2820701225>3.0.CO;2-F"},{"key":"2026030518051616000_2026.02.26.26346987v1.13","doi-asserted-by":"crossref","first-page":"6396","DOI":"10.1158\/1078-0432.CCR-07-1036","article-title":"A contemporary prognostic nomogram for men with hormone-refractory metastatic prostate cancer: a TAX327 study analysis","volume":"13","year":"2007","journal-title":"Clin Cancer Res Off J Am Assoc Cancer Res"},{"key":"2026030518051616000_2026.02.26.26346987v1.14","doi-asserted-by":"crossref","first-page":"3972","DOI":"10.1200\/JCO.2002.11.021","article-title":"Nomogram for overall survival of patients with progressive metastatic prostate cancer after castration","volume":"20","year":"2002","journal-title":"J Clin Oncol Off J Am Soc Clin Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.15","doi-asserted-by":"crossref","first-page":"3876","DOI":"10.1158\/1078-0432.CCR-10-2815","article-title":"Targeting continued androgen receptor signaling in prostate cancer","volume":"17","year":"2011","journal-title":"Clin Cancer Res Off J Am Assoc Cancer Res"},{"key":"2026030518051616000_2026.02.26.26346987v1.16","doi-asserted-by":"crossref","first-page":"2494","DOI":"10.3390\/cells9112494","article-title":"Clinical Actionability of the Genomic Landscape of Metastatic Castration Resistant Prostate Cancer","volume":"9","year":"2020","journal-title":"Cells"},{"key":"2026030518051616000_2026.02.26.26346987v1.17","doi-asserted-by":"publisher","DOI":"10.1016\/j.eururo.2013.05.029"},{"key":"2026030518051616000_2026.02.26.26346987v1.18","doi-asserted-by":"crossref","first-page":"386","DOI":"10.1002\/psp4.49","article-title":"Disease Progression\/Clinical Outcome Model for Castration-Resistant Prostate Cancer in Patients Treated With Eribulin","volume":"4","year":"2015","journal-title":"CPT Pharmacomet Syst Pharmacol"},{"key":"2026030518051616000_2026.02.26.26346987v1.19","doi-asserted-by":"publisher","DOI":"10.1158\/1078-0432.CCR-16-1224"},{"key":"2026030518051616000_2026.02.26.26346987v1.20","doi-asserted-by":"crossref","first-page":"454","DOI":"10.1093\/annonc\/mdv594","article-title":"A prognostic index model for predicting overall survival in patients with metastatic castration-resistant prostate cancer treated with abiraterone acetate after docetaxel","volume":"27","year":"2016","journal-title":"Ann Oncol Off J Eur Soc Med Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.21","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1200\/JCO.2013.52.3696","article-title":"Updated prognostic model for predicting overall survival in first-line chemotherapy for patients with metastatic castration-resistant prostate cancer","volume":"32","year":"2014","journal-title":"J Clin Oncol Off J Am Soc Clin Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.22","doi-asserted-by":"publisher","DOI":"10.1093\/jnci\/djt280"},{"key":"2026030518051616000_2026.02.26.26346987v1.23","doi-asserted-by":"publisher","DOI":"10.1158\/1078-0432.CCR-09-2514"},{"key":"2026030518051616000_2026.02.26.26346987v1.24","doi-asserted-by":"crossref","first-page":"2672","DOI":"10.12688\/f1000research.8628.1","article-title":"Predicting survival time for metastatic castration resistant prostate cancer: An iterative imputation approach","volume":"5","year":"2016","journal-title":"F1000Research"},{"key":"2026030518051616000_2026.02.26.26346987v1.25","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejca.2009.11.007"},{"key":"2026030518051616000_2026.02.26.26346987v1.26","first-page":"613","article-title":"Prognostic models for predicting overall survival in metastatic castration-resistant prostate cancer: a systematic review","volume":"38","year":"2020","journal-title":"World Journal of Urlology"},{"issue":"63","key":"2026030518051616000_2026.02.26.26346987v1.27","doi-asserted-by":"crossref","first-page":"106468","DOI":"10.18632\/oncotarget.22474","article-title":"Prognostic factors in men with metastatic castration-resistant prostate cancer treated with cabazitaxel","volume":"8","year":"2017","journal-title":"Oncotarget"},{"key":"2026030518051616000_2026.02.26.26346987v1.28","doi-asserted-by":"crossref","first-page":"403","DOI":"10.1200\/JCO.18.01279","article-title":"Overall Survival of Black and White Men With Metastatic Castration-Resistant Prostate Cancer Treated With Docetaxel","volume":"37","year":"2019","journal-title":"J Clin Oncol Off J Am Soc Clin Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.29","first-page":"351","article-title":"A novel approach to assess real-world efficacy of cancer therapy in metastatic prostate cancer","volume":"46","year":"2019","journal-title":"Analysis of national data on Veterans treated with abiraterone and enzalutamide. Semin Oncol"},{"key":"2026030518051616000_2026.02.26.26346987v1.30","doi-asserted-by":"publisher","DOI":"10.1634\/theoncologist.2008-0075"},{"key":"2026030518051616000_2026.02.26.26346987v1.31","doi-asserted-by":"publisher","DOI":"10.1158\/1078-0432.CCR-10-1762"},{"key":"2026030518051616000_2026.02.26.26346987v1.32","doi-asserted-by":"publisher","DOI":"10.1016\/S1470-2045(17)30074-8"},{"key":"2026030518051616000_2026.02.26.26346987v1.33","doi-asserted-by":"crossref","first-page":"105288","DOI":"10.1016\/j.ebiom.2024.105288","article-title":"Assessing olaparib efficacy in U.S. Veterans with metastatic prostate cancer utilizing a time-indifferent g-rate method ideal for real-world analyses","volume":"107","year":"2024","journal-title":"eBioMedicine"},{"key":"2026030518051616000_2026.02.26.26346987v1.34","doi-asserted-by":"crossref","unstructured":"Alba PR , Gao A , Lee KM , Anglin-Foote T , Robison B , Katsoulakis E , et al. Ascertainment of Veterans With Metastatic Prostate Cancer in Electronic Health Records: Demonstrating the Case for Natural Language Processing. JCO Clin Cancer Inform. 2021;xs1005\u201314.","DOI":"10.1200\/CCI.21.00030"},{"key":"2026030518051616000_2026.02.26.26346987v1.35","unstructured":"Wilkerson: Tumor growth rate analysis - Google Scholar. cited 2025."},{"key":"2026030518051616000_2026.02.26.26346987v1.36","doi-asserted-by":"publisher","DOI":"10.1214\/08-AOAS169"},{"key":"2026030518051616000_2026.02.26.26346987v1.37","doi-asserted-by":"crossref","first-page":"1697","DOI":"10.3348\/kjr.2021.0223","article-title":"Review of Statistical Methods for Evaluating the Performance of Survival or Other Time-to-Event Prediction Models (from Conventional to Deep Learning Approaches)","volume":"22","year":"2021","journal-title":"Korean J Radiol"},{"key":"2026030518051616000_2026.02.26.26346987v1.38","doi-asserted-by":"publisher","DOI":"10.1177\/0962280216680245"},{"key":"2026030518051616000_2026.02.26.26346987v1.39","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2010.03.014"},{"key":"2026030518051616000_2026.02.26.26346987v1.40","doi-asserted-by":"publisher","DOI":"10.1016\/j.spl.2010.02.020"},{"key":"2026030518051616000_2026.02.26.26346987v1.41","doi-asserted-by":"publisher","DOI":"10.1186\/s12885-019-5729-7"},{"key":"2026030518051616000_2026.02.26.26346987v1.42","doi-asserted-by":"crossref","first-page":"1476","DOI":"10.2967\/jnumed.119.240242","article-title":"Early Prostate-Specific Antigen Changes and Clinical Outcome After 177Lu-PSMA Radionuclide Treatment in Patients with Metastatic Castration-Resistant Prostate Cancer","volume":"61","year":"2020","journal-title":"J Nucl Med"},{"key":"2026030518051616000_2026.02.26.26346987v1.43","doi-asserted-by":"crossref","first-page":"100804","DOI":"10.1016\/j.isci.2019.100804","article-title":"Treatment Stratification of Patients with Metastatic Castration-Resistant Prostate Cancer by Machine Learning","volume":"23","year":"2020","journal-title":"iScience"},{"issue":"1","key":"2026030518051616000_2026.02.26.26346987v1.44","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1111\/bju.14477","article-title":"Use of machine learning to predict early biochemical recurrence after robot-assisted prostatectomy","volume":"123","year":"2019","journal-title":"BJU Int"},{"key":"2026030518051616000_2026.02.26.26346987v1.45","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1016\/j.clgc.2016.08.018","article-title":"Predicting Time From Metastasis to Overall Survival in Castration-Resistant Prostate Cancer: Results From SEARCH","volume":"15","year":"2017","journal-title":"Clin Genitourin Cancer"},{"key":"2026030518051616000_2026.02.26.26346987v1.46","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-32987-6"},{"key":"2026030518051616000_2026.02.26.26346987v1.47","doi-asserted-by":"crossref","first-page":"103550","DOI":"10.1016\/j.jbi.2020.103550","article-title":"Time-range based sequential mining for survival prediction in prostate cancer","volume":"110","year":"2020","journal-title":"J Biomed Inform"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.02.26.26346987","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,6]],"date-time":"2026-03-06T02:06:18Z","timestamp":1772762778000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.02.26.26346987"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,3]]},"references-count":47,"URL":"https:\/\/doi.org\/10.64898\/2026.02.26.26346987","relation":{},"subject":[],"published":{"date-parts":[[2026,3,3]]},"subtype":"preprint"}}